Document title - cordis.europa.eu€¦ · Web viewGreen driving advisory systems were found to...

Final Report Public Copyright TeleFOT

Contract N. 224067

Large Scale Collaborative Project

7th Framework Programme

INFSO-ICT 224067

Final ReportDeliverable n. D1.15 Final Report

Subproject SP 1 Coordination

Work package WP1.1 Management

Task n. T1.1.1; T1.1.2 Coordination, Management

Authors Petri Mononen, Stig Franzen, Katia Pagle, Andrew Morris, Satu Innamaa, MariAnne Karlsson, Katerina Touliou, Roberto Montanari , Serena Fruttaldo

Status Final

Distribution Public (PU)

Issue date M54 Creation date 2012/11/11

Project start and duration

1st of June, 2008 – 54 months

TABLE OF CONTENTSProject co-funded by the European Commission

DG-Information Society and Media

in the 7th Framework Programme


Contract N. 224067

TABLE OF CONTENTS..........................................................................................................2

LIST OF FIGURES.................................................................................................................5

LIST OF TABLES...................................................................................................................5

LIST OF ABBREVIATIONS.....................................................................................................6

REVISION CHART AND HISTORY LOG..................................................................................8

EXECUTIVE SUMMARY.........................................................................................................9

1. INTRODUCTION AND BACKGROUND...........................................................................13

1.1. Objectives..........................................................................................................................13

1.2. The main focus...................................................................................................................14

1.3. The consortium and the work plan.....................................................................................15

1.4. Impacts..............................................................................................................................16

1.5. www.telefot.eu...................................................................................................................17

1.6. Project structure................................................................................................................17

2. FRAMEWORK FOR A FOT WITH NOMADIC DEVICES.....................................................19

2.1. The TeleFOT project...........................................................................................................19

2.2. TeleFOT Characteristics.....................................................................................................21

2.3. Integrated top-down and bottom-up approach..................................................................22

2.4. Data management.............................................................................................................24

2.5. Collaboration......................................................................................................................25

3. TEST SITES..................................................................................................................26

3.1. Test Communities Concept................................................................................................26

3.2. The key characteristics of TeleFOT FOTs...........................................................................28

3.3. TeleFOT Test Sites and FOTs..............................................................................................29

4. ANALYSIS AND EVALUATION.......................................................................................33

4.1. User uptake........................................................................................................................33

Project co-funded by the European Commission




Contract N. 224067

4.2. Impacts in travel and driver behaviour..............................................................................38

5. IMPLICATIONS OF RESULTS.........................................................................................48

5.1. Navigation..........................................................................................................................48

5.2. Traffic information..............................................................................................................49

5.3. Speed information and alert..............................................................................................51

5.4. Green driving.....................................................................................................................52

6. DISCUSSION ON IMPACT ASSESSMENT RESULTS........................................................54

6.1. User uptake........................................................................................................................54

6.2. Mobility..............................................................................................................................57

6.3. Efficiency............................................................................................................................59

6.4. Environment.......................................................................................................................60

6.5. Safety.................................................................................................................................61

6.6. Safety Comparability of Results across the Impact Assessments......................................64

6.7. General Limitations of the Results.....................................................................................66

7. REACH-OUT................................................................................................................71

7.1. Dissemination....................................................................................................................71

7.2. Exploitation........................................................................................................................79

7.3. Stakeholder Fora................................................................................................................81

7.4. Other Stakeholder-related activities..................................................................................84

8. LESSONS LEARNT.......................................................................................................86

8.1. SP2 Framework..................................................................................................................86

8.2. SP3 FOTs............................................................................................................................86

8.3. SP4 Analysis.......................................................................................................................88

8.4. SP5 Dissemination.............................................................................................................89

9. CONCLUSIONS............................................................................................................91

9.1. General conclusions...........................................................................................................91





Contract N. 224067

9.2. Summary and conclusions on key findings........................................................................92

9.3. Data access for third parties..............................................................................................93

REFERENCES.....................................................................................................................96

ANNEXES..........................................................................................................................97

ANNEX 1: List of TeleFOT deliverables........................................................................................98ANNEX 2: Results by Research Question and Hypothesis.........................................................101





Contract N. 224067

LIST OF FIGURES

Figure 1: TeleFOT test sites.......................................................................................................................... 15

Figure 2: Workplan general structure with Sub Projects, Work Packages and coordinating organisations......18

Figure 3: The FESTA V................................................................................................................................... 21

Figure 4: The combined top-down and bottom-up approach........................................................................23

Figure 5: The total process from impacts over hypotheses to sensors...........................................................24

Figure 6: Illustration of the TeleFOT Tests Communities and Test Sites.........................................................27

Figure 7: Messages and tools....................................................................................................................... 73

Figure 8 – TeleFOT stakeholders................................................................................................................... 74

LIST OF TABLES

Table 1: Summary of TeleFOT FOTs.............................................................................................................. 30

Table 2: Comparability of results across the impact assessments..................................................................65

Table 3: The test-site FOT data that were used to address each Research Question......................................69





Contract N. 224067

LIST OF ABBREVIATIONS

ABBREVIATION

DESCRIPTION

ADAS Advanced Driver Assistance System

CAN Controller Area Network

DFOT Detailed Scale FOT

EC European Commission

FOT Field Operational Test

GPRS General packet radio service

GPS Global Positioning System

GSM Global System for Mobile Communications

ICT Information and Communication Technology

LFOT Large Scale FOT

NDA Non-Disclosure Agreement

OBD-II On-Board Diagnostics (interface)

PND Personal Navigation Device

PSAP Public Safety Answering Point





Contract N. 224067

ABBREVIATION

DESCRIPTION

RAM Random Access Memory

RDS Radio Data System

SD Secure Digital

SIM Subscriber Identity Module

SP Sub Project

TMC Traffic Message Channel





Contract N. 224067

REVISION CHART AND HISTORY LOG

REV DATE AUTHOR REASON

0.1 11th Nov 2012

Petri Mononen Template, chapter 1

0.4 11th Jan 2013

Petri Mononen First draft version based on SP input

0.5 16th Jan 2013

Petri Mononen Draft version to the EC

0.614th February 2013

Andrew Morris and Satu Innamaa

Modification of Chapters 4-7, 9 and 10

0.715th February 2013

Katia Pagle Quality review & format modifications

0.8 22nd March 2013

Stig Franzén Up-date of section 8.1

1.0 25th March 2013

Petri Mononen Final edits (Executive summary, Conclusions, Annex 1, formatting.)





Contract N. 224067

EXECUTIVE SUMMARY

Background and general information

This deliverable is the final report of the TeleFOT project. TeleFOT Consortium presents the outcome of a four-year study which focused on the assessment of the impacts of driver support functions provided by smartphones, navigators and other in-vehicle aftermarket and nomadic devices on the driving task and driver behaviour.

With a budget of 15 million euro, the four-year TeleFOT project, coordinated by VTT Technical Research Centre of Finland, is one of the biggest ICT projects in Europe addressing traffic (as well as transport and travel). The completed field operational tests produced a unique set of data, based on a comprehensive assessment of driver behaviour and the efficiency, quality, robustness and user-friendliness of interactive in-vehicle traffic systems and services.

Many intelligent transport services provided by nomadic devices are already part of the daily lives of road users, but information about their actual impacts on road safety, for example, has not previously been available. It is generally acknowledged that nomadic devices such as portable navigators and smart phones have thoroughly penetrated the market during the last decade. For example, hundreds of millions of smart phones that are capable of GPS navigation are sold worldwide on a yearly basis. Simultaneously the same has also happened to thousands of services that are provided through those nomadic devices - including services directed to drivers and travellers. Even though both the devices and services usually go through a thorough testing cycle during their design and production phases, up to now there still has been a very small amount of transparent unbiased scientific information available on the positive or negative effects of those services to the driving task, e.g. in terms of safety, efficiency and mobility. TeleFOT is now able to shed light on these issues based on analysed real-life subjective and objective data – data that has been collected from millions of kilometres driven by thousands of normal everyday users while interacting with services such as navigation,





Contract N. 224067

green driving support, real-time traffic information and speed limit information. Furthermore, TeleFOT has during its lifetime in many ways enhanced the FOT methodology itself.

Key findings

The extensive research material reveals that intelligent transport systems allowed drivers to find quicker and less congested routes, and prevented them from speeding accidentally. Fuel costs also dropped, as did driving-related stress and anxiety. The drivers’ sense of safety and driving comfort increased. The study was based on extensive field trials, with almost 3,000 drivers covering a combined distance of more than 10 million kilometres in eight European countries. Test drivers were recruited for the project from Finland, Sweden, Germany, the UK, France, Greece, Italy and Spain.

The project studied the impacts of driver support functions provided by in-vehicle aftermarket devices on safety, efficiency, mobility, the environment and driver behaviour in road traffic. The user acceptance and uptake of the services were also studied. The services tested included Static and Dynamic Navigation Support, Green Driving Support, Speed Limit Information, Traffic Information and eCall.

The main benefits of the functions were perceived by the participants to be Convenience (easy access to information), Comfort (less uncertainty, fewer driving errors), Economic (less cost) and Environment (fewer emissions).

Of the tested devices, navigators and traffic information systems, in particular, increased efficiency by allowing drivers to find quicker and less congested routes. Up to 45% of participants, particularly those in large cities, reported that the Traffic Information function helped them to avoid travel delays and traffic jams. Green driving systems guided drivers to routes that lowered their emissions, and towards driving more economically. Green driving advisory systems were found to reduce fuel consumption by up to 6%.





Contract N. 224067

The use of a green driving system in a bus fleet helped to lower fuel consumption and to reduce speeding, which also improves road safety. Another significant finding is that the systems reduced driving-related stress and anxiety across the board and, in all the participating countries, increased the drivers’ sense of safety and driving comfort. From the perspective of mobility, the results were positive for all systems.

The users’ expectations for the services were high at first. After using the services for some time, they were slightly disappointed not to have seen a direct benefit. The longer they used the services, the more clearly they could see the benefits and advantages, and the more satisfied they were. Participants assessments of the designs of the devices were positive but there were some negative views. Acceptance of the devices changed over time – acceptance results in usage rather than vice-versa.

There is no evidence to suggest that the TeleFOT functions affected Mode of Transport and Timing of Commuting journeys.

Eyes off road time was found to increase when the navigation function was introduced although the Green-driving function did not change visual behaviour.

Navigation support has positive implications in all areas of Mobility and many aspects of Efficiency but the function effects are small for Environment. Some effects for Safety are evident in terms of distance travelled (reduces) and distraction (increases).

Traffic Information is positive for Mobility in terms of reduced journey duration and reductions in stress and uncertainty. It was also positive for Efficiency in terms of reduced travel durations, reduced headway variations and perception of avoidance of congestion. The impact of this function on Environment was inconclusive.

Speed Information/Alert ha a small but mainly positive effect for Mobility, Efficiency and Safety.

The Green-driving function was found to decrease fuel consumption but increase journey duration. Average speed was found to decrease with use of the Green-driving function. Average speed and speed variance were both smaller with this function. The Green-





Contract N. 224067

driving function had a positive effect on Efficiency and Environment. This function had a mixed effect on Mobility and a negative effect on Safety (by changing exposure).

Limitations

Overall the TeleFOT project was not without limitations but most of these were recognised at the beginning of the project and specific measures were introduced to overcome these. However, some of the limitations are still important to highlight as follows.

TeleFOT was reliant on third-parties for execution of the FOTs in some cases. This was highly beneficial on the one-hand as it provided extra input to the project at minimal cost. However, one implication was that the project had little control over this input.

Smaller sample sizes were necessary in the Detailed FOTs compared to the Large-scale FOTs although this was largely offset by enhanced data acquisition in the detailed FOTs.

The majority of the FOTs were performed using a within-subject (or before-and-after) design. In the some cases, a matched control-group throughout the whole test-period might have been more ideal but practical reasons (e.g. resources in time and money) made did not allow this throughout the project.

The fast-moving pace of technology could limit the results to some degree - the TeleFOT project began in 2008 just as technology proliferated. By the end of TeleFOT, technology had evolved substantially so that the results, whilst largely generalizable to the first or second generation of after-market devices could not easily be applied to the state-of-the-art. Many of the functions tested have already migrated from smart-phones to apps built on application programming interfaces and a new world that is dominated by platforms and technology-enabled services.

Use of the project results





Contract N. 224067

TeleFOT findings and recommendations help steering policy making and investment to a more cost-effective direction and help the industry design better and more attractive products. Thus the project outcomes help enhancing the well-being of Europe in general by contributing towards a smarter and more cooperative transport system. This was the main TeleFOT objective from the outset and upon closure TeleFOT as a project was able to meet its set objectives.

The results reported in this document should be taken in the context of the particular systems tested, the participants involved and the road environments/countries covered in the large scale and detailed trials. Please refer to the full set of TeleFOT deliverables for the detailed information. Further information can be found on the website www.telefot.eu.

1. INTRODUCTION AND BACKGROUNDTeleFOT is a large-scale collaborative project under the seventh Framework Programme, co-funded by the European Commission DG Information Society and Media within the strategic objective "ICT for Cooperative Systems".

Started on 1 June 2008, TeleFOT aimed to test the impacts of driver support functions on the driving task with large fleets of test drivers in real-life driving conditions. In particular, TeleFOT assessed via Field Operational Tests (FOTs) the impacts of functions provided by aftermarket and nomadic devices, including present mature services and future interactive traffic services that will become part of driving environment systems within the next five years.

Field Operational Tests developed in TeleFOT led to a comprehensive assessment of the efficiency, quality, robustness and user friendliness of in-vehicle systems, such as ICT, for smarter, safer and cleaner driving.

TeleFOT concluded on 30th November 2012 and this is the Final Report of the project.





Contract N. 224067

1.1. ObjectivesThe objectives of the TeleFOT project were to assess the impacts of functions provided by aftermarket and nomadic devices in vehicles and raise wide awareness of their traffic safety potential. In particular, TeleFOT investigated how retrofitted equipment, such as navigators and smart phones that are not in all cases originally designed for in-vehicle use, can support the driver and the detailed effects of the information produced for this purpose on the driving task. These devices can provide different types of driver support functions and almost nothing is known about their safety and other impacts as yet.

The project also aimed at speeding up the penetration of systems able to "see" beyond drivers’ field of vision in conditions where good situation awareness is needed.

TeleFOT also provided opportunities to test the impacts of similar functions that future cooperative systems will provide after their development challenges have been solved in the coming years. In fact, aftermarket and nomadic devices provide an alternative to some important cooperative driving and ADAS functions for many years ahead. The concept comprised of creating three European test communities: Northern, Central and Southern. Almost 3 000 drivers participated in the tests.

1.2. The main focusThe market penetration of portable navigators and smart phones has been and still is exploding, therefore the timing for the project was ideal. The tested functions cover two broad areas: functions aiming at promoting (i) safe driving and (ii) economic and fuel efficient driving. These are Speed information, Traffic information, Road weather information and “Green driving” support. The impacts were assessed on levels ranging from usability; behaviour & incidents; safety; Green Driving and efficiency; to the impacts on the transport system. Attention was also paid to possible negative impacts, since especially smart phones are not originally designed for vehicle use and navigators may have problems in fixing and positioning in the cockpit. Business models were also studied.





Contract N. 224067

Field Operational Tests

Field Operational Tests developed in TeleFOT were aimed at a comprehensive assessment of the efficiency, quality, robustness and user friendliness of in-vehicle systems, such as ICT, for smarter, safer and cleaner driving. FOTs were organized in three test communities in Northern (Finland, Sweden), Central (Germany, UK, France) and Southern (Greece, Italy, Spain) Europe.

Figure 1: TeleFOT test sites

The tests were planned in two phases: first, short and long term testing were performed with a large number of vehicles. In the second phase, detailed testing with a limited





Contract N. 224067

number of subjects with instrumented cars was carried out. In the tests, drivers had access to smart phones and navigators and the effects of the services they provide to support driving were tested. Prior to any field operational tests, the usability and safety of the devices and services was studied carefully in laboratory conditions.

1.3. The consortium and the work plan

1.3.1. The ConsortiumThe project Consortium was composed of Research Institutes, Universities as well as representatives from the European automotive industry, equipment manufacturers, road maintenance, and service operators.

TeleFOT was led by VTT Technical Research Centre of Finland and Mr Petri Mononen was the project Coordinator. In addition to VTT, the project partners were: Adac, Blom, BroadBit, Centro Ricerche FIAT, CERTH/HIT, Chalmers, CIDAUT, MediaMobile Nordic, Emtele, Electronic Trafic S.A., Institute for Communication and Computer Systems, RWTH Aachen - Institut für Kraftwahrwesen, Logica Suomi Oy, Loughborough University, Magneti Marelli, Metasystem, MIRA Ltd, Navteq, Universite de Technologie de Belfort-Montbeliard (joined during 2nd project year), Rücker Lypsa, Swedish Road Administration, and University of Modena and Reggio Emilia.

1.3.2. The WorkplanAs a Large Scale Project, TeleFOT was divided into five Sub-Projects:

In SP1 (Coordination) management procedures were used to ensure shared and clear decisions among the partners, receiving support from the Core Group and the EU team at VTT.

In SP2 (FOT Framework) a detailed TeleFOT operational framework was created to guide the activities, firstly based on the FESTA general handbook, then, constantly updated when new knowledge was accumulated and data has been collected.





Contract N. 224067

SP3 (Field Tests) coordinated three Test Communities (in Northern, Central and Southern Europe) to collect the data of the introduction of nomadic devices to the vehicle environment. Data collected were analysed in SP4.

SP4 (Evaluation and Assessment) ensured an appropriate analysis of data collected, in order to identify the impacts of aftermarket and nomadic devices/functions.

In SP5 (Dissemination), user awareness and exploitation activities were coordinated, also supporting liaison among all subprojects and raising awareness of its goals and results since the earliest phases of its activities.

1.4. ImpactsThe project will speed up the penetration of systems able to support drivers’ field of vision in conditions where good situation awareness is needed. The collected sizable database will be useful and valuable also after the project completion.

1.5. www.telefot.euDetailed information on the project results and the consortium are available on its website http://www.telefot.eu .

1.6. Project structureError: Reference source not found provides the main framework for the work plan, which is constructed in five Sub Projects (SP) and several technical Work Packages (WP). All of the WPs were divided into several Tasks (T). Two of the SPs (SP1 Coordination and SP5 Dissemination) were horizontal in nature, i.e. they ran on more or less the same level of activity for the whole duration of the project, from start to finish. The remaining three SPs progressed in a logical manner, starting from the FOT framework build-up (SP2), continuing to the setup, implementation and operation of the Field Operational Tests (SP3) and completing with the Evaluation and assessment of the data (SP4).




http://www.telefot.eu/


Contract N. 224067

WP1.1 Management (VTT)WP1.2 Project administration (VTT)WP1.3 Quality, Risk and Knowledge Management (VTT)

SP1 Coordination (VTT)

WP 5.1 Coordination and communication (UNIMORE)WP 5.2 Dissemination (UNIMORE) WP 5.3 Facilitation of exploitation (UNIMORE)WP5.4 Stakeholder Forum (ADAC)

SP5 Dissemination, user awareness and exploitation (UNIMORE)

WP3.1 Coordination & comm.(ICCS)WP3.2 Test tools development (MET)WP3.3 Test sites set up (ETR)WP3.4 FOT Plans (CERTH/HIT)WP3.5 Large-scale FOT Execution (ICCS)WP3.6 Detailed FOT execution (VTT)WP3.7 Data and User Mngmnt (EMT)

SP3 Field Operational Tests(ICCS)

”Execute”

”Manage”

WP2.1 Coordination and communication (CHA)WP2.2 Methods and tools (CHA)WP2.3 Data specification (LOUGH)WP2.4 Implementation issues (ETR)WP2.5 Liaison (CHA)WP2.6 Technology and service observatory (TID)

SP2 FOT Framework(CHA)

”Plan”

WP4.1 Database usability & accessibility (VTT)WP4.2 Co-ordination and review (LOUGH)WP4.3 Safety impacts (LOUGH)WP4.4 Mobility impacts (VTT)WP4.5 Efficiency impacts (CERTH/HIT)WP4.6 Environmental impacts (IKA)WP4.7 Business models & user uptake (CHA)WP4.8 Benchmarking (VTT)WP4.9 Special analyses and transport system level implications (LOUGH)WP4.10 Technical evaluations (MIRA)

SP4 Evaluation & Assessment(LOUGH)

”Evaluate”

”Open up”

Figure 2: Workplan general structure with Sub Projects, Work Packages and coordinating organisations.





Contract N. 224067





Contract N 224067

2. FRAMEWORK FOR A FOT WITH NOMADIC DEVICES

2.1. The TeleFOT projectA framework for the testing and evaluation of functions provided by nomadic devices has been developed based on the FESTA Handbook and related documentation. The resulting strategy has been adapted to the needs of the test communities and to the functions/systems to be tested. It was introduced in the beginning of the project to support the preparation and planning of the project. It was already from the beginning clarified that the execution and the evaluation of the tests would generate experience and new knowledge that would be fed back to improve the strategy for future use.

The FESTA Handbook sets out the basic principles for a Field Operational Test (an FOT) and in the context of TeleFOT an FOT is generally described as a test conducted under normal operating conditions in an environment typically encountered by the subjects and the equipment being tested. Normally an FOT involves a larger number of users using the services and systems in their daily life in actual use conditions.

2.1.1. L-FOTs vs. D-FOTsThe TeleFOT project consists of Large Scale FOTs (L-FOTs) and Detailed FOTs (D-FOTs). The L-FOTs constitute the core of the TeleFOT project. Thus, the D-FOTs are in specific cases a complement to the L-FOTs, providing additional information for the analysis and interpretation of the results.

L-FOTs are naturalistic studies in the sense that they are studies which involve investigation of normal and everyday use of a set of different functions provided by the platforms of nomadic and aftermarket devices. The studies concerned conditions in which the participants received, used and reacted to functions and services provided to them.

Data was collected over a long period of time from a large number of participants. The studies were experiments in the sense that tests were undertaken in order to find out the answers to questions and hypotheses posed. Nevertheless, they were not controlled experiments even though a rigid test procedure was executed. And some of the research questions were addressed without the introduction of a hypothesis.

2013 of 160

VTT


Contract N 224067

The D-FOTs were carried out as controlled experiments and were undertaken with more control than the L-FOTs, e.g. the participants were asked to drive certain routes, as well as under certain conditions. Furthermore, fewer vehicles and fewer participants were involved and the vehicles were equipped with additional equipment, why more detailed and specific data could be collected.

2.1.2. The FESTA V appliedThe TeleFOT framework builds upon the so called FESTA V (Figure 3) where the first “leg” to the left constitutes the preparing and planning of the Field Operational Tests (FOTs) to be performed and evaluated, i.e. it constitutes the testing and evaluation framework (or strategy). The bottom of the FESTA V is related to the execution of the FOT and the “leg” to the right refers to the evaluation work to be performed.

2013 of 160

VTT


Contract N 224067

Context

Using

Figure 3: The FESTA V

2.2. TeleFOT Characteristics

2.2.1. Impact assessmentIt is important to emphasise that the assessment in TeleFOT was extended beyond the in-vehicle use to include decisions made during the trip but also pre-trip and post-trip. The impact areas addressed were Efficiency, Environment, Mobility and Safety.

2.2.2. User uptakeAnother area of interest was User Uptake which is not an impact area and therefore cannot be dealt with in the same way, i.e. by hypotheses testing and related approaches.

2013 of 160

VTT


Contract N 224067

However, user uptake (user acceptance and adoption) is important to understand as the results demonstrate the potential of the function to be used in reality. It is only when a positive attitude towards a function is observed that any positive outcome of the impact assessments made will have an effect in real life.

2.2.3. Functions testedThe following functions were tested in TeleFOT:

• Traffic information

• Speed information and Speed alert

• Navigation support

• Green driving support

• eCall

In the detailed FOTs aftermarket device lane departure warning and forward collision warning were tested in addition to the aforementioned functions. In some cases in-vehicle ADAS functionality co-existed with the functionalities provided through nomadic devices.

2.2.4. Participants in the testsThe participants in the tests were chosen so that they should mirror the intended user population. Therefore a description of the user population was an important first step in the recruitment and choice of participants. As a consequence a non-randomised sampling procedure was used in the different FOTs. When appointed to take part in the test the participants were informed about the overall purpose of the project and its organisation, as well as about possible risks, the costs covered and not covered, whom to contact in case of breakdown, etc. A formalised agreement about the arrangement between the involved organisations and the participants themselves was made.

2.3. Integrated top-down and bottom-up approachFor the impact assessment a partly new and innovative approach was introduced. An integrated top-down/bottom-up approach was used for generating the research questions

2013 of 160

VTT


Contract N 224067

and hypotheses. The top-down approach was driven by the issues of relevance to the impact area irrespective of the system functionality. And the bottom-up approach was driven by the functionality, the system design, the use cases and the impact area. The process is indicated in Figure 4.

Figure 4: The combined top-down and bottom-up approach

The number of hypotheses was kept to a minimum and the same hypotheses served several impact areas. In Figure 4the generation of the research questions is highlighted, but the most important part of the figure is the iterative process indicated to the right. It emphasises the reduction of research hypotheses by a cost-benefit analysis, i.e. by comparing the cost for measures, loggers, data collection and analysis time with the benefits of understanding the implications of a specific impact area. The process was iterative and took into account the project as a whole, the individual test sites as well as the stakeholders concerned. The result was a suitable number of research questions and hypotheses to be manageable with the existing project resources.

The study design and the selection of participants was a crucial issue at the beginning of the project and it was (due to the limitations in number of potential subjects) recommended that a Within Subject Design should be used in preference to a Between Subject Design. The required number of participants in the FOT depends upon a number

2013 of 160

VTT


Contract N 224067

of factors. A rule of thumb is that a test to be classified as a large scale FOT should involve about 100 participants as a minimum. The calendar time should be long enough to allow for a stable performance to be settled (at least longer than 6 months).

2.4. Data managementPerformance indicators, measures and sensors (i.e. data collection means) were identified and the process from impacts and research questions to measures and sensors is indicated in Figure 5.

Figure 5: The total process from impacts over hypotheses to sensors

2.4.1. The Data Working GroupThe data management process was very early identified as very crucial for the success of the project and it was deemed necessary to create a data co-ordination group (called the Data Working Group) from the very beginning of the project. It was composed of persons involved in all phases of the data management from SP2, SP3 and SP4 to deal with harmonisation of the different elements in the data flow. The topics addressed were part of the total data flow process from data configuration, data acquisition and database structure and management right through to the analytical end-stop.

2013 of 160

VTT


Contract N 224067

2.4.2. Data collectionThree different phases of data collection were identified; In the pre-test phase data about the user was collected, e.g. gender, age, driving experience, experience with functions, in the next phase - d uring the tests - objective data were logged inside the car and were completed by subjective data, like travel diaries and questionnaires (filled in parallel to the test runs). Finally in the post-test phase a subjective evaluation of the system by the user was done.

Subjective data was collected by means of diaries, questionnaires, individual interviews, and focus group interviews. A major part of the subjective data was collected by web questionnaires. A set of questionnaires was designed specifically for TeleFOT and made available to the test sites in the local language in order to ensure that the same type of data was collected across all test sites that tested the same type of functions/systems.

2.4.3. Pilot studiesPilot studies are necessary in preparing a large test exercise and they were completed for all FOTs. These “preliminary field tests” checked the technical function of the data collection systems in real driving and operating situations. Also the feasibility of the evaluation process was tested and it included the use of different data collection methods and tools (questionnaires, diaries, etc.) as well as a preliminary analysis of the data.

2.5. CollaborationFinally the development of the framework has gained by close collaboration with other FOT project, especially the FOT-Net project . The workshops organised on specific topics of interest for the FOT community has been very valuable. Also the work on a revision of the FESTA Handbook based on experiences form TeleFOT and other FOT projects has made it possible to introduce some lessons learned from the TeleFOT approach in the latest version of the FESTA Handbook (version 5.0). The amendments made have made it possible to widen the use of the FESTA Methodology to a broader context than was originally thought.

2013 of 160

VTT


Contract N 224067

3. TEST SITESTeleFOT project performed series of Field Operational Tests (FOTs). As mentioned above, in TeleFOT two types of FOTs were executed, namely large-scale and detailed FOTs (abbreviated as LFOTs and DFOTs respectively) and they were performed in 8 different countries throughout Europe.These FOTs helped in addressing the impacts of different functions all around Europe. A list of different functions was tested in the TeleFOT FOTs.

• Traffic information• Speed limit information• Speed alert• Navigation support• Green driving support• eCall• Forward Collision Warning• Lane Departure Warning

In order to organize and coordinate the planning, set up and execution of the TeleFOT tests, a close cooperation between the TeleFOT Sub-Projects was essential and proved to be very effective since the beginning of the project.

3.1. Test Communities ConceptThe different FOTs implemented within the project can be split into large and detailed FOTs implemented into eight different Test Sites across Europe, covering the North, the Central and the South Europe. In order to coordinate all the activities in the different FOTs, TeleFOT has grouped the Test Sites into three different Test Communities:

Northern Communityo Finlando Sweden

Central Communityo France

2013 of 160

VTT


Contract N 224067

o Germanyo UK

Southern Communityo Greeceo Italyo Spain

Figure 6: Illustration of the TeleFOT Tests Communities and Test Sites.

2013 of 160

VTT


Contract N 224067

The organization of the different FOTs and Test Sites in communities facilitated both the coordination and management of all the activities of the project and the comparison of the FOTs results per geographical areas.

3.2. The key characteristics of TeleFOT FOTsTeleFOT is an example of an FOT, or rather several FOTs. In order to define the key characteristics of the TeleFOT project and its different FOTs, it is necessary to firstly describe the following key concepts: field test, naturalistic study, experiment, and field operational test.

By field tests is generally meant to test something, e.g. a product, under actual operating conditions or in actual situations reflecting intended use.

By naturalistic study is most often referred to a study where researchers/corresponding observe and record some behaviour or phenomenon, often over a longer period of time, in its natural setting while interfering as little as possible with the subjects/participants or the phenomena.

Experiment is defined as a test or trial carried out for the purpose of discovering something unknown or of testing a principle, supposition, etc.

By controlled experiment is meant an experiment that isolates the effect of one variable on a system by holding constant all variables but the one under observation.

A Field Operational Test is generally described as a test run under normal operating conditions in the environment typically encountered by the subjects and the equipment being tested. Normally a FOT involves a larger number of users using the systems and services in their daily life in actual use conditions.

The TeleFOT project consists of large-scale FOTs (LFOTs) and detailed FOTs (DFOTs). LFOTs are naturalistic studies in the sense that they are studies in which will be investigated normal, everyday, use of a set of nomadic and after-market devices and different functions. The studies concern conditions in which the participants receive, use and react to functions and services provided to them and data is to be collected over a longer period of time from a larger number of participants. The studies are also experiments in the sense that tests are undertaken in order to find out the answers to questions and hypotheses posed. Nevertheless, they are not controlled experiments even though as rigid a test procedure is to be executed.

2013 of 160

VTT


Contract N 224067

Also the DFOTs were carried out as experiments in the sense that the tests were undertaken in order to find out the answers to questions and hypotheses posed. However, even though not all DFOTs were carried out as completely controlled experiments, the DFOTs run with more control than the LFOTs, e.g. the participants were asked to drive certain routes, as well as under certain conditions. Furthermore, less vehicles and less participants were involved but the vehicles were equipped with additional equipment allowing more detailed data collection than in LFOTs. The LFOTs constitute the core of the TeleFOT project. The main purpose of running LFOTs and DFOTs across different test sites, and in parallel, is to benefit from the particular strengths of the respective approaches. Thus, the DFOTs complement the LFOTs, providing information for the analysis and interpretation of the results.

3.3. TeleFOT Test Sites and FOTsFifteen LFOTs and thirteen DFOTs were carried out in seven European countries, grouped in the three TeleFOT test communities, covering the North (Sweden, Finland), the Central (France, UK and Germany) and the South (Spain, Italy, and Greece) Europe. In the fifteen LFOTs, the project partners recruited 2382 test users, while data was collected from more that 9 million kilometres. In the thirteen DFOTs combined, the project partners recruited 452 test users and data collected for more than a million kilometres. Therefore overall more than 2800 test users were recruited within TeleFOT in order to participate in the FOTs conducted within the project. The tests targeted specific research questions within TeleFOT project and used a variety of equipment to collect required data. Apart from the objective data recordings, 1760 travel diaries were collected (considered as semi-objective data) and 6400 background and user uptake questionnaires (considered as subjective) from the test participants’. In addition to these, the consortium partners conducted several focus groups and interviews with FOT participants.

The table below provides a summary of the FOTs executed in TeleFOT.

2013 of 160

VTT


Contract N 224067

Table 1: Summary of TeleFOT FOTs.

FOT Function(s)

Number of participants

Total recorded kilometres

Test duration

Finland, LFOT

OulouFOT

GD, SI/SA, TI 140 343802 kilometres September 2010 – October 2011

Finland, DFOT2

eCall Receiving and Handling in PSAP

eCall from PSAP perspective

2 - ½ hour on May 16 2011

Finland, DFOT3

Personal Navigation Device Benchmarking

Static navigation from usability perspective

9 - 3 weeks, June–October 2009

Finland, DFOT4

Tele-ISA

Feedback on speeding for novice drivers

81 593943 kilometres July 2010 –December 2011

Finland, DFOT5

TeleBUS Green Driving Application

Green driving assistance

143 475000 kilometres June 2010 - September 2011

Sweden, LFOT1

The Stockholm test

SA, GD 54 109177 kilometres February 2011 – December 2011

Sweden LFOT2

The Gothenburg test

GD, NA, TI 96 622244 kilometres November 2010-June 2011

(baseline: November 2010-January 2011 & treatment:

February-June 2011)

Sweden, LFOT3 SI/SA, NAV 657 - March - August 2009

2013 of 160

VTT


Contract N 224067

FOT Function(s)



Test duration

MOTION

Sweden, LFOT4 TI 554 428092 kilometres September 2010 – June 2011

(baseline: September 2010-December 2010 & treatment: January-June

2011)

UK, LFOT

BLOM

NA, SI/SA 80 300000 kilometres (approximately)

December 2010 - November 2011

UK, DFOT1

BLOM

NA, SI/SA 25 900 kilometres December 2010 - November 2011 (following

the LFOT conduction)

UK, DFOT2

Foot-LITE

GDS, LDW, FCW

40 4910 kilometres October 2011 – February 2012

UK, DFOT3

Mobileye

FCW, TTC, LDW

23 1980 kilometres January – February 2012

France, LFOT eCall 233 244750 kilometres (approximately1)

July 2011 – November 2012

Germany, DFOT

BLOM and ADAS

NA, SI/SA, plus ADAS (ACC, FCW, LKA)

9 11400 kilometres

August 2011-April 2012

Spain, LFOT1 NA, SI/SA 120 837 729 kilometres September 2010 - September 2011

1 Value based on participants’ estimation reported on background questionnaires2013

of 160VTT


Contract N 224067

FOT Function(s)



Test duration

Valladolid, BLOM

Spain, LFOT2

Madrid, ETRA - CRAMBO

NA, GD 97 Baseline: 863703.654 kilometres

Treatment: 1189101.801 kilometres

Baseline: January 2011 – May 2011

Treatment: June 2011 – December 2012

NA, TI 35 Baseline: 526060.137 kilometres

Treatment: 670665.483 kilometres

Baseline: January 2011 – May 2011

Treatment: June 2011 – November 2012

Spain, DFOT

Valladolid, BLOM

NA, SI/SA 32 450 kilometres Oct 2011 – Jan 2011

Italy, LFOT NA, SI/SA 168 4155133.254 kilometres

September 2010 – December 2011

Italy, DFOT

Navigation, Green Driving and Traffic info support

NA, GD, TI 48 3800 kilometres October 2011- March 2012

Greece, LFOTs NA, SI, SA, TI 148 806 776 kilometres February – December 2011

Greece, DFOTs NA, SI, SA, plus ADAS (LDW, CAS)

40 1305.6 kilometresOctober 2011-March 2012

2013 of 160

VTT


Contract N 224067

FOT Function(s)



Test duration

TOTAL

NA, SI, SA, GD, TI, eCall, LDW, FCW

(plus in vehicle ADAS)

LFOT: 2382

DFOT: 452

Total: 2834

More than 10 million kilometres

recorded

2013 of 160

VTT


Contract N 224067

4. ANALYSIS AND EVALUATIONThe data analysis undertaken within TeleFOT was completed mainly within the final year of the project. The approach that was taken was to conduct the analysis separately within each Impact Assessment on a “research question” basis. Each individual research question was handled by individual partners within the TeleFOT consortium. This was to ensure that there was no duplication of effort across the project and also to ensure that the analysts responsible took a consistent approach to analysing the data when addressing the data from the different test-sites. The full results of the data analysis can be found in each of the individual Impact Assessment Deliverables (D4.3.3, D4.3.4, D4.4.3, D4.5.3, D4.6.3 and D4.7.3). All of these Deliverables are organised such that the main Deliverable document summarises the main results from the Impact Assessment domain and then the full results (by research question) can be found in separate annexes to the main Deliverable document. To analyse the data, the partners downloaded the large-scale FOT data from the central server hosted in by Emtele in Finland. Some Detailed FOT data was also hosted on the central server but where this was not possible (e.g. large video data-files) the data was stored locally and was accessed according to analysis requirements. For ease of use, the Large-scale FOT data was organised into ‘Legs’ files which were essentially files containing individual components of individual trips. These were synthesised and summarised from the logged data ensuring that the analyst did not need to process the raw data in order to undertake analysis. A substantial amount of analysis of the questionnaire data was also made within the project and the questionnaire data was also hosted on the central server. In this chapter, a summary of the main results from TeleFOT are described according to the individual Impact Assessments. The implications of this analysis are presented in Annex 1 (tables A1 to A26).

4.1. User uptake

4.1.1. Use of devices and functionsUse of devices and functions was studied as user uptake research questions “To what extent have the functions/devices been used (before, during, after journeys)?”, “Is there a

2013 of 160

VTT


Contract N 224067

change in the use of functions/devices over time?”, “For which types of journeys are the functions used?” and “In what types of situations are the functions used?”.

According to the travel diary data, a large proportion of the participants did not use the functions offered by the nomadic devices. This could in itself be an indication of a lack of user uptake – with implications for the assessments of the impacts of the respective functions on e.g. efficiency, environment, mobility and safety. Furthermore, there were some reported changes in usage across the test period. Increases as well as decreases in usage can be noted, but in general no statistically significant differences could be found. Quite a few of the participants used the respective functions for particular trips and/or situations. However, the navigation function was used especially for “unfamiliar roads”.

4.1.2. Knowledge and awarenessChanges in knowledge and awareness were studied as user uptake research question “Will problem awareness/problem perception change?”.

In most cases no change in participants’ awareness of different problems associated with use of the car as a means for transportation was found, at least no statistically significant change, can be noted. In those cases where a change has been identified it was not related to the negative environmental impacts of car use but to economic aspects. Overall the participants seemed even more convinced about using their car after the trial than before which can be interpreted as a “negative” change in awareness” but it can also be interpreted as though the participants’ awareness increased but they were not willing to change and had, hence, a need to confirm their choice of mode of transport.

Noticeable is also that a majority of the participants had a fairly strong belief that new technology would play a part in solving both environmental and efficiency problems caused by traffic. The participants may be representative of the ‘average user’ in the different countries but more probable is that the participants have been individuals with a positive, or even very positive, attitudes towards new technology and the functions provided by different devices. They may hence represent – if not the innovators (cf. Rogers 1975) – so the early majority, i.e. individuals who are not among the very first to adopt innovations but who are well ahead of the late majority and the laggards. Such a positive attitude towards new technology can be anticipated to result in a high level of acceptance – but it can evidently not ensure user uptake. It could also be that a strong

2013 of 160

VTT


Contract N 224067

belief in new technology results in high expectations for the technical solutions offered, expectations that may not be fulfilled and hence a disappointment and a rejection of the solutions.

4.1.3. Attitudes and assessmentsParticipants’ attitude was assessed as user uptake research questions “What are the participants’ attitudes towards the devices and functions?” and “Do the attitudes change over time?”. The physical design of the device and the design of the user interface (ease-of-use) have both implications for the usability and acceptance of the function and they were incorporated into a hypothesis relating to user acceptance. Perceived usefulness or benefit was assessed as user uptake research questions “How do the participants rate the benefit(s) of having access to the devices and functions?” and “What type of benefits do the participants notice?”

Even though there are differences, overall the participants’ expectations for the different functions seem high. In general they had a positive attitude towards the different functions, they expected that they would experience large benefits, and they believed that they could trust the information provided. When the functions were first introduced, there was a change; attitudes became less positive, the benefits were rated lower, and there were also negative changes in the participants’ trust in the information. In the final questionnaires, the picture changed again. In general the attitudes became more positive, the benefits larger, and the trust in the information higher. The evaluation curve described, hence, a “U”-shape.

The participants’ assessments of the design of the respective devices (some of which were their own mobile phones) were fairly positive but there were also large individual differences. Their assessment of the design of the user interfaces, and their respective ease of use, more or less mirror their assessment of the design of the devices. Consistent with ratings of other aspects, though, the Finnish and UK L-FOTs participants are less positive than the other L-FOT participants which can be interpreted as though the rating of the one aspect projects on to another. If you cannot interpret the interaction cues and/or cannot read the text, the information may not be available to you or available at a cost that is too high in comparison to the potential benefits. Over time, however, one can assume that at least a major part of the users (the ones who continue to make the effort)

2013 of 160

VTT


Contract N 224067

learn how to make use of the function, the effort associated with using the function is therefore reduced, and the perceived benefit increases. This could be part of the explanation behind the “U”-shaped curve discussed earlier.

4.1.4. Acceptance and user uptakeWillingness to keep was assessed as user uptake research questions “To what degree are the participants willing to keep the devices and functions for future usage?” and “Is there a change in users’ acceptance over time?”. Willingness to pay – affordability was assessed as research questions “What is the participants’ willingness to pay?” and “Is there a change in perceived affordability over time?” and the impact of perceived benefit as “Is user acceptance influenced by perceived benefit?” Impact of physical design of device was assessed as research questions “Does the (physical) design of the device affect user’s acceptance of function/device?”, impact of user interface design as “Does the design of the user interface affect user’s acceptance of function/device?” and impact of trust in information provided “Is user acceptance influenced by perceived trust in function/device?”.

The assumption that acceptance will change over time appears to hold true. If one compares the initial intention of the participants to keep the devices/functions with their stated willingness to keep and willingness to pay for the access, there is in general a decline. The hypothesis that acceptance of function/device will increase with increased use experience should, based on the results, probably be rephrased – it is not use experience that results in acceptance but acceptance that results in usage. Those participants who were the most positive towards the devices/functions were also the ones who used them the most.

The results from the analysis show strong support for the hypotheses that user’s acceptance is influenced by the perceived usefulness – or benefit – of the device/function. Ratings of benefit explained 38% of the variation regarding willingness to keep and 24% of the variation regarding willingness to pay. Some support was also found for the hypothesis that acceptance is influenced by trust in the information provided by the device/function. Ratings of benefit explain 23% and 17% of the variation in willingness to keep and willingness to pay. The physical design of the device accounted for less than

2013 of 160

VTT


Contract N 224067

10% of the differences in user uptake. The results show some support for the hypothesis that users’ acceptance is influenced by the design of the user interface.

4.1.5. Journey qualityImpacts on stress and uncertainty were studied as mobility research questions “Is there a change in user stress?” and “Is there a change in user uncertainty?”. Impacts on feeling of safety and comfort were studied as mobility research questions “Is there a change in feeling of subjective safety?” and “Is there a change in feeling of comfort?”.

A considerable proportion of participants reported that the TeleFOT functions had decreased their stress and uncertainty related to travel. For navigation support the proportion indicating a decrease in stress was 24–50% depending on the FOT (37–50% without the UK LFOT), for speed information/alert 13–38% (27–38% without the Finnish LFOT), for traffic information 18–40%, and for green driving support 10–18%. For navigation support the proportion indicating a decrease in uncertainty was 41–66% depending on the FOT, for speed information/alert 23–39%, for traffic information 14–43% (33–43% without the Finnish LFOT), and for green driving support 4–19%. Consequently, it seems that navigation support reduced uncertainty more than stress, and for the other functions the impact was similar for both attributes. Navigation support had the strongest impact on user stress and uncertainty related to travel, and the green driving support the smallest among the TeleFOT functions.

The results related to feeling of safety and comforts were also positive. For navigation support, the proportion indicating an increase in feeling of safety was 17–37% depending on the FOT (without Spanish LFOT1 27–37%), for speed information/alert 22–62% (42–62% without the Finnish LFOT), for traffic information 15–23%, and for green driving support 10–28%. For navigation support the proportion indicating an increase in feeling of comfort was 18–59% depending on the FOT (42–59% without the UK LFOT), for speed information/alert 6–48% (40–48% without the UK and Finnish LFOTs), for traffic information 24–46%, and for green driving support 10–21%. Consequently, it seems that navigation and traffic information increased feelings of comfort more than feelings of safety, but that speed information/alert increased feelings of safety more than feelings of comfort. Overall, green driving support had the least impact on these attributes.

2013 of 160

VTT


Contract N 224067

4.2. Impacts in travel and driver behaviour

4.2.1. Number of journeysImpacts on number of journeys was studied as mobility research questions “Is the number of journeys undertaken affected in total?”, “Is the number of other home-related journeys affected?” and “Is the number of other journeys (than home- or work-related) affected?)”, and as environment research question “Is number of (car) journeys affected?”. The analyses were based on travel diary data.

We cannot confirm that the TeleFOT functions have an impact on number of journeys, although this is still likely – especially for navigation support. Specifically, in the Italian LFOT, the statistically significant (roughly 10%) increase from the baseline to treatment phases 1 and 2 in the number of journeys in total was due to an equal increase in the number of home-related journeys other than commuting. However, the treatment phase travel diaries were collected during varying months for different test participants. When seasonal trend was eliminated there was still a difference of equal magnitude, but the result was no longer statistically significant.

Furthermore, the data revealed no clear evidence that the availability of a green driving function would influence the number of journeys. The group of participants in the Finnish LFOT who had traffic information and speed information/alert during the baseline and traffic information and green driving during treatment had a temporal decrease in the number of all types of journeys. The group with traffic information alone throughout the trial did not have such changes, nor did any other group with different bundles including the green driving function. Therefore, it is plausible that this decrease was a random difference due to small sample size.

The data showed no clear evidence that the availability of speed information or alert system or traffic information would affect the number of journeys either.

4.2.2. Mode of transportImpacts on the mode of transport were studied as mobility research question “Is there a change in commuting mode of travel?” and as environment research question “Is transport mode affected?” The analyses were based on travel diary data and

2013 of 160

VTT


Contract N 224067

questionnaires. However, the results did not show any evidence of the availability of TeleFOT functions having caused changes in the mode of transport.

4.2.3. Timing of journey Impacts on timing of journey were studied as mobility research question “Is there a change in the departure time of a commuting journey?” The analyses were based on logged and travel diary data. The result showed no clear changes in timing of commuting journeys. This can be attributed to people normally knowing the commuting route and traffic conditions very well and thus having already made a personal optimization of departure time.

4.2.4. Length and duration of journeysImpacts on length in distance and duration of journeys were studied as mobility research questions “Is the length of journeys in terms of distance affected?” and “Is the duration of journeys affected?”, as efficiency research question “Is the travel time from origin to destination affected?”, as environment research question “Is the distance travelled affected?”, and as safety research question “Is the amount of time on the road affected (how long travel takes place for)?” The analyses were based on logged data, travel diary data and questionnaires.

Results indicated that green driving support affects the distance by increasing it and navigation by decreasing it. However, it could be assumed that speed information does not impact the distance driven. In addition, there was no indication that traffic information affected the length of journey either. Specifically, there were statistically significant differences in length of journey in terms of distance in the Spanish LFOT1, where the logged distance was 3% shorter with the bundle “navigation and speed information/alert” than without any functions. Another statistically significant difference was that the logged distance was 10% longer for the function pair “traffic information and green-driving” than for “traffic information and speed information/alert” in the Finnish LFOT. These impacts were also seen when the timing of journeys was included in the analysis. As navigation was not included in the function bundle of the Finnish LFOT and the green driving application showed e.g. momentary fuel consumption on the route, it seems that participants started to favour (longer) routes for better performance. In the

2013 of 160

VTT


Contract N 224067

Spanish LFOT1, logged data were in line with the questionnaire data, but the Finnish LFOT participants reported no change in distance. The UK and Italy LFOT results were contrary to the Spanish LFOT1 results but not statistically significant, and they were very small as absolute numbers. Furthermore, green driving support (with traffic information and speed information/alert as baseline, Finnish LFOT) also showed an increase in distance, while the bundle “green driving and speed information/alert” (traffic information as baseline, Finnish LFOT) showed a decrease. These results were not statistically significant.

Results indicated that green driving may increase journey duration due either to a new route or to slower speed, while bundle navigation and speed information/alert and traffic information may decrease it. Specifically, the statistically significant differences were that the duration was 18% longer in the Finnish LFOT with “green driving” compared to “speed information/alert”, and 10% shorter with “green driving, traffic information and navigation” in the Swedish LFOT2, when compared to the baseline without any functions. These results remained statistically significant when timing of the journey was included in the analysis. The statistically significant impacts of 13% longer duration with green driving than without it in the Finnish LFOT (TI + SI/SA as baseline functions), 5% shorter with “navigation and speed information/alert” in the Spanish LFOT1, and 10% shorter with “navigation and traffic information” in the Greek LFOT3 (NA as baseline function) were not statistically significant when timing of the journey was included in the analysis, although mostly in the same direction. Some other significant impacts were found for certain timings only, e.g. in the UK LFOT there was a 6.8% and in the Italian LFOT a 2.5% decrease in duration during rush hours and a 15–29% decrease in duration during night-time in the Greek LFOT, all of these examples probably due to navigation. The questionnaire data were in line with the logged data results.

However, it must be noted that green driving in a bundle with navigation and traffic information in the Swedish LFOT2 showed a contradictory decrease in duration, while in the Finnish LFOT participants who had “traffic information and speed information/alert” in the baseline and “traffic information and green driving” during treatment showed an increase in duration (in line with the above result). In line with the Spanish LFOT1 result, the same bundle “navigation and speed information/alert” decreased journey duration also in the UK LFOT, although the difference was not statistically significant impact except for rush hour journeys. The Swedish LFOT2 result described above indicates that the

2013 of 160

VTT


Contract N 224067

decrease caused by navigation and traffic information was stronger than that caused by green driving support.

4.2.5. SpeedImpacts on speed were studied as efficiency research question “Are the vehicles’ speeds in the network increased/decreased?” as environment research questions “Is average speed affected?” and “Is speed homogeneity affected?”, and as safety research question “Is speed affected?” The analyses were based on data logged in detailed FOTs.

There were no changes in speed limit compliance due to the use of navigation function. Specifically, there were differences in only one section of one of the DFOTs. The anticipated effect of using speed information and alert function was not ratified by the LFOT and DFOT data analysed. These functions might contribute to increased speed compliance but not enough to have a significant impact upon well known, common journeys.

Results indicated that average speed and speed variance were smaller while using the green driving function than driving without it. Combination of traffic information and navigation seemed to increase speeds, especially in urban environment. Surprisingly, no effect was found for speed information/alert. Specifically, the Swedish LFOT2 and UK DFOT2 showed a significant increase (2%, UK DFOT2) in average speed when driving on urban road types with green driving support. Furthermore, both the German DFOT1 and Finnish LFOT revealed no change in average speed when evaluating the use of speed information/alert. However, the most convincing results come when we consider the effects of driving with traffic information being offered to the driver and when using a navigation system. Results from Swedish LFOT2 show that average speed increased from the treatment phase both for the entire journey and also during urban driving. The only other function used with Swedish LFOT2 was green driving support which as suggested previously will not lead to an increase in average speed. Therefore we can assume that the observed increase was as a direct result of the traffic information and navigation support offered. Results also suggest that these increases in average speed were more apparent in urban driving, with a 6.7% increase. This is entirely plausible as higher traffic densities will usually be present in the urban environment, where up-to-date traffic information and efficient routing to avoid any potential traffic jams will maximise these

2013 of 160

VTT


Contract N 224067

benefits. Results from Swedish LFOT4 show a significant increase in average speed of 4.2% when using the evaluated functions, but only in the urban environment.

The velocities in a traffic network could not be studied in field, or by simulation in TeleFOT.

4.2.6. Route choiceRoute choice as use of different road types was studied as mobility research question “Is there a change in route choice in commuting?”, as environment research question “Is road type and choice of routes affected?”, and as safety research question “Is the route affected (where travel takes place)?” The analyses were based on logged data and questionnaires.

There were two FOTs with statistically significant changes in route choice measured with road types for logged positions due to having TeleFOT functions. In the Spanish LFOT1 there was a shift from higher class roads to lower class roads, e.g. urban streets and unpaved roads in rural areas, probably due to navigation support. In the Finnish LFOT there was a shift from interurban and rural roads to highways probably due to green driving support. In addition to changes in road type distribution, impacts on distance indicate use of new routes. The same applies to duration, as none of the functions had an impact on the traffic situation; if a driver makes journeys faster than during the baseline, it is likely that he/she has found a better route. Here, slight impacts on distance were reported by many participants. Therefore the same functions that affected distance driven, i.e. navigation, green driving support and traffic information, can be assumed to impact route choice also.

Navigation stands out as the function that seems to have the greatest impact. In the Spanish LFOT1 it resulted in shorter and faster routes through the city centre or via shortcuts in rural areas using unpaved roads. Although the new routes used due to having green driving support are longer than the routes used before the impact (using highways instead of smaller roads), this can be seen as positive for mobility because the quality of route can be assessed by multiple criteria (travel time, safety, fuel consumption, etc.), and it is hard so see why any person would change to a route he/she considers worse. Traffic information service seemed to decrease journey duration. Similarly, although no impacts were found on distance or road type choice, the driver

2013 of 160

VTT


Contract N 224067

must have taken a different route, as the information does not impact on the traffic situation itself.

4.2.7. Driving conditionsImpacts on driving conditions were studied as mobility research question “Is there a change in travelling in adverse conditions (dark, fog, slippery road, etc.)?” The analyses were based on questionnaires.

The results showed that none of the TeleFOT functions changed the amount of journeys in the dark or in adverse weather conditions. Nevertheless, there was a small but statistically significant minority for several functions or FOTs who expected at the beginning of the trial that there would be either a slight decrease or a slight increase. However, most of these participants changed their mind towards “no change” during the trial period. The result that the functions do not influence driving in the dark or in adverse weather conditions makes sense, as the dominating factor for driving in the dark or in adverse weather conditions is the time of year.

4.2.8. Vehicle positioning and car following behaviourImpacts on vehicle positioning and car following behaviour were studied as efficiency research questions “Is the time headway between the vehicles increased/decreased?” and “Is the distance from the preceding vehicle smaller/larger?”, and as safety research question “Is vehicle positioning affected (proximity and lane positioning)?” The analyses were based on questionnaires and Greek detailed FOT data.

Several differences in longitudinal position were found in detailed FOT data analysis. The differences found were not supported by the subjective assessment of longitudinal control because of using the TeleFOT functions in different LFOTs. Specifically, a 13% increase in time headway was found with green driving support and ADAS when compared with no functions in the UK DFOT2 and 11% increase of mean time headway with navigation support and ADAS but 16% decrease of mean time headway with speed information and ADAS compared to pure baseline in rural roads in the Greek DFOTs. Furthermore, a decrease was found in standard deviation of time headway when using navigation, speed information and ADAS when compared to pure baseline or

2013 of 160

VTT


Contract N 224067

navigation support in Greek DFOTs. Standard deviation of time headway was found to be smaller with green driving support and ADAS when compared to navigation support in the Greek DFOTs.

The Greek DFOT results indicated that 5% less time was spent with dangerously small headways (<1 sec) in all DFOT conditions compared to pure baseline. The UK DFOT results confirmed this for green driving support and ADAS.

The German DFOT results indicated 10% decrease in mean distance with navigation and speed information/alert compared to ADAS only. A 27% decrease in standard deviation in distance was found with navigation support, speed information and ADAS conditions compared to all other conditions (pure baseline, ADAS, navigation support, navigation support with speed information/alert) in the Greek DFOT.

4.2.9. BrakingImpacts on braking behaviour were studied as safety research question “Is braking affected?” The analyses were based on logged data. After analysing three DFOTs and one LFOT, it could not be confirmed that any of the functions of the nomadic device changes braking behaviour of the subjects. There were some statistically significant changes, but it was not reasonable to reduce those changes to the functionalities of the nomadic device. They could not be verified by other data and the amount of available data was not huge enough to make a reliable statement regarding the influence of the nomadic device on the braking behaviour.

4.2.10. Non-driving manual activityImpacts on non-driving manual activity were studied as safety research question 7 (Is non-driving manual activity affected?) The analyses were based on logged data and hands and eyes videos of Italian detailed FOT.

No significant impact in hands off wheel time for navigation or green driving support. Drivers had to accept the system input for traffic information and confirmed it with a manual operation. Thus 100% interaction when traffic information activated was found but the difference was not significant when compared to all other hands off road actions.

2013 of 160

VTT


Contract N 224067

4.2.11. DistractionImpacts on distraction were studied as safety research question “Does the device cause distraction?” The analyses were based on logged and video data recorded in detailed FOTs.

We could conclude that the navigation function increases eyes off road time when compared to receiving verbal instructions from a passenger as an increase was found in both frequency and duration of glances off road with treatment. For green driving support no change in frequency or duration of glances was found compared to baseline.

4.2.12. Fuel consumptionImpacts on fuel consumption were studied as environment research questions “Is total fuel consumption affected?” and “Is average fuel consumption affected?” Analyses were based on detailed FOT data.

The use of green driving functions was shown to decrease fuel consumption. Specifically, in a Finnish DFOT the fuel consumption in summertime was found to be 0.5–1.2 l/100 km less over the whole route for long-term users, being greatest in night time traffic when using the system compared to driving without it. The impact was greatest in speed limit 30 km/h areas (2.0–5.5 l/100 km less with the device than without it, night time traffic results 2.0 l/100 km being statistically significant). In wintertime, the effect was not as clear. Drivers using the green driving application performed better in night time and rush hour, fuel saving being 0.1 l/100 km and 1.8 l/100 km respectively. A novelty effect was found as novel users of the green driving support had greater fuel savings than the long-term users.

Other DFOTs confirmed the result. The results of UK DFOT2 showed also a significant reduction of fuel consumption on the driven route. Between the baseline and the treatment conditions, 3.5% fuel could be saved on average. In the Italian DFOT, 31 of 40 subjects (77.5%) improved their performance regarding fuel consumption. The mean saving was approximately 4% fuel saving for the 10.8 km section. There was a consistency between the logged data and the perception of the subjects in Italian DFOT.

In the German DFOT, speed limit information/speed alert system in combination with static navigation support was tested. However, no change in average fuel consumption

2013 of 160

VTT


Contract N 224067

was found caused by the functions of the nomadic device (navigation, speed information/alert). Additionally, there was no significance between the different configurations, i.e. that the bundle of functions navigation and speed information/alert had no influence on the average fuel consumption.

4.2.13. CO2 emissionsImpacts on CO2 emissions were assessed as environment research question “Is amount of CO2 emissions affected?” Analyses were based on detailed FOT data. However, it was not possible to carry out a simulation due to the limited data of the FOTs. Therefore no results with regard to the total CO2-emissions could be determined.

With regard to the preceding research question regarding fuel consumption, it is obvious that the green driving support also contributes to a reduction of CO2-emissions. The fuel consumption is directly linked to the CO2-emissions and if a function leads to a reduction in fuel consumption (at least when it is distinct), it also leads to a reduction of CO2-emissions. Due to the reasons mentioned above, these results could not be confirmed with a simulation to quantify the amount of the saving potential.

4.2.14. Delays and jamsImpacts on encountered delays and traffic jams were studied as efficiency research questions “Are there any delays avoided?” and “Are there any traffic jams avoided?” Analyses were based on questionnaires.

Both Navigation and Traffic Information support have the potential to reduce delays. Certainly the use of functions did not increase delays as the number of users who reported a decrease surpassed those reporting an increase. A slight decrease in encountered delays was experienced (perceived) by a significant minority of users, as the majority perceived no change. The percentage of participants perceiving the benefit is varying from 6% in the UK to 40% in Greece. This is reasonable because other factors affecting delays are also crucial. Avoidance of delays largely depends on the specific road and traffic situation driven by the users. In many cases, due to the physical structure of the road network, alternative roads are not available.

2013 of 160

VTT


Contract N 224067

On the whole, FOT participants were relatively positive about the impact that nomadic device functionalities had on their perceptions of traffic jams avoidance. Most users reported that the traffic information function made no change. One third of users, though, said that traffic information function slightly decreased their “getting stuck” in traffic jams. Some participants did indicate that due to access to the traffic information service, their feeling of “getting stuck” in traffic jams was decreased radically.

2013 of 160

VTT


Contract N 224067

5. IMPLICATIONS OF RESULTSImplications were studied for mobility, efficiency, environment and safety by function.

5.1. NavigationNavigation support had positive implications for mobility in all areas of mobility: amount of travel, travel patterns and journey quality. Positive impacts on the amount of travel came from a decrease in journey length and duration. In travel patterns, the implications came from positive impacts on route choice, i.e. finding shorter routes. There were no impacts on other attributes related to travel patterns. Journey quality was improved as user stress and uncertainty were decreased and feelings of safety and comfort were increased.

Navigation support had positive implications also for efficiency by increasing it. Results indicated that navigation support decreases duration by providing alternative routes (e.g. provided “fastest” or “shortest” route) and alleviating existing volume (volume re-distribution). Navigation alone did not affect how drivers perceived avoidance of traffic jams and delays per se. Drivers reported slightly less delays when using navigation support in conjunction with dynamic traffic information. It should be noted, though, that the difference was slight and variant (i.e. variation across sites was observed). Decrease in time spent with dangerously small headways is of importance for efficiency by minimisation in possible disruptions in the network. Navigation support was found to affect standard deviation of time headway and decrease variation. Decrease in variation in headway could lead to increased stability of the network and a more harmonised flow, but increased stability not necessarily offers increased capacity and volume. Users should be informed and trained about using navigation support for more efficient route planning. Drivers are familiar with shortest and fastest routes but not with efficient routes.

The implications of navigation support are small for environment if the navigation support leads to a shorter distance to reach the destination without wasting fuel in traffic jams or on congested roads. In two of the nine research questions, a possible impact of the navigation function could be identified on the environment. While average speed, speed homogeneity, speed distribution, number of journeys, transport mode, fuel consumption and CO2 emissions were not influenced, a change in distanced travelled

2013 of 160

VTT


Contract N 224067

could be observed. Shorter distances to reach the same destinations are a benefit for the environment. Furthermore, drivers increased their use of low class roads with low speed limits when driving with navigation support. Using low class roads which go through the city centres instead of using higher class roads around the city centre may lead to a shorter distance, but it may also take more time if the city centres are crowded. However, at two other test sites (Italy and UK) with the same device the numbers go in the opposite direction, so if there is an influence of the nomadic device and its functions then it is really subordinate. Navigation support could not show any influence on fuel consumption and therewith no influence on CO2 emissions.

The implications of navigation support for safety are twofold. The results of the analysis indicate that navigation has an effect in the following areas relevant for safety: distance travelled, route choice and distraction. Participants felt that both the length and the duration of journeys had decreased with use of navigation. This in turn reduces the amount of exposure to the road which is seen as beneficial in terms of safety. Considering route choice, one conclusion is drawn from the questionnaire data where again the participants considered all journeys they undertook during the study. Here they reported a slight increase in the use of rural roads but no change in the use of highways. Rural roads are associated with a higher accident risk and hence this is potentially detrimental for safety. On the other hand, logged data indicated a slight increase in the use of city roads for comparable origin destination pairs. Turning to distraction, the use of the navigation was shown to increase both the percentage of eyes off road time and the average length of glances away from the road and that these were related to the function use. The literature quantifies a doubling of the risk of accident involvement for glances in excess of 2 seconds. The trials highlighted very few such glances. However, this seems to be a very broad brush rule and a more detailed study should be conducted.

5.2. Traffic informationFor traffic information a positive implication for mobility was found on duration of journeys; as information itself does not affect the traffic situation, this decrease in duration must be a consequence of less congested route choice although length of journey was not impacted. Therefore the implications for amount of travel and travel patterns were seen as positive. In addition, journey quality improved as user stress and

2013 of 160

VTT


Contract N 224067

uncertainty decreased and feelings of safety and comfort increased. The overall implication of traffic information was positive for mobility.

Traffic information has a primary positive implication for efficiency in terms of travel duration, in headway variations, decreases dangerous car following and affects perceived change in delays and traffic jams (small change and great variations among LFOTs and not included in efficiency model as an effect). Traffic information affected journey duration significantly with the least decrease at 10%. One third of drivers mentioned a decrease in encountered delays and traffic jams because of using traffic information. The use of the system was perceived as effective in assisting drivers to avoid traffic jams in large urban areas (i.e. Athens).

Traffic information showed potential to be beneficial for the environment in the analyses of two research questions. On the one hand, it increases the average speed on urban roads; on the other hand the subjects changed their route choice from low class roads to higher class roads which allow driving faster and more fluently. Both aspects have a positive implication for the environment. Traffic information could not affect speed homogeneity, number of journeys, distance travelled or the transport mode. A statement regarding fuel consumption and CO2 emissions cannot be given because no FOT data is available which has information about a traffic information function and a fuel gauge. It is a fact that driving between 60 km/h – 80 km/h (on rural roads) instead of 30 km/h – 50 km/h (on city roads) is more fuel efficient and therewith traffic information might also have an influence on fuel consumption, but that could not be proven with logged data from an FOT.

The implications of navigation support for safety are twofold. The results of the analysis indicate that traffic information has an effect in the following areas; route choice and distance travelled. Both the logged data and the questionnaire data indicate a slight increase in usage of rural roads which could be indicative of a switch to road types with a higher accident risk. On the converse, traffic information also showed through logged and questionnaire data to result in shortened journey durations which in turn represent a reduction in the exposure to accidents. It is not possible to quantify the extent of the positive and negative outcomes for safety in order to assess an overall impact.

2013 of 160

VTT


Contract N 224067

5.3. Speed information and alertSpeed information/alert systems had positive implications for mobility, but this was related only to journey quality. Journey quality improved as user stress and uncertainty decreased, and feeling of safety and comfort were increased. No impacts were observed on amount of travel or journey patterns. The overall implication of speed information/alert was positive for mobility.

Speed limit and alert have only secondary effects in efficiency. The indirect effect because of longitudinal control in speed variations and speed limit compliance was not evident in these studies. Therefore, the implications for efficiency were not found as initially expected. The result could be because of both not affecting longitudinal control to such an extent that significance was attained but also because of the combination of functions. This is not necessarily a negative finding for efficiency. Speed limit information and alert affect time spent with very low time headways when combined with ADAS. Drivers spent almost half time less with very low time headways when using the TeleFOT functions and ADAS. Decrease of accidents could be attained and the stability of traffic flow could be less affected. This finding has a twofold advantage: it affects both safety and efficiency.

The speed information and alert function has limited implications for environment. The speed limit information/speed alert function could only show a change in the speed homogeneity in one of the two analysed DFOTs. The data on which the analysis is based has a poor quality and therefore there is no clear implication to the environment ascertainable. On no other parameter of the environment (e. g. average speed, number of journeys or fuel consumption) is an influence of the function detectable, so that no statistically significant environmental impact of this functionality could be assessed or derived from the results gained in this project.

There are no results that indicate a change in any of the safety indicators with the use of Speed Information / Speed alert. It is therefore concluded that, for the TeleFOT trials do not present any evidence to suggest that there is an impact upon safety.

2013 of 160

VTT


Contract N 224067

5.4. Green drivingFor green driving support the implications for mobility were twofold: negative impacts were found related to amount of travel but impacts related to travel patterns (route choice) and journey quality were positive. The length and duration of journeys increased (negative for mobility), but the changes in route choice were seen as positive for mobility because the driver was able to find a route that he/she considered more economic in terms of fuel consumption. The overall implications of green driving support for mobility were seen as positive despite negative implications for amount of travel.

Green driving support has twofold implications for efficiency. Specifically, green driving support increases journey duration by up to 18% (negative implication). When green driving is bundled up with speed limit information/alert then the increase is greater than when it is used with traffic information. Green driving support was not anticipated to affect avoidance of traffic jams and delays. On the other hand, increase in mean headway and decrease in time spent with very small and dangerous headways is positive for both safety and efficiency. Yet, increase in headways would decrease volume and capacity in heavier traffic with high penetration rates for these vehicles. It should be borne in mind, though, that the primary impact of green driving should be on environment and not efficiency.

Green driving support had positive implications for environment. The green driving application which is primarily developed to save fuel and therewith reduce emissions showed its potential in three FOTs in this project. The fuel reduction was statistically significant with an average of around 4%. Also to reduce fuel, the smaller variance of the speed distribution can contribute. However, the distance travelled was 9.6% longer when using green driving support as the second function compared to speed limit information/speed alert as the second function. A longer distance might be a negative impact on the environment in some cases, but a slightly longer route instead of a congested shorter route has a positive impact on the environment. Since only one FOT could identify a change in distance, it is vague to state that as a possible negative implication to environment.

The implications of green driving support were negative for safety. In terms of safety indicators the TeleFOT trials indicate a potential move towards roads with a higher accident risk and also to longer exposure on the road. Specifically, the results of the

2013 of 160

VTT


Contract N 224067

analysis indicate that traffic information has an effect in the following areas; route choice and distance travelled. In relation to green driving around 5% of participants reported through the questionnaires that they perceived a very slight decrease in the use of highways with a corresponding increase in the use of rural roads. Logged data in the Finnish LFOT showed the function to increase the duration of comparable journey durations.

2013 of 160

VTT


Contract N 224067

6. DISCUSSION ON IMPACT ASSESSMENT RESULTS

6.1. User uptakeThe initial assumption, that user uptake is influenced by several factors, appears to hold true. According to the results, acceptance, measured as willingness to keep the device for usage, seems to be influenced by several factors and perceived benefit is the key. Trust in information provided and design of the user interface and hence perceived ease-of-use, are two other important factors whereas the physical design of the device itself does not seem to play such a significant role. These findings are in accordance with the findings in other studies. Future development of the type of functions that were tested in TeleFOT must take these aspects into account. However, what is perceived as e.g. beneficial from a user perspective depends most probably on a number of factors. For instance, some participants used the functions across all types of journeys whereas others experienced the functions as beneficial only for some. In addition, the traffic situation in the area where the individual undertakes his or her journey may not be perceived as problematic and information on e.g. road works or queues are hence not considered to provide any additional value. These contextual factors on user uptake also include the personal factors that determine whether a device is of benefit to the driver. For example an individual may feel more or less uncertainty when driving an unknown road or may be particularly concerned with adhering to speed limits.

The same factors play a significant role for willingness to pay but overall, the factors explain less of the variability regarding willingness to pay than the willingness to keep, possibly because a major part of the participants were not willing to pay for access to the devices/functions. Nevertheless, willingness to pay for access to the functions appears to exist among a group of participants. The sum that a majority of these is willing to pay each month for access to a function is within the range of 1–10 EURO. A small part of the participants have indicated a higher monthly fee (predominantly 11–25 EURO). Evidently this reluctance to pay could be interpreted as a lack of user uptake but at least part of the explanation must also be sought in the fact that several information services, such as for instance navigation support and traffic information, are now (2012) offered as e.g. apps and for free – provided that you have the required type of smart phone. Willingness

2013 of 160

VTT


Contract N 224067

to pay may therefore no longer be a relevant factor to consider in order to understand user uptake of the type of functions tested in TeleFOT.

Benefit, trust in information provided and design of user interface are factors which have been identified as influencing User Uptake as willingness to keep and willingness to pay. In addition, previous knowledge of devices and functions had some impact but in this case the more previous knowledge, the more negative the impact. Based on the TeleFOT results it is not possible to conclude if problem perception (as measured) has any considerable impact on acceptance or the use of devices and functions. Access to the different devices and functions did not appear to increase the participants’ problem awareness of problems related to car use or their compliance with societal aims – apart from the impact of some of the functions on their compliance with speed regulations.

Choosing willingness to keep as an indicator of acceptance can evidently be questioned, in particular if the participants are aware that they may keep the devices and functions free of charge. One could argue, for instance, that acceptance of a device and/or function could be understood as usage, i.e. if a device/function is used this is an indication of acceptance. The devices and functions were not used by a large proportion of the participants which could indicate a low level of acceptance. However, if one compared the use of the green driving function and the speed information/alert function they were used more frequently than, for instance the navigation function. This function was used for less frequent and for more specific situations: for longer journeys and for unfamiliar journeys. The difference in use frequency make sense in that the green driving function as well as the speed information/alert function provides information which the driver can potentially make use of every journey (support green driving and inform/warn against speeding) whereas navigation does not make much sense if a major part of the journeys made are commuter and other frequently made trips. This does not mean, however, that navigation is not used for the few trips that are, indeed, made to unknown destinations and where this benefit is large enough to motivate acceptance of the devices and functions. The comments made by some participants support such a claim. Thus, when understanding users’ adoption and rate of adoption of the kind of functions tested in TeleFOT, the different characteristics of the functions must be taken into consideration. Furthermore, use frequency is, as such, most probably not an appropriate indicator.

2013 of 160

VTT


Contract N 224067

As has already been addressed, the participants’ expectations before the trials were high. When the functions were first introduced attitudes became less positive, the benefits were rated lower, and there were some negative changes in the participants’ trust in the information. Post-trial the attitudes became slightly more positive, the benefits larger, and the trust in the information higher. The evaluation curve describes a “U”- or a “TeleFOT V”-shape. There are evidently several possible explanations to this “V”-shape but the notion of adoption or user uptake of new technology as a process which requires time is worth considering. Adoption is more than a positive attitude towards the idea of new technology – overall the TeleFOT participants had a positive attitude towards and a belief in technology as a tool for change. Adoption of new technology is a process during which the individual learns about the new technical solution, familiarizes with it, finds out about its possibilities and limitations, and draws conclusions regarding when to use the new technical device and for what purposes in order to benefit from it. At any stage the individual may decide to reject the solution. Such a process takes time, for some individuals longer than for others, probably longer than the months allocated in the TeleFOT FOTs. If this is so, the importance of running field tests for longer periods of time in order to understand adoption (or user uptake) processes must be underlined.

Furthermore the importance of collecting data at certain intervals is emphasized. A before and after approach may indicate a very negative trend whereas if one takes into consideration data collected at several instances during the trials, another pattern may emerge. In this case the trend may be understood as a positive one. Without several “measurement points”, the TeleFOT “V”-shaped evaluation curve would not have been identified. The experimental approach, with a before and after design, may be the most efficient way to determine (possible) effects of different types but can be questioned as at all relevant for studying user uptake. An individual may have a certain attitude before the introduction of the new technology and another one after a certain period of time. An individual may have certain expectations before a trial and may then experience whether or not the expectations were fulfilled but none of these actually reflects adoption. User uptake is when the new technology, the new device or function, is adopted and integrated into everyday life, i.e. invested, used and made use of – in the case of the functions tested in TeleFOT - in relation to planning and undertaking journeys by car and/or by other means of transport. What appears as more important in this case is to follow the process over time, from when the functions and devices are first introduced

2013 of 160

VTT


Contract N 224067

over time to when a “steady state” can be anticipated, and to investigate how and when the new devices and functions are used, if the information provided is made use of, and why – or why not. These are questions that cannot be addressed in a short experimental study during which an individual is allowed to, or obliged to, use a function or system for a short period of time in controlled conditions. These are questions that require a field test which lasts for a long period of time.

6.2. MobilityThe main results show that aftermarket or nomadic devices providing driver support functions have a positive impact on mobility at a personal level. Specifically, the quality of travel improved for all functions tested, as stress and uncertainty related to travel decreased and feeling of safety and comfort increased. Furthermore, the amount of travel improved with navigation support and traffic information. However, it deteriorated for green driving support, resulting in the use of longer routes and longer travel times. The travel patterns improved with new routes with navigation, traffic information and green driving support.

The increase in distances and durations with green driving may be partly explained by the features of the particular green driving application used in the Finnish LFOT. As navigation was not included in the function bundle and the green driving application showed e.g. momentary fuel consumption on the route, it seems that participants started to favour (longer) routes for better performance. Sometimes longer route with slower speed can be a better choice from the standpoint of total fuel consumption. However, if momentary fuel consumption is in focus, the total fuel consumption may not always be minimised although momentary fuel consumption would. Thus an intelligent combination of navigation support and green driving is recommended.

The findings suggest that participants had high expectations of positive impacts for all driver support functions. However, when the functions were activated a clear deterioration was seen in attitudes. Nevertheless, the FOTs lasted a relatively long time and the participants had time to learn how to benefit from the functions — when, where and for what kinds of journeys. At the end attitudes were much more positive and participants agreed on the benefits of the functions.

2013 of 160

VTT


Contract N 224067

The fact that TeleFOT functions did not affect many aspects of travel patterns (timing, driving conditions and transport mode) is understandable, as the devices were located in the car and the functions were used after starting the journey by the participant’s own car. Thus the functions were not targeting the original decision to make the journey, the timing or the mode. In addition, information on adverse conditions was provided during the journey, not before it.

Impacts were assessed based on commuting and other frequently made journeys, as their number was high and the same leg could be identified during both the baseline and treatment phases. However, the elasticity of commuting is the least with fixed timing and destination. Therefore, the impact potential of other journeys might have been higher although difficult to assess with FOTs.

In conclusion, although the absolute impacts of driver support functions that aftermarket and nomadic devices provide are not high, they are positive for various dimensions of personal mobility. It should be noted that these small impacts may be an indicator of longer term impact potential. As travel patterns are hard to impact, it takes time – probably longer than a 14-month period, which was the longest duration of FOTs in TeleFOT.

Many impacts especially in Mobility impact assessment domain were assessed based on commuting and other frequently made journeys, as their number was high and the same leg could be identified during both the baseline and treatment phases. However, the elasticity of commuting is the least with fixed timing and destination. Therefore, the impact potential of other journeys might have been higher although difficult to assess with FOTs.

Travel diary data was the most important data set for the Mobility impact assessment. However, each travel diary dataset (one week) represented a single season while logged data was divided into baseline and treatment and especially the treatment phase was long covering several seasons of the year. This partly explains the different results obtained from these two data sources.

2013 of 160

VTT


Contract N 224067

6.3. EfficiencyIncreasing traffic efficiency is a goal aiming at harmonised traffic flow, optimal traffic volume, increased road capacity, less accidents, and greater accessibility for all involved road users. Consideration in TeleFOT is made solely for passenger cars and no other involved road users that are affecting traffic elements (e.g. modal choice or change). Efficiency is considered the traffic efficiency determined from the driver’s perspective.

Apparently not all functions affect traffic efficiency and not all functions play the same role (either primary, secondary, or neutral) in different impact areas. As anticipated, navigation and traffic information seem to be the most important function for efficiency. Speed limit information and speed alert have a secondary role. The effect of green driving support is not clear; it seems conflicting. It appears that green driving support increases travel duration (negative effect) but decreases variation – as much as almost the rest of functions – in headways. The effect of functions and ADAS appeared to be complementary and not conflicting wherever it was possible to investigate separate impacts. The combined effect of ADAS and TeleFOT functions is more complicated than to be expressed as additive.

It is important to mention that a set of sustainable structural indicators could be set for monitoring the effect of functions to traffic and driver efficiency that could infiltrate the development of the next generation of assistive and information systems and would prove useful for strategies within policy making for nomadic device deployment. These indicators would be time duration (e.g. journey) and headways as they are both easily and universally measurable. Objective assessment for delays and avoidance of traffic jams should be based on more complex algorithms than certain sets of assumptions and restrictions. Subjective evaluation is actually correlated to exposure but these would hold true also for objective assessment for different pilot sites.

The main limitation for Efficiency impact assessment area was the lack of simulated large scale scenarios for certain traffic related parameters and indicators. The absence of a macro and microscopic perspective restricted the estimation of actual impact on traffic and the quantification of direct effect to traffic efficiency. Therefore, only estimations were made for the impact of certain functions and/or bundles of functions on traffic, in general. Certain parameters were revealed to have more potential for efficiency impact

2013 of 160

VTT


Contract N 224067

assessment (e.g. distance travelled) than others but their importance is valid for the framework they were investigated with.

A second limitation lies in the adopted experimental assumption. In many cases, it was impossible to estimate actual use of the nomadic devices and its functions but it was possible only to accept users’ intention to use the function. This assumption limits also the generalisation of findings as repeatability is uncertain when based solely on intent.

Indirect extrapolations to EU27 were carried out for very rough estimations of the potential benefit for European traffic networks. These estimations, though, lack statistical power and as no simulation modelling was performed they could not be implemented for different types of roads, networks, and road conditions.

A third limitation lies in the sample size of some of the studies. Unavoidably, large scale studies had many more participants than small scale efforts. Some of the research questions were answered based on small samples and others with larger samples. Therefore, the whole power within the efficiency impact assessment area was impossible to be calculated. The power is closely related to the repeatability of the findings.

It is important to consider these limitations when interpreting the results. The fourth limitation follows the third one. The combination of LFOT and DFOT data for answering the same Research Questions did enrich the data sources but also created unexpected confounders for data analysis. For example, distance keeping behaviour is probably different in DFOTs where testing is somewhat controlled and in LFOTs where the setting is naturalistic. The methodology should be refined for combining not only data sources but also designs in order to become more robust to environmental violations and still preserve the ecological validity of large scale efforts.

6.4. EnvironmentFrom four tested functions three (navigation support, traffic information and green driving support) showed an implication on the environment in different manners. First of all, green driving support which was designed to help the driver to drive in a more economic and fuel efficient way, shows a huge impact on fuel efficiency. The FOTs which tested this function had the opportunity to log the fuel consumption directly and therewith they provided a good data quality to evaluate the impact of this function. The average fuel

2013 of 160

VTT


Contract N 224067

saving was about 4% compared to the baseline condition. With regard to the society, a single application on the navigation device or a smart phone can easily contribute tremendously to a more fuel efficient driving (if the advice is not ignored or disregarded by the driver) and along with this to lower CO2 emissions.

The traffic information function could show its benefit to the environment when guiding the driver from low class roads onto higher class roads which have higher speed limits and a more fluent traffic flow. The optimal ratio between driving speed and fuel consumption is between 60 km/h – 80 km/h. So, the benefit for changing from low speed roads to high speed roads is obvious.

Whether shorter journeys are beneficial for the environment or not, needs a closer review. Shorter journeys are beneficial if the absolute fuel consumption of the journey is less compared to the original, longer journey. The opposite is true if the journey is shorter, but due to slow moving traffic for instance, the absolute fuel consumption is higher. This must be considered when drawing a conclusion about the impact of the navigation function on the environment. With the available information, a final statement cannot be given for this function.

When talking about the benefit not only for the driver himself, but for the whole society, every function which saves fuel and therewith emissions, contributes to one big objective: to reduce as much emissions as possible. This societal value is clearly identified for the above mentioned functions and therewith a really valuable outcome of this project.

6.5. SafetyThe conclusions from the data analysis in respect of the impact of single functions supported aftermarket nomadic devices upon safety are as follows: Throughout the project there has been no measurable event (e.g. accidents or injuries) during the baseline and experimental conditions since these are relatively infrequent discreet events. Therefore it is impossible to quantify impacts on safety since cannot say injuries went up or accidents fell in relation to any function use.

As a result we are left to infer the impact on safety associated with the use of the functions supported by the TeleFOT devices under assessment through specific safety

2013 of 160

VTT


Contract N 224067

research questions. These research questions are in themselves only generic indicators of the risk of accident and as such are not easily quantified.

According to mobility research question on feeling of safety, the function that is perceived to have the greatest benefit for safety is speed alert but no logged data supported this. Traffic information showed a slight increase in the use of rural roads in the Swedish FOT which could represent a switch to road types with a higher accident risk but also resulted in shorter journey times which reduce the exposure to accidents. These two effects have apposing impacts on safety. Green driving indicated a potential increase in exposure and accident risk due to a shift in road type and longer journeys. The result on distraction is perhaps the most measurable in relation to safety since eyes off road time increased due to the function and the length of gazes also increased. This was the case for Navigation but not Green Driving.

The HMIs are different, green driving is an image associated with a simple response required whereas navigation requires more cognitive processing to interpret the information. Although there were a few glances over 2 seconds, these occurred both to the navigation and equally to other areas of the visual scene,the 2 second guide is a rough rule of double risk which needs to be contextualised further with vehicle speed and complexity of road environment which influence crash risk. The devices used to provide the functions were limited; alternative devices with a diversity of HMI’s may provide wider ranging and perhaps different results.

The market is evolving quickly and has even moved on during the life of the project. There is now more use of smart phone applications to provide functions, an example of a change in HMI. We have to ask the question ‘Is the HMI comparable to TeleFOT devices in relation to for example distraction?’ Hence there is a need to keep evaluating how the changing HMI effects in particular driver distraction. Overall, the main limitation of the Safety Impact Assessment was that the usual metric for studying the ‘safety’ of a system did not apply. That is, within the TeleFOT FOTs there were no accidents and incidences reported therefore it is difficult to determine whether safety is really well and truly affected by After-market and Nomadic devices. Instead, inferences were used from metrics that are traditionally thought to affect safety although the relationship is not always clear or well–established. For example, both vehicle speed and long-term exposure of a vehicle to the risk of accident are documented throughout the research

2013 of 160

VTT


Contract N 224067

literature as factors that influence accident rates, but the relationship is such that the probability of an accident with increased speed or exposure is never 100%. Therefore some caution needs to be applied when interpreting the results of the Safety Impact Assessment.

Small sample size particularly was the case for the analysis relating to Safety research question 3 (“Does the device cause Distraction?”) In order to address this question in its entirety, it would have been necessary to examine on a frame-by-frame basis all of the video data for all of the drivers at all of the test-sites collecting video data. However, with the available resources and project duration, it was only possible to analyse the video data for a sample of participants for which visual behaviour data was available. For these drivers, it was also only possible to select certain segments of the route driven. Whilst every effort was made to ensure that the segments (road junctions) that were chosen were both the most demanding in a visual and cognitive workload sense and where reliance on the nomadic device was high (and hence whether the risk of distraction was thought to be at its maximum), the fact remains that the sample size is small overall and nothing can be learned regarding the driving during other segments of the route. An initial attempt was made to overcome this dilemma by automating the visual data collection process (using eye-tacking equipment) but this was not successful as it became apparent that eye-tracking equipment, which works extremely well in carefully controlled laboratory environments does not work well in situations where the environment is very variable and changes rapidly (due to differential lighting).

Another limitation concerned Safety research question 5 (“Is lane position affected?”) This is an important research question rightly included in the original Analysis Plan since it was hypothesised that the lateral and/or longitudinal positioning of the vehicle could be affected through using the device thereby increasing crash-risk. However, once the FOTs had commenced, it became apparent that there would be little available data that could satisfactorily identify whether a lane deviation had occurred, particularly within the Large-scale FOTs. Whilst the logged data was an excellent recourse for examining certain characteristics of individual driving, the sampling rate for most test-sites was not sufficient to identify whether a lateral lane deviation had occurred. For longitudinal positioning, no logged data was available at all and therefore the analysis required to address the research question was reliant on subjective information. In this case, it is

2013 of 160

VTT


Contract N 224067

postulated that participants would firstly not be aware whether a lane-deviation had actually occurred and secondly would vary in their individual interpretations of the term “deviation” in a lane-positioning sense.

6.6. Safety Comparability of Results across the Impact Assessments

In several cases, the results of the TeleFOT were applicable across several impact assessments. For example, a Navigation System was found to increase the speed of vehicle in and around urban areas which is beneficial for traffic efficiency but on the other hand, because of increased vehicle speed, may increase the risk of crashes (and hence affect Safety) within the network. Therefore, whilst benefits may be experienced through use of some after-market devices, there is a need to consider the impact that the device or function has in other domains rather than viewing them in isolation. Table 2 highlights the impacts of the functions and devices tested in TeleFOT and shows where the corresponding benefits and dis-benefits were found to occur. User uptake has not been considered in this table since it is not considered to be an impact per se.

In many cases, it is difficult to determine the optimum set-up. For example, a navigation system may significantly reduce travel time but it is difficult to quantify the benefit that this gives the driver when compared to the dis-benefit that may be experienced by possibly causing the driver to become distracted through using the device. An attempt to examine this issue in more detail has been undertaken in Deliverable D5.3.2 where benefit-cost analyses of Navigation and Green-driving systems have been undertaken.

2013 of 160

VTT


Contract N 224067

Table 2: Comparability of results across the impact assessments.

Function tested

Positive Benefit for Negative Benefit For Comments

Navigation Safety (decrease in distance travelled) Mobility (better informed choice about

transport mode, route choice)

Safety (increase in use of rural roads, increased EORT)

Mobility (increase duration of journey -exposure)

Navigation function plays a secondary role on Efficiency

Traffic Information

Safety (decrease in distance travelled, decreased exposure to urban roads)

Efficiency (fewer traffic jams encountered, fewer traffic delays encountered, shorter journey duration)

Mobility (shorter journey duration)

Safety (increased average speed on urban roads, increases exposure to rural roads)

Positive benefits thought to outweigh negative aspects for this function

Green Driving Advisory

Safety (increases time-headway, decrease speed)

Environment (variance in speed distribution smaller, Environment (decrease in fuel consumption, CO2 emissions)

Safety (increased exposure due to greater distances travelled)

Efficiency (increased journey duration)

Environmental benefits are thought to outweigh the negative aspects of this function

ADAS Safety (increases time-headway) No objective effects observed

Speed Limit/Alert

Perception of Safety and reduction in driver stress

Usually available as one of a number of functions on a bundle

2013 of 160

VTT


Contract N 224067

6.7. General Limitations of the ResultsNo study of this type could produce results that were not limited to some degree as have already been described within the discussion of the Impact Assessments. It should be noted that most of these limitations were recognised at the beginning of the project and specific measures were introduced to overcome these. However, some of the limitations are still important to highlight as follows.

A first limitation involved the project relying on third parties for input to the project. In many cases, the input of third-parties was highly beneficial since it implied enhanced user-awareness in the local communities and extra input to the project with little extra cost. However the conduction of an FOT requires very careful planning accompanied by a risk analysis and mitigation strategies in case the risk occurs. There was the potential for this input to affect several parts of the FOT, depending on the role of the third party. For example recruitment could be affected when external organizations (as opposed to individuals) are involved, or functionality provision in the case that a service provider is involved. These limitations were evident in some of the TeleFOT FOTs.

Another limitation concerned data-logging and system usage. In some cases, Logging did not include actual use of the system, thus the impacts of actual use could not be analysed and implications for function level had to be assessed based on function bundle level results. The analysis phase very clearly highlighted the need to log the use of the system, as it was difficult to differentiate between the effects of the service itself from the effects of the service just being available. To a large extent, this limitation was overcome through use of the subjective data, for example User Uptake questionnaires and travel diaries.

The majority of the TeleFOT FOTs were performed as a within-subject (or before-and-after) study design. There are limitations following such an approach, such as the variability of the context linked to the timing of the tests and the potential learning effects form one experimental condition to another. A between-subject design was used only in some rare cases but with the limitations inherent in that approach, i.e. the challenge of identifying a matching control-group and the need for double the number of subjects to be involved. Ideally, with “unlimited” resources (of time and money), a

2013 of 160

VTT


Contract N 224067

between-subject design could have been used more frequently in the FOTs throughout the whole testing period. This would have ensured that the analyses could separate the effects of different services from changes due to factors such as economic trends, seasonal changes, etc. Such an approach was in most cases abandoned from the very start of the project for practical reasons. However, the within-subject design chosen (in the Large Scale FOTs) made possible a comparison of data between baseline and treatment conditions (before and after) as data from data loggers was used in combination with subjective data collected by questionnaires, travel diaries, etc.. Therefore the confidence in the results overall is high.

Another possible general limitation concerns the sample sizes within the projects. As described elsewhere within the various TeleFOT Deliverables, the test-sties varied with regard to the size of the sample of participants that were tested. As a general rule of thumb within the project, it could be assumed that there is greater confidence in the results from the Large-scale FOTs compared to the Detailed FOTs – the L-FOTS used larger numbers of participants and the tests were conducted over a relatively long time-frame (some up to 17 months) therefore confidence in the results of these longitudinal (long duration) FOTs is relatively high given the timeframe and numbers of subjects involved. However, it should also be remembered that the Large-scale FOTs were conducted using Data Acquisition Systems with lower-frequency data-recording capabilities (1hz in some cases) and this might have restricted some of the analysis of the logged data at least (for example, this sampling-rate is not discrete enough to allow determination of vehicle positioning on the road). However, this limitation was balanced with subjective data (collected by means of questionnaires and travel diaries, etc.) to be included in the analyses made.

For the Detailed FOTs, most were conducted over a much shorter time-frame using smaller numbers of subjects and hence the confidence in the results is possibly not as high. Conversely for the DFOTs, it should be remembered that the Data Acquisition Systems used were much more precise with higher sampling rates for logged data and this may offset some of the limitations of using smaller samples and shorter timeframes. The sample size for each FOT together with the duration of test and the research questions that used the data from each of the FOTs are shown in table 3. As can be seen from the table, data from the large-scale FOTs were much more likely to be used for

2013 of 160

VTT


Contract N 224067

addressing multiple research questions compared with data from the Detailed FOTs whereby the data was predominantly used for addressing only one or two research questions (DFOTs being complementary to LFOTs on selected crucial issues). This was the intention from the beginning and purpose of using the two types of FOTs (i.e. LFOT and DFOT) within the project.

Another potential limitation in terms of interpretation of the results is the nature of the samples and whether they are truly representative of the general population from which they were drawn. Such a general representation is not always the best to capture the true nature of the functions/systems in use as certain function can have a dedicated target group of users. In most cases, the participants were recruited from the test-site locality and were given incentives to participate in the FOTs. Given that the participants were willing to be involved in the first place could potentially raise some questions about whether their experiences with the functions tested within the FOTs can be truly reflective of the “ideal” function user and also creates some difficulties with scaling up the results to the European dimension.

A final general limitation of the results and interpretations of them is the fast-moving pace of technology. The TeleFOT project began in 2008 just as the technology surrounding after-market and nomadic devices was beginning to proliferate. At the end of the study, the technology had evolved substantially so that the results, whilst possibly generalizable to the first or second generation of after-market devices could not so easily be applied to the state-of-the-art. This is especially true at the end of the project (late-2012), as many of the functions tested had already migrated from smart-phones to apps built on APIs and to a new world that is dominated by platforms and technology-enabled services. It is notable that any industrial stakeholders who want their product to still be relevant on the market have to meet the rapid technological development in the ICT sector with new demands on agility, i.e. they need to create a mix of flexibility and speed in their activities in order to survive.

2013 of 160

VTT


Contract N 224067

Table 3: The test-site FOT data that were used to address each Research Question.

Test-site and FOT

DFOT/LFOT

Sample Size

Duration RQ Addressed

Finland LFOT 140 14 months SRQ1, SRQ2, SRQ4, SRQ5, MRQ1, MRQ2, MRQ3, MRQ4, MRQ5, MRQ6, MRQ7, MRQ10, MRQ11, EFFRQ1, EFFRQ2, EFFRQ5, ENRQ4, ENRQ3, ENRQ5. ENRQ6, ENRQ7, UURQ1, UURQ2, UURQ3, UURQ4, UURQ5, UURQ6, UURQ7, UURQ8, UURQ9, UURQ10, UURQ11, UURQ12, UURQ13, UURQ14, UURQ15, UURQ16

Finland DFOT2 2 1 day SRQ4, ENRQ1, ENRQ2Finland DFOT3 9 3 weeks ENRQ8Finland DFOT4 81 18 months ENRQ8Finland DFOT5 143 16 months ENRQ8Sweden LFOT1 54 11 months SRQ5, MRQ1, MRQ2, MRQ3, MRQ4, MRQ5, MRQ6, MRQ7, MRQ10, MRQ11, EFFRQ2,

ENRQ6Sweden LFOT2 96 8 months SRQ1, SRQ2, SRQ4, SRQ5, MRQ1, MRQ2, MRQ3, MRQ4, MRQ5, MRQ6, MRQ7, MRQ10,

MRQ11, EFFRQ1, EFFRQ2, EFFRQ5, ENRQ1, ENRQ2, ENRQ4, ENRQ5, ENRQ3, ENRQ6, ENRQ7, UURQ1, UURQ2, UURQ3, UURQ4, UURQ5, UURQ6, UURQ7, UURQ8, UURQ9, UURQ10, UURQ11, UURQ12, UURQ13, UURQ14, UURQ15, UURQ16

Sweden LFOT3 657 6 months MRQ1, MRQ2, MRQ3, MRQ4, MRQ5, MRQ6, MRQ7, MRQ10, MRQ11, EFFRQ2Sweden LFOT4 554 10 months SRQ2, SRQ4, SRQ5, MRQ1, MRQ2, MRQ3, MRQ4, MRQ5, MRQ6, MRQ7, MRQ10, MRQ11,

EFFRQ1, EFFRQ2, EFFRQ5, ENRQ1, ENRQ5, ENRQ6UK LFOT1 80 12 months SRQ1, SRQ2, SRQ5, MRQ1, MRQ2, MRQ3, MRQ4, MRQ5, MRQ6, MRQ7, MRQ10, MRQ11,

EFFRQ1, EFFRQ2, ENRQ4, ENRQ5, ENRQ6, ENRQ7, UURQ1, UURQ2, UURQ3, UURQ4, UURQ5, UURQ6, UURQ8, UURQ13, UURQ14, UURQ15, UURQ16

UK DFOT1 25 12 months SRQ32013

of 160VTT


Contract N 224067

Test-site and FOT

DFOT/LFOT

Sample Size

Duration RQ Addressed

UK DFOT2 40 5 months SRQ3, SRQ4, SRQ6, ENRQ1, ENRQ2, ENRQ3, ENRQ8UK DFOT3 23 2 months SRQ4, EFFRQ4, EFFRQ6France LFOT 233 17 months eCRQ9, eCRQ11, eCRQ12, eCRQ13Germany DFOT 9 9 months SRQ4, SRQ6, EFFRQ3, EFFRQ4, EFFRQ6, ENRQ1, ENRQ8Spain LFOT1 120 12 months SRQ1, SRQ2, SRQ4, SRQ5, MRQ1, MRQ2, MRQ3, MRQ4, MRQ5, MRQ6, MRQ7, MRQ10,

MRQ11, EFFRQ1, EFFRQ2, ENRQ4, ENRQ5, ENRQ6, ENRQ7 UURQ3, UURQ4, UURQ8, UURQ13, UURQ14, UURQ15, UURQ16

Spain LFOT2 SRQ2, MRQ1, MRQ2, MRQ3, MRQ4, MRQ5, MRQ6, MRQ7, MRQ10, MRQ11, EFFRQ1, EFFRQ2, EFFRQ5, ENRQ5, ENRQ6, UURQ1, UURQ8

Spain DFOT 32 4 months SRQ3, SRQ4, ENRQ2, ENRQ3Italy LFOT 168 16 months SRQ1, SRQ2, SRQ4, SRQ5, MRQ1, MRQ2, MRQ3, MRQ4, MRQ5, MRQ6, MRQ7, MRQ10,

MRQ11, EFFRQ1, EFFRQ2, ENRQ2, ENRQ4, ENRQ3, ENRQ5, ENRQ6, ENRQ7, UURQ1, UURQ2, UURQ3, UURQ4, UURQ5, UURQ6, UURQ8, UURQ13, UURQ14, UURQ15, UURQ16

Italy DFOT 48 16 months SRQ7, ENRQ8Greece LFOT 148 11 months SRQ2, SRQ6, MRQ2, MRQ3, MRQ4, MRQ5, MRQ6, MRQ7, MRQ10, MRQ11, EFFRQ1,

EFFRQ2, EFFRQ5, ENRQ4, ENRQ5, ENRQ6, ENRQ7, UURQ1, UURQ2, UURQ3, UURQ4, UURQ5, UURQ6, UURQ8, UURQ13, UURQ14, UURQ15, UURQ16

Greece DFOT2 40 6 months SRQ6, EFFRQ4, EFFRQ6

2013 of 160

VTT


Contract N 224067

7. REACH-OUTOne of the main goals of TeleFOT was to create awareness about the project and its results, by spreading and promoting them to partners, stakeholders and the European audience.

Activities devoted to Dissemination User Awareness and Exploitation were planned and carried out under SP5.The main objectives of SP5 can be grouped into three main areas:

7.1. DisseminationDissemination activities focus on the need to 1) create awareness of the project content, achievements and the work done, by spreading and promoting them to partners, stakeholders and the European audience; 2) share experiences, results, achievements in general, gained within TeleFOT, so that they can be utilized by other organizations and individuals interested in such an area.

2013 of 160

VTT


Contract N 224067

In such a perspective, thorough Planning, wide Spreading and continuous Assessment, as key elements of TeleFOT’s Dissemination strategy, exposed by a so conceived Dissemination Plan (D5.2.1, distributed since the project beginning), constitute the Critical Success Factors for the TeleFOT Project.

7.1.1. Dissemination StrategyThe different tools and Media have been used and combined in different ways during the four project years, in order to effectively reach one of the two overall aims of the project: to raise awareness on the project results towards an as wide as possible audience.

During the first project year of TeleFOT the Dissemination activities have been aimed to make the public aware of the project’s objectives and expected results. The TeleFOT official website was developed and leaflets and posters were designed and presented during events: most of the dissemination material produced at this stage is going to be used throughout the project life.

It has to be pointed out that, at this stage of the project, Dissemination activities were not limited to specific groups, but rather addressed to the public in general in order make it aware of the project’s existence and goals. Publications and presentations aimed at describing mainly the project’s concept and approach, as well as research methodology.

After this period, it was needed to use these tools and activities to effectively spread the TeleFOT identity, scopes and first relevant results towards more targeted groups.

During the second project year efforts have been carried out to prepare the Field Operational Tests and to allow them running, defining the Framework as well as Evaluation and Assessment activities. First feedback has been received from Stakeholders via the First General Stakeholder Forum. The newsletter played an important role in spreading information about progresses on-going in the project and the number of planned events increased.

In the third project year FOTs run and results from them were expected. This implied a more targeted use of the Media and a major participation to events, while important feedback was expected from the local Stakeholder Fora.

2013 of 160

VTT


Contract N 224067

During the fourth project year it has been important to spread the results achieved, aiming to reach an as wide as possible audience. Differently from the first project year, on the base of results achieved and on the impact assessed in the project, it has been also important to target them towards specific audiences, as well as to show how the TeleFOT results represented a starting point for future research and deployment.

Error: Reference source not found shows messages and tools identified for Dissemination.

Figure 7: Messages and tools

7.1.2. Target GroupsAs TeleFOT is mobilizing a relevant amount of resources working together to achieve quite ambitious and important results, the awareness of content, intermediate and final results towards the European audience has a crucial significance for success.

In particular, when new knowledge is created it has to be spread towards specific target audience: this implies that contents, media, formats and language used in getting the

2013 of 160

VTT


Contract N 224067

outcomes into the hands (and minds) of those target audience has to be shaped on them, in order to favour awareness, commitment and sharing.

Target groups of Stakeholders have been identified since the project beginning. Target Groups and the strategy to address them are extensively described id D5.2.1. and D5.3.1.

The main Target Groups are represented in Error: Reference source not found, where also the scope of their involvement in TeleFOT is pointed out.

Figure 8 – TeleFOT stakeholders

According to the Dissemination Strategy they have been reached in different ways: presentations during events and conferences, events organized (including Stakeholder Fora), liaison and networking activities, dissemination tools.

7.1.3. Dissemination tools and materialProject logo

2013 of 160

VTT


Contract N 224067

The project logo was created at the very beginning of TeleFOT, representing the visual reproduction of TeleFOT’s core elements: road and devices working on board vehicle. It is the starting point for the definition of the project’s image: the characterizing element for its brand.

The Brand Handbook, as an Annex of the Dissemination Plan and Brand handbook (Deliverable 5.2.1) contains all the rules for the correct use and replication of the TeleFOT logo over the whole internal and external material regarding the project, distributed to any persons and/or organizations.

Website and Wiki

A TeleFOT website was created early in the project and regularly updated, serving as the front face of the project, and has been given high priority from the very beginning of the project. The website is uploaded to the URLs: www.teleFOT.eu and www.teleFOT.eu.org.

An area reserved to Partners is accessible via a redirect to the project Wiki from the website homepage, via the “Partners Only” orange button on top right.

The Wiki (http://telefot.openinno.fi) continues to serve as an internal web site for the project members. In the Wiki files are uploaded, such as templates, documents and Deliverables. Furthermore, the Wiki represents a collaborative area where texts and media can be shared in a continuously updated and collaborative way.

Posters, leaflets, hand-outs

At the project beginning a leaflet with an introduction to the project and contact information has been produced, as well as a poster with brief information on the project, to be used e.g. for meetings or events where TeleFOT is presented. They have been updated and are available on the project website and on the wiki and follow the general “look and feel” stated for TeleFOT materials.

The last version of the General project poster has been revised in occasion of the TeleFOT Final Event.

For this occasion also other posters have been created to show the work done in the project:

• SPs Roll up posters

2013 of 160

VTT

http://www.teleFOT.eu.org/

http://www.teleFOT.eu/


Contract N 224067

• Posters on Test Sites organization and work (Error: Reference source notfound- Test Sites posters)

• Posters on Impact areas showing objectives, main outcomes and partners involved

The posters have been designed in respect of the project identity, using the official project logo, colours, fonts..., with a graphical layout coherent to that defined and used for all the Dissemination materials.

A notepad has also been realized and distributed to partners, having a graphic layout (fonts, colours, graphic elements, TeleFOT logo, EC logos,...) coherent to other Dissemination tools.

The leaflets, posters and any hand-outs are thoroughly described in particular in D5.2.2.

When a TeleFOT event was organized - such as Stakeholder Fora - a Dissemination package was distributed to participants. This was distributed also during the Final Event, containing:

Project leaflet; full programme and map of the exhibition; agenda; speakers’ biographies; badge; gadgets: a usb stick and an electric power adapter customized with the

TeleFOT logo; TeleFOT notepad; a pen customized with the TeleFOT logo.

FOTs have been provided with guidelines to produce their own FOT logo and a Business Cards, showing the key elements identifying their FOT and in particular:

TeleFOT logo

Map of the Test Community your FOT belongs to

Car picture: one (if DET) or three (if LSC) cars

2013 of 160

VTT


Contract N 224067

Map of your test site (e.g. map of Italy)

Name of your FOT (e.g. Reggio Emilia Test Site)

Logo of the partners involved in the Test Site

Business card

including also details to contact the team managing the FOT.

In Annex 4 of D5.2.2 and in D5.2.5 TeleFOT logo and Business card are showed.

Newsletters were published on the TeleFOT website and distributed periodically, especially when a relevant event occurs, with the aim at stimulating a debate (which is supposed to continuously grow during the project life) on project topics, results, and news related to TeleFOT scope.

Events and Liaison

During its whole life, TeleFOT targeted other projects in order to engage them and benefit from potential cooperation during or after the TeleFOT project: participation to conferences, seminars and other activities and in particular creating a link with other European and International projects, working in the same or related areas (e.g. euroFOT, FOT-Net, TeleFOT INCO, Prologue, Dacota), contributes in building awareness of the new projects, their results, methodology and tools among Scientific Community, OEMs, decision makers.

Participation and presentation to events such as conferences, workshops, etc. (e.g. ITS WC 2008, 2009, 2010, 2011, ICT4EE, TRA 2010, FISITA 2010, ITS Lyon, TRA 2012, FOT-Net workshops and seminars, … ) has been encouraged and supported. The list of events is reported in D5.2.1.

Video

A video introducing the whole project has been produced, and updated during the 2nd

project year to illustrate the project’s functions and work: the project concept and general information, including its EC framework and overall aims, FOTs organization and Test

2013 of 160

VTT


Contract N 224067

Communities, the project objectives, its organization, the Consortium and functions to be tested.

During the PY4 a new version of the video has been created.

A thorough description of the video is available in the D5.2.3 – Short Report on Project Video.

Media center

A Media Center was created at the University of Modena and Reggio Emilia, coordinated by the TeleFOT Dissemination Manager, leader of SP5.

Through the Media Center, all the relevant material from the project and in particular from Test Site Communities was collected, selected and transformed into information, results and events, to be spread to stakeholders and to as wide an audience as possible.

Assessment of Dissemination activities

A complex project as TeleFOT, with many partners and resources involved, needs clear and transparent communication between participants and awareness about the matching between goals planned and reached.

In such a sense, all TeleFOT Dissemination activities have been subjected to an assessment procedure, in order to ensure their highest level of efficiency and effectiveness.

The criteria that objectively assess the effectiveness of communication strategies adopted referred to:

• size of audience achieved (e.g. web site visitors, participants of a user forum, television share in case of a programme in which TeleFOT is presented, etc.);

• Audience responses and attitudes (e.g. interest gathered from a message, positive and proactive initiatives stimulations, positive feeling on TeleFOT);

• Direct and indirect involvement of opinion-leaders (e.g. political actions taken after TeleFOT media exposure, etc.);

2013 of 160

VTT


Contract N 224067

• Project visibility.

Criteria have been coupled with the techniques needed to assess them and the threshold that should be overtaken to consider communication as a success each year.

The assessment procedure was composed by a qualitative and a quantitative analysis and is extensively described in D5.2.1. Results are available in D5.2.5.

Moreover, according to Task 5.2.4, periodical reports have been provided about the satisfaction rate of dissemination activities, in order to allow modifications in processes and tools or in strategy.

7.2. ExploitationExploitation activities carried out in TeleFOT (in particular in WP 5.3 (Facilitation of exploitation) covered exploitation planning, which aims to raise awareness among partner industries and other relevant parties on the undertaken work, promotion of results achieved, and ensuring their sustainability from the commercial point of view.

During the first Project year foundations had been laid in order to have Business cases and potential new functions identified at M24 as planned. E.g. crash tests have been performed by ADAC, being an example of exploitable results for the industry. D 4.8.1 Review of consumer test and standards" and "D 4.8.3 Crash of fitting systems" (as part of Benchmark) have been produced.

Since Exploitation was considered as a relevant activity for the project, in the second Project year a Workshop on Exploitation was organized on March 18th 2010 in Brussels.

It gave very important feedback on WP 5.3 activities and on D5.3.1, in particular on:

Main needs to be covered in WP 5.3, namely:

Raising awareness on results achieved;

Ensuring their commercial sustainability.

2013 of 160

VTT


Contract N 224067

Tools, including TIP and ERCs; CRM; Business Models (WP 4.7) and Cost Benefit Analysis (T5.3.4).

Deliverable D5.3.1 - Business cases and potential new functions (which was renamed into D5.3.1. Business cases, potential new functions and Technology Implementation Plan in order to include the TIP).

This deliverable was updated periodically, since different ERCs are supposed to be individuated by partners during the project life.

During the Workshop on Exploitation all the SP leaders identified potential Exploitable Result Cases on the base of activities carried out in their own Sub Projects.

Once ERCs have been shared with the project Consortium, they have been presented to target stakeholders.

The ERCs partners described are available in D5.3.1 and on the TeleFOT CRM.

A CRM (Customer Relationship Management) was set up. It is accessible online by project partners, allowing them to identify strategies to catch customers’ attention, to log each step of the relations from the very informal beginning and to improve the marketing strategies time after time and learning from previous experiences (all properly stored).

During the third project year collection of ERCs has continued among TeleFOT partners. Partners were invited to find new Exploitable Result Cases and to enlarge the list of ERCs discussed during the Workshop on Exploitation.

Partners were also invited to access and use the TeleFOT CRM and to insert contacts (including those from the SHF).

In PY4, when the project closed, it was important to understand the potential of marketable project results, in a perspective of future deployment.

Deliverable 5.3.2 – ‘Markets and market potential of TeleFOT functions and services’ was delivered at the end of the project. It aimed to understand the potential of the marketable project results, bridging the gap between research and a marketable innovation.

The evolving picture in the Nomadic Devices market was taken into consideration, from the project beginning to its end.

The main objectives of D5.3.2 were to:

2013 of 160

VTT


Contract N 224067

Point out the TeleFOT major outcomes (ERCs and BCs identified in the project are described in a dedicated chapter);

Understand the target stakeholders to which such results should be addressed. In D5.2.1 ‘Dissemination Plan and Brand Handbook’, target groups have been identified. They have also been presented in D5.3.1 – ‘Business cases and potential new functions’;

Identify new Business Cases for TeleFOT: after depicting the scenario at the beginning of the TeleFOT project, this chapter aims to outline the market trends in smartphone usage and to understand how the diffusion of apps is changing our lives. This can help to identify the evolution of the potential market of TeleFOT, from its beginning to its end (2008-2012). Moreover in this chapter the role of TeleFOT results in this new world is highlighted;

Illustrate what are the likely benefits and associated costs of potential new functions and the conditions under which benefits/costs are effective to help justify decisions for their possible implementation: the Benefit-Cost-Analysis (BCA) conducted in TeleFOT is described in the deliverable.

The importance of the Exploitation activities in TeleFOT and the tools to support them was highlighted also during external events (e.g. the FOT-Net 5th Stakeholders workshop).

7.3. Stakeholder ForaStakeholder Foras were planned in TeleFOT in order to effectively help understand possibilities and limitations of aftermarket and nomadic devices among the key players and the driving public.

Key objectives of the Stakeholder Fora were to:

Keep stakeholders informed on project results and achievements

Consult stakeholders

2013 of 160

VTT


Contract N 224067

Achieve wide awareness of the possibilities of the aftermarket and nomadic devices among different interest groups

Make TeleFOT objectives effective and in line with its target audiences’ needs.

Two kind of Stakeholder Fora have been organized during the project life: General and Local Stakeholder Fora.

During the first Project Year the foundations had been laid to set up the First general Stakeholder Forum at the beginning of the Second Project Year.

The TeleFOT First General Stakeholder Forum on Aftermarket Nomadic Devices in Field Operational Tests took place at M20, on December, 9. It was organized in combination with the Nomadic Device Forum meeting, which was held on December, 8 2009.

An area dedicated to Stakeholder Fora news has been created on the TeleFOT website, to extend the aim to effectively raise awareness of the possibilities and limitations of aftermarket and nomadic devices, among the key players and the public reached by the events, as well as to all people interested.

At the beginning of the third project year, during which the TeleFOT vehicles were entering the road, core issues of the project need to be incorporated into a wider framework which could help to identify the pillars required to build safer, more efficient and greener mobility in conjunction with other EC-funded projects focusing on the Field Operational Test and Naturalistic Driving methodology.

In this sense there was a perfect opportunity to include the Second TeleFOT General Stakeholder Forum within the 4th FOT-Net Stakeholder Forum, organized in cooperation with the European projects euroFOT, TeleFOT, PROLOGUE and DaCoTA. Moreover the TeleFOT event closed three days of events focusing on FOT and Naturalistic Driving experiences and the FESTA methodology (read the article below).

On December 2nd, 2010 the FOT-Net event in Brussels was attended by a wide range of stakeholders and experts who were interested in the general FOT approach.

2013 of 160

VTT


Contract N 224067

From November 2010 to July 2011 workshops with local TeleFOT stakeholders were organized in order to support results of the General Fora with input from local participants and stakeholders:

Finland: Oulu, November 10th 2010

Italy: Reggio Emilia, November 22nd 2010

Spain: Valladolid, November 23rd 2010

UK: London, January 21st 2011

Germany: Aachen, February 3rd 2011

Sweden: Stockholm, February 10th 2011

Greece: Athens, May 13th 2011

France: Lure, Malbouhans, July 6th 2011

Figure 4 shows the differences between the two kinds of events (General and Local):

2013 of 160

VTT


Contract N 224067

Figure 9 – Characteristics of the Fora

7.4. Other Stakeholder-related activitiesIt is important to mention here that several other initiatives were addressed within TeleFOT in order to approach stakeholders and be aware of the status of aftermarket and nomadic devices market.

In 2010, in addition to the Second General Stakeholder Forum organized in Brussels together to the 4th FOT-Net Stakeholders meeting, it is important to mention that on March 24, 2010 TeleFOT joined the FOT-Net 5th Stakeholders workshop and presented its experience in the session “How can we transfer the results of FOTs to stakeholders”.

In 2011 and 2012 the dialogue with stakeholders has continued, in particular:

Additional local meetings related to TeleFOT National stakeholders have been arranged, in connection to other existing initiatives coming from individual partners. Although these additional meetings have had a wider scope than just TeleFOT, these have been useful in getting an idea of the future priorities and plans of relevant stakeholders.

2013 of 160

VTT


Contract N 224067

The liaison and cooperation activities with FOT-Net and other FOT-related projects have continued. TeleFOT consortium partners have participated in different European and international events and conferences (such as FOT-Net workshops, ITS World Congresses, etc.), either in an active way (as speakers) or as attendants.

Finally it is important to mention that TeleFOT has received funds from the European Union for International Cooperation activities focused on sharing of projects results: TeleFOT INCO (TeleFOT International Cooperation project), as an International extension of TeleFOT, has given the opportunity to meet with partners from different extra-European organizations, as TeleFOT INCO Sister projects partners or external organizations involved in different initiatives.

2013 of 160

VTT


Contract N 224067

8. LESSONS LEARNTThis section is a summary of the lessons learned from the project experience in regard carrying out a FOT, and a FOT for nomadic and aftermarket services. The detailed information can be found from TeleFOT deliverables. A lot of this has also been fed directly into the general European FOT guidance i.e. the FESTA methodology.

8.1. SP2 FrameworkGenerally, the lessons learned in all of TeleFOT have been conveyed to the European FOT guidance (FESTA) via SP2.

The FESTA Handbook was used as the first approach for the implementation, execution and evaluation of the FOTs. The functions tested were Traffic information, Speed Limit Information, Speed Alert, Navigation Support (static), Navigation Support (dynamic), Green Driving Support, and eCall. The characteristics of these functions imply that they must be studied in a wider transportation context than just traffic and cover also transport and travel. Furthermore, the user must be studied in different roles as driver, passenger and traveller and the trip time line must be considered (i.e. trip planning and use during and after the trip). A consequence of that the platform used for the functions was not a vehicle but a nomadic device (to be used in a vehicle but also elsewhere) made it necessary to complement the FOT activities with controlled experiments. This resulted in a proposal for modification of the FESTA Handbook where also the well-known “FESTA V” was expanded to include a block “CONTEXT” to capture the wider use of the functions addressed (i.e. also outside the vehicle). Finally, a new and innovative top-down approach was added to improve the identification of research questions and hypotheses and in order to minimise the work at hand a prioritisation of the hypotheses using a cost-benefit approach was proposed and applied. These four proposed modifications have been accepted and are now found in the latest version 5 of the FESTA Handbook (2012).

8.2. SP3 FOTsBased mostly on the D3.5.1 / D3.6.1 conclusions section, which in turn are based on the conclusions / lessons learned provided per test site:

2013 of 160

VTT


Contract N 224067

Generally the project partners have a long history with testing new functions in vehicles as controlled tests. In such tests generally some 20 or more drivers are invited to a test track to try out a novel system. However, running large-scale or detailed field operational tests (LFOTs and DFOTs respectively) at the same time and collaboration between partners helped us to improve especially in two aspects, which are easy to neglect in small-scale tests: user agreement details and data handling.

Having a group of lawyers to go through LFOTs and DFOTs user agreements that had more than ten pages with attachments, we got plenty of feedback and were able to improve partner’s existing checklists for topics to cover in test agreements. In short, the agreements got longer to cover more situations and e.g. describe data privacy in depth. Although it might be an obvious point, it is worthy to stress out that the legal issues related to the tests should be followed carefully; the laws should be respected.

With regard to data handling, there are similar needs in larger scale and small scale tests e.g. to verify test data for correctness right after a test, in order to avoid having to redo tests. In subsequent experiments, more software tools may be created for quick validation purposes, to reduce manual work. As a large scale project, data harmonization and analysis by multiple partners, was considered at every step. This work resulted in detailed metadata documentation to present tests for a group of analysts. Post-processing was harmonized so that results and indicators would be easier to compare and same analysis methods could be used.

On the execution side, relying on third parties for the test conduction may have significant advantages, but it requires careful planning followed by a carefully planned risk / mitigation strategy. A variety of incentive schemes were used by the local test site management teams to attract test participants addressing the project needs while taking into account the local peculiarities. When testing with nomadic devices one should consider the rapid changes in nomadic devices market, as they may induce significant modifications to such long term tests like the FOTs conducted in TeleFOT; potential changes should be taken into account during the planning phase as it might be difficult to apply modifications in execution phase.

Such lessons learned become highlighted after running 15 large scale and 13 detailed FOTs with more than 2800 test participants recruited and above 10 million kilometres of data recorded in TeleFOT.

2013 of 160

VTT


Contract N 224067

8.3. SP4 AnalysisThe impact assessment phased several challenges that needed to be overcome to be able to perform the work. First challenge dealt with how to co-ordinate data issues within the project. The challenge was overcome by initiating the data working group with SP2. Second challenge was also related to data, specifically with late flowing of data within project. However, with careful planning of the data analysis, substantial piloting and hard work at the end of the project we were able to manage the work within timeframe. The amount of data created the third challenge: how to analyse 1Hz data. We ended up in producing and analysing summary data (summary tables describing legs) which turned out to be efficient way to deal with the 1Hz challenge.

Fourth challenge was related to large number of partners in impact assessment and overlapping research questions in different impact assessment domains. We overcame that challenge by dividing the work by research question allocating similar research questions to the same partner whenever possible and by careful co-ordination and communication throughout.

Last and positive challenge deals with stakeholders and how they now have a clearer idea of questions as they know what data are available. It seems that we produced the answers that are now stimulating more questions! Thus to answer all the questions posed by the stakeholders, we leave to future projects.

Lessons learnt from the above:

Establish a data working group from the outset – no matter how much the project feels it does not need it.

Make sure the analysts are fully involved from the beginning – not joining at a later stage

Limit the numbers of research questions and hypotheses to a manageable number Study first the impacts on travel and driver behaviour together and only after that

divide the analysis work to impact areas to avoid overlapping work Specify the data ahead of the FOT beginning Engage fully with stakeholders – ensure that their needs are met Decide how logged data will be managed Pilot the FOTs – as much as possible

2013 of 160

VTT


Contract N 224067

The pilot the data analysis – as much as possible Never under-estimate the amount of time needed for analysis Never under-estimate the amount of time needed to summarise the analyses as a

readable deliverable Never try to be over-ambitious with the data analysis Decide how to handle your data before data collection starts Plan for the data analysis as far as is reasonably practical – ad hoc creativity just

creates unharmonised approaches Prepare for the unexpected Have a coping strategy for the unexpected Accept the fact that despite best intentions, the final analysis will not give all of

the answers required and more analysis will be required and/or will be possible

8.4. SP5 DisseminationActivities devoted to Dissemination and Exploitation of project results as well as to Stakeholders involvement should be carried out throughout the project:

WHEN HOW

At project beginning Definition of clear objectives, target groups and respective strategies

During the project Continuous involvement of stakeholders

Make project objectives effective and in line with its target audiences’ needs

Evaluation of effects of Dissemination activities:

Check that strategies are working. Adjust them if needed

Make sure the target groups are reached and that they understand the

2013 of 160

VTT


Contract N 224067

outputs and results. Understand how project findings may

impact policy-making and influence deployment

Ensure that Press is interested towards the project (e.g. check if any article is published on the project)

During and at the end of the project Exploit the project results:

derive further and greater benefits from project achievements and results

Think about how your project’s findings may be re-used from other projects and initiatives

2013 of 160

VTT


Contract N 224067

9. CONCLUSIONS

9.1. General conclusionsIt is generally acknowledged that nomadic devices such as portable navigators and smart phones have thoroughly penetrated the market during the last decade. For example, hundreds of millions of smart phones that are capable of GPS navigation are sold worldwide on a yearly basis. Simultaneously the same has also happened to thousands of services that are provided through those nomadic devices - including services directed to drivers and travellers. Even though both the devices and services usually go through a thorough testing cycle during their design and production phases, up to now there still has been a very small amount of transparent unbiased scientific information available on the positive or negative effects of those services to the driving task, e.g. in terms of safety, efficiency and mobility.

TeleFOT is now able to shed light on these issues based on analysed real-life subjective and objective data – data that has been collected from millions of kilometres driven by thousands of normal everyday users while interacting with services such as navigation, green driving support, real-time traffic information and speed limit information.

TeleFOT findings and recommendations help steering policy making and investment to a more cost-effective direction and help the industry design better and more attractive products. Thus the project outcomes help enhancing the well-being of Europe in general by contributing towards a smarter and more cooperative transport system. This was the main TeleFOT objective from the outset and upon closure TeleFOT as a project was able to meet its set objectives. Furthermore, TeleFOT has during its lifetime in many ways enhanced the FOT methodology itself.

2013 of 160

VTT


Contract N 224067

9.2. Summary and conclusions on key findingsThe main benefits of the functions were perceived by the participants to be Convenience (easy access to information), Comfort (less uncertainty, fewer driving errors), Economic (less cost) and Environment (fewer emissions).

Of the tested devices, navigators and traffic information systems, in particular, increased efficiency by allowing drivers to find quicker and less congested routes. Up to 45% of participants, particularly those in large cities, reported that the Traffic Information function helped them to avoid travel delays and traffic jams. Green driving systems guided drivers to routes that lowered their emissions, and towards driving more economically. Green driving advisory systems were found to reduce fuel consumption by up to 6%.

The use of a green driving system in a bus fleet helped to lower fuel consumption and to reduce speeding, which also improves road safety. Another significant finding is that the systems reduced driving-related stress and anxiety across the board and, in all the participating countries, increased the drivers’ sense of safety and driving comfort. From the perspective of mobility, the results were positive for all systems.

The users’ expectations for the services were high at first. After using the services for some time, they were slightly disappointed not to have seen a direct benefit. The longer they used the services, the more clearly they could see the benefits and advantages, and the more satisfied they were. Participants assessments of the designs of the devices were positive but there were some negative views. Acceptance of the devices changed over time – acceptance results in usage rather than vice-versa.

There is no evidence to suggest that the TeleFOT functions affected Mode of Transport and Timing of Commuting journeys.

Eyes off road time was found to increase when the navigation function was introduced although the Green-driving function did not change visual behaviour.

Navigation support has positive implications in all areas of Mobility and many aspects of Efficiency but the function effects are small for Environment. Some effects for Safety are evident in terms of distance travelled (reduces) and distraction (increases).

2013 of 160

VTT


Contract N 224067

Traffic Information is positive for Mobility in terms of reduced journey duration and reductions in stress and uncertainty. It was also positive for Efficiency in terms of reduced travel durations, reduced headway variations and perception of avoidance of congestion. The impact of this function on Environment was inconclusive.

Speed Information/Alert ha a small but mainly positive effect for Mobility, Efficiency and Safety.

The Green-driving function was found to decrease fuel consumption but increase journey duration. Average speed was found to decrease with use of the Green-driving function. Average speed and speed variance were both smaller with this function. The Green-driving function had a positive effect on Efficiency and Environment. This function had a mixed effect on Mobility and a negative effect on Safety (by changing exposure).

9.3. Data access for third partiesThe aim of the TeleFOT project is to disseminate the results of the project to a wide audience outside of the TeleFOT partnership. This includes the European Commission, TeleFOT associate partners (e.g. national organisations that have directly or indirectly supported the TeleFOT data collection), stakeholders of the road transport system and the scientific community. Such an audience may need the TeleFOT results for understanding the wide-scale implications of the uptake of aftermarket and nomadic systems, particularly with regard to Safety, Mobility, Efficiency, Environment and User Uptake.

The Large-scale FOT (LFOT) data has been collected and incorporated into a set of electronic systems which are held centrally and which have been accessed by the partners for analysis purposes. Due to the complexity of these data and the statistical requirements for their analysis it is envisaged that future access after the project duration will be through the IP co-ordinator.

The Detailed FOT (DFOT) data is held and has been analysed mainly at a local level although data has been made available to other partners for analysis during the project lifetime (to address specific research questions) on request.

2013 of 160

VTT


Contract N 224067

Overall, the working principle remains that the data are fundamentally a European resource and that they should be used and exploited as widely as possible both within and external to the partnership. However, data distribution is to be handled according to a specific protocol. The following principles are therefore proposed:

The process in regard sharing the TeleFOT data (to be validated by TeleFOT consortium).

TeleFOT Partners may under conditions that are specified here, transfer a subset of anonymized TeleFOT data for research use for a third party. This provision only applies to the part of data, that the TeleFOT Partner has a full ownership of. [Typically the test site leader or test leader organisations VTT, CHALMERS, TRA, UTBM, LOUGHBOROUGH, MIRA, CIDAUT, ETRA, UNIMORE and ICCS; additionally also IKA, CRF and CETRH/HIT, but only detailed FOT data.] Consent and permission for the data transfer also need to be acquired from any third party local funding and/or cooperation organisations; and also from other members of the TeleFOT consortium that have ownership of the intended data to be transferred. After these permissions have been granted but prior to any data transfer, there will need to be an NDA (Non-Disclosure Agreement) signed between the TeleFOT Partner and the intended third party recipient of the TeleFOT data; this NDA will need to detail the data access rights and restrictions and the data usage rights and restrictions. The transferred data or any part of it may not be handed over in any format to fourth parties without TeleFOT Partner and VTT consent and a new NDA or an amendment to the original NDA.

The intended research use needs first to be presented in the format of a research proposal written by the third party. Third party shall submit to TeleFOT Partner and VTT a written research proposal wherein the need for access to the TeleFOT data is described. As such, the proposal format is free, but the language needs to be English. The proposal must at the bare minimum i) contain a description of how the data will be used in the intended research; ii) describe the objective for this research; iii) describe the method, format and content of publishing the research results and iv) to nominate explicitly the persons who will have access to the data

2013 of 160

VTT


Contract N 224067

with full viewing rights. This research proposal needs to be annexed to the associated NDA.

In addition, prior to any data transfer TeleFOT Partner and VTT will check and verify that the transfer will not be in conflict with either the “Test user agreements” (written contracts between TeleFOT Partner and the test users of TeleFOT test sites) or with anynational and/or European Union level data and privacy protection legislation. TeleFOT Partner and VTT commitment for any data transfer will be valid for up to 18 months after the end of the TeleFOT project.

Some general principles regarding data from the TeleFOT subjects are also necessary and these are proposed to be as follows;

Publication:

Published results will be anonymized

No information will be published that would allow individuals to be identified

If the results contain any personal data (see below), the specific consent of the data subject will need to be obtained.

Personal data:

Data relating to a living individual who can be identified from that information or from that data and other information in possession of the data controller. Includes subject name, address, telephone number and id number. Also includes expression of opinion about the individual, and of the intentions of the data controller in respect of that individual.

2013 of 160

VTT


Contract N 224067

REFERENCES

TeleFOT Deliverables 2008-2013 - TeleFOT Consortium

2013 of 160

VTT


Contract N 224067

ANNEXES

ANNEX 1: List of TeleFOT deliverables

ANNEX 2: Results by Research Question and Hypothesis

2013 of 160

VTT


Contract N 224067

ANNEX 1: List of TeleFOT deliverables

Deliverable ID Deliverable name PUbl

ic /

COnf

iden

tial

IPD1.1 Project presentation PUD1.2 Website PU/COD1.3 Crash tests of nomadic and aftermarket devices PUD1.4 Test communities description PUD1.5 Testing and evaluation strategy PUD1.6 Field Operational Tests plans PUD1.7 Evaluation plan PUD1.8 Usability of nomadic and aftermarket devices PUD1.9 Stakeholder Forum activities *) PUD1.9 Stakeholder Forum activities PUD1.14 Upcoming functions & business potential COD1.15 Final report PU

SP2D2.1.2 Contribution to the revised FESTA Handbook PUD2.2.1 Testing and Evaluation strategy I PUD2.2.2 Testing and Evaluation strategy II PUD2.3.1 Data specification and quality PUD2.4.1 Recommendations for the implementation PUD2.5.1 Functions specification PUD2.6.1 Communication technologies for cooperative driving PUD2.6.2 Upcoming innovations in cooperative driving,technologies and services PU

SP3D3.2.1 Test vehicle concepts PUD3.2.2A Test tools COD3.2.2B Test tools COD3.2.3 Instrumented vehicles for the detailed tests COD3.3.1 Test communities overview (initial) PUD3.3.2 Test communities final description PUD3.4.1 Field operational test plans PUD3.5.1 Large scale FOT execution PUD3.6.1 Detailed FOT execution PUD3.7.1 Data and user management description PUD3.7.2 Data and user management CO

2013 of 160

VTT


Contract N 224067


ic /

COnf

iden

tial

SP4D4.1.1 Tools for database handling PUD4.1.2 Tools for data analysis - software package and user guide COD4.1.3 Trial analysis report PUD4.1.4 Trial analysis report PUD4.2.3 Overview of results from 4.3 to 4.7 on a half-yearly basis. PUD4.2.3b Overview of results from 4.3 to 4.7 on a half-yearly basis. PUD4.3.1 Safety Data Analysis plan PUD4.3.2 Impacts on safety-preliminary results PUD4.3.3 Impacts on safety-results and implications PUD4.3.4 Report on eCall Large Scale FOT PUD4.4.1 Mobility Data Analysis plan PUD4.4.2 Impacts on mobility - preliminary results PUD4.4.3 Impacts on mobility - results and implications PUD4.5.1 Efficiency Data Analysis Plan PUD4.5.2 First simulation results / Impacts on Efficiency - preliminary results PUD4.5.3 TELEFOT applications efficiency impact PUD4.6.1 Environment - Data Analysis Plan PUD4.6.2 Impacts on environment-preliminary results PUD4.6.3 Impacts on environment-results and implications PUD4.7.1 Take-up of functions-data analysis plan PUD4.7.2 Implications of take-up PUD4.7.3 Impacts on user uptake - results and implications PUD4.7.4 Operational business models PUD4.8.1 Report of consumer tests and standards PUD4.8.2 Report on customer survey PUD4.8.3 Report of benchmarking of nomadic and aftermaket systems PUD4.8.4 Report on Usability Benchmarking PUD4.9.2 A catalogue of fact sheets detailing projects results and outcomes PUD4.10.1 Framework for collection of initial FOT system technical performance PUD4.10.2 Collated national FOT system technical performance specifications PUD4.10.3 Final summary of achieved technical performance from National FOT´s PU

2013 of 160

VTT


Contract N 224067


ic /

COnf

iden

tial

SP5D5.2.1 Dissemination plan and brand handbook PUD5.2.1 Dissemination plan and brand handbook PUD5.2.1c Dissemination plan and brand handbook PUD5.2.1d Dissemination plan and brand handbook PUD5.2.2 Dissemination tools e.g. logo, web-site, internal communication tool, etc PU/COD5.2.3 TeleFOT video PUD5.2.3b TeleFOT video - second release PUD5.2.5 Dissemination activities final report and assessment results PUD5.3.1 Business cases and potential new functions PUD5.3.1b Business cases and potential new functions PUD5.3.2 Markets and market potential of TeleFOT functions and services PUD5.4.1a (Internal Deliverable on Stakeholder Forums) COD5.4.1b (Internal Deliverable on Stakeholder Forums) COD5.4.1c (Internal Deliverable on Stakeholder Forums) COD5.4.1d (Internal Deliverable on Stakeholder Forums) COD5.4.2 Stakeholder Forum Activity PUD5.4.2b Stakeholder Forum Activity PUD5.4.3 (Internal Deliverable on Stakeholder Forums) COD5.4.4 Final Event PU

SP1 Admin, Finance, ManagementD1.1.1 Project Presentation PUD1.1.2 Six Monthly Report 1/8 COD1.1.2 Six Monthly Report 2/8 COD1.1.2 Six Monthly Report 3/8 COD1.1.2 Six Monthly Report 4/8 COD1.1.2 Six Monthly Report 5/8 COD1.1.2f Six Monthly Report 6/8 COD1.1.2g Six Monthly Report 7/8 COD1.1.2h Six Monthly Report 8/8 COD1.1.2i Six Monthly Report 8/8 COD1.1.3 Periodic Activity Report 1/4 COD1.1.3 Periodic Activity Report 2/4 COD1.1.3c Periodic Activity Report 3/4 COD1.1.3d Periodic Activity Report 4/4 COD1.1.4 Periodic Management Report 1/4 COD1.1.4 Periodic Management Report 2/4 COD1.1.4c Periodic Management Report 3/4 COD1.1.4d Periodic Management Report 4/4 CO

2013 of 160

VTT


Contract N 224067

2013 of 160

VTT


Contract N 224067

ANNEX 2: Results by Research Question and Hypothesis

SRQ1 - Is the route affected (where travel takes place)Table A1 –Effect on route

Research Question/hypothesis

Data analysed

Devices tested Summary of analysis techniques used

Summary of Final Results

Comments

H1.1 - There is a change in the proportion of road types driven on when the device is used compared to when it is not (commuting)

Finland LFOT

Italy LFOT

Spain (Valladolid) LFOT

Sweden LFOT2

UK LFOT

Spain LFOT2

Sat-Nav Logger Paired-sample t-test No significant change All other treatments

Significantly more type 2 & 3, significantly less type 6 (TI)

Significantly more type 2,3 & 5, significantly less type 6 (SP, NA)

Significantly more type 0 (GD, SP,TI)

Functions tested in LFOT may change route choice.

Navigation allowed travel through city centre rather around on longer roads.

Traffic info shows it’s faster to take larger roads so use increase.

2013 of 160

VTT


Contract N 224067

H1.2 – People choose different routes (based on road type) when the device is used compared to when it is not (commuting)

Finland LFOT

Italy LFOT


Sweden LFOT2

UK LFOT

Spain LFOT2

Questionnaires Post questionnaire distributions

Slight increase in use or rural roads, no effect on highways: Function = NA (GrFOT, ITFOT, SEFOT2, UKFOT; SPFOT1)

No effect on road type use Function = SP (FIFOT, GRFOT, ITFOT, SEFOT2, SEFOT1, UKFOT, SPFOT1)

No effect on highway use, slight increase in rural use Function = TI (FIFOT, GRFOT, SEFOT2, SEFOT4)

V slight decrease in highway & rural use Function = GD (FIFOT, SEFOT2, SEFOT1).

Logged data is contradicted by post questionnaire (i.e. traffic info increased rural road use).

Navigation devices seen as limited use.

Navigation support allowed use or rural roads more

2013 of 160

VTT


Contract N 224067

SRQ2 - Is the amount of time on that road affected? (How long travel takes place for)Table A2 - Effect on the amount of time on the road


Data analysed Devices tested Summary of analysis techniques used


Comments

H2.1 Subjects report a change in the number of trips undertaken because they have the device

Finland

Greece LFOT1-4

Italy

Spain (Valladolid)

Sweden LFOT2

Sweden LFOT4

UK LFOT

Questionnaires Friedman Test (differences between questionnaires)

Wilcox Test with a Bonferroni adjustment (analyse found differences)

No perceived change in number of journeys for any of the functions (green driving, SI/SA, TI or Navigation)

2013 of 160

VTT


Contract N 224067




Comments

H2.2 There is a change in the distance travelled between comparable origins and destinations

Finland LFOT

Italy LFOT


Spain (Madrid) LFOT

Sweden LFOT2

UK LFOT

Greece LFOT

Logged Data Paired-sample t-test Significant change in mean distanceo (TI, GD)o (NA, SI/SA)

No Significant change in mean distanceo All other tested

treatments in all other data sets

2013 of 160

VTT


Contract N 224067




Comments

H2.3 Subjects report a change in the distance travelled between comparable origins and destinations

Finland

Greece LFOT1-4

Italy

Spain (Valladolid)

Sweden LFOT2

Sweden LFOT4

UK LFOT

Questionnaire Friedman Test (differences between questionnaires)


No perceived change in distance travelled (Green driving, SI/SA, TI)

Some perception distance travelled is decreased (Navigation)

2013 of 160

VTT


Contract N 224067




Comments

H2.4 There is a change in the duration of journeys travelled between comparable origins and destinations

Finland LFOT

Italy LFOT


Sweden LFOT2

Sweden LFOT4

UK LFOT

Greece LFOT1-4

Logged Paired-sample t-test Significant Influence on comparable journey duration (longer)o (TI, GD)o (TI, SI/SA, GD)

Significant Influence on comparable journey duration (shorter)o (NA, SI/SA)o (GD, NA, TI)o (NA, TI)

No significant influence on comparable journey durationo All other treatments

Green driving, bundle navigation and speed info/alert and traffic info may effect duration.

2013 of 160

VTT


Contract N 224067




Comments

H2.5 Subjects reports a change in the duration of journeys travelled between comparable origins and destinations

Finland

Greece LFOT1-4

Italy

Spain (Valladolid)

Sweden LFOT2

Sweden LFOT4

UK LFOT



Perceived change in journey duration – shorter (TI, Navigation)

No perceived change in journey duration (Green Driving, SI/SA)

2013 of 160

VTT


Contract N 224067




Comments

H2.6 There is a change in the length of time driven without a break

Finland

Greece LFOT1-4

Italy

Spain (Valladolid)

Sweden LFOT2

Sweden LFOT4

UK LFOT



No perceived change in time driven without a break (Green Driving, SI/SA, TI, Navigation)

2013 of 160

VTT


Contract N 224067

SRQ3 - Does the Device cause Distraction?Table A3 – Effect on Distraction


Data analysed



Comments

H3.1 The duration and/or frequency of glances to defined target areas of the driver’s visual scene changes

Spain (Valladolid) DFOT

UK DFOT1

UK DFOT2

Logged No statistical testing available

Navigation by deviceo Increases in both

frequency and duration of glances off road

Green Driving, Foot-Liteo No change in frequency

or duration of glances

Some junctions excluded from test and done in daylight during working hours, so impossible to say that test is fully representative

2013 of 160

VTT


Contract N 224067

SRQ4 - Is speed affected?Table A4 - Effect on speed


Data analysed



Comments

H 4.1. The number of speed violations/proportion of time spent in excess of the speed limit changes with the access to the Navigation Device.

Valladolid DFOT

Valladolid LFOT

UK DFOT2

Sweden DFOT2

Finland DFOT2

Italy LFOT

Logged UK DFOT2 & LFOT’s: Paired-sample t-test

Valladolid DFOT: t-test for independent users

No significant change in number of times speeding (All other treatments)

No significant change in average speed in urban roads & motorways. Significant change in inter-urban road (Green Driving)

Care needs to be taken when interpreting findings as a result of using navigation devices

Generally no change in speed limit compliance due to use of Navigation Device.

2013 of 160

VTT


Contract N 224067


Data analysed



Comments

H 4.2. There is a change in average speed.

Valladolid DFOT

UK DFOT2

Sweden LFOT2

Sweden LFOT4

Finland LFOT

Italy LFOT

Valladolid LFOT

UK DFOT3

GEDFOT1

Logged Paired Sample t-test No significant change in average speed All other treatments

Significant decrease in average speedo(GD)

Significant increase in average speedo(Speed Limit info, GD &

Nav)o(SI/SA & Navigation)

Devices analysed in this research question is contradictory

2013 of 160

VTT


Contract N 224067

SRQ5 - Is Vehicle Positioning Affected (proximity and Lane positioning)Table A5 – Effect on vehicle positioning



Summary of Final Results Comments

H5.1 The longitudinal positioning of the vehicle will change as a result of having the nomadic device

H5.2 The lateral positioning of the vehicle will change as a result of having the nomadic device.

IT FOT

SPA

LFOT

FI LFOT

BRITISH LFOT

GRE LFOT

SWE LFOT1

SWE LFOT2

SWE LFOT4

questionnaire Descriptive statistics Little or no impact on longitudinal or lateral positioning (Green Driving, SI/SA, TI, Navigation)

Users perceive Navigation system as rarely impacting vehicle lane positioning

Users perceive speed limit/ alert as no or low impact on positioning

2013 of 160

VTT


Contract N 224067

SRQ6 - Is braking affected?Table A6 – Effect on Braking


Data analysed



Comments

H6.1 The device changes braking behaviour

German DFOT1

Greek DFOT

UK DFOT2

Greek LFOT1-4

No Information available

No Information available

No change in braking behaviour

o Static Navigation, SI/SA, ADAS

o GDo NAV, SL, TI, SA

No evidence that features in the Nomadic device change braking behaviour

2013 of 160

VTT


Contract N 224067

SRQ7 - Is Non-Driving manual activity affected?Table A7 Effect on non-driving manual activity


Data analysed



Comments

H7.1 There is a change in the duration of hands off wheel time

Italian DFOT Logged + hands & eyes videos

T-tests No Significant impact in hands off wheel time

o Navigationo Green Driving

100% interactions when TI activated, but not significant when compared to other hands off road actions

o Traffic Information

Use of functions has very low impact on hands-off steering wheel time.

2013 of 160

VTT


Contract N 224067

MRQ1 - Number of JourneysTable A9 – Number of Journeys


Data analysed

Devices tested

Summary of analysis techniques


M-RQ1.1 Is the number of journeys undertaken affected in total?

M-RQ1.4 Is the number of other home-related journeys affected?

M-RQ1.5 Is the number of other journeys (than home- or work-related) affected?

All test sites (not Greek)

Travel diaries ANOVA Chi-Squared

No Significant increase of journeyso Italian LFOT - Sig increase was

found, but most likely result of an increase in better weather conditions

o Spanish LFOT1 – Sig decrease was found, but most likely due to economic downturn of region.

o Finnish LFOT – Sign temporal degrease observed (S/A, TI, GD), but no effect with just TI, plausibly the result of random effect due to sample size

Changes in number of journeys seen less in young and old groups than middle age

>20k per year drivers had decrease (UK LFOT, Spanish LFOT1)

Increase in home-related journeys also made some public transport journeys, those with no change never used public transport (Italian LFOT).

2013 of 160

VTT


Contract N 224067

MRQ2 -Length and Duration of JourneysTable A10 – Length and Duration of Journeys


Data analysed

Devices tested



M-RQ2.1 Is the length of journeys in terms of distance affected?

All test sites

Logged data

Travel diaries

questionnaires

ANOVA (distance data from a sequence of travel diaries)

Sequential t-test (travel diary data based duration analysis)

Friedman’s test (questionnaire data results)

LoggedoDistances 9.6% longer with TI & GD than TI & SA (Finnish LFOT)oDistances 2.5% shorter with NAV & SA than with no functions

(Spanish LFOT)oNo other significant outcomesoSignificant increases seen during rush hour and daytime journeys,

not night-time (Finnish LFOT)oSignificant decrease in distance only seen at night (Spanish

LFOT1, Greek LFOT1)oSignificant increase in distance travelled at night (Finnish LFOT,

Italian LFOT) Travel Diary (holiday periods filtered out)

o Travel distance increased in 4th period (Spanish LFOT1)o No significant change (UK LFOT, Italian LFOTs, Greek LFOT,

Swedish LFOT2, and Finnish LFOT)

More long journeys in treatment phase (Finnish LFOT and Spanish LFOT1) in line with sig differences found

Impact on numeric results is due to short journeys

More journeys of 5-15Km found using GD than with SA (Finland LFOT)

Few short journeys with NAV support found (Spanish LFOT1)

2013 of 160

VTT


Contract N 224067


Data analysed

Devices tested



M-RQ2.2 Is the duration of journeys affected?

All test sites

Logged data

Travel diaries

questionnaires

ANOVA (distance data from a sequence of travel diaries)

Sequential t-test (travel diary data based duration analysis)

Friedman’s test (questionnaire data results)

LoggedoNo sig impact n journey length (Finnish LFOT + TI only)o longer journeys with GD than without or with SI (Finnish LFOT,

Swedish LFOT2)oShorter journeys (Greek LFOT3)oSig increase in duration for rush hour and daytime (Finnish LFOT

with TI+GD against TI + SA baseline, Swedish LFOT2)oSig decrease in duration (Finnish LFOT with other treatments)oSig decrease in rush hour duration (UK LFOT, Italian LFOT)oSig increase in rush hour duration (Greek LFOT1)oSig decrease in night journey duration (Greek LFOT1)oSig. increase in night journey duration (Greek LFOT2)

QuestionnaireoGD and NAV: questionnaire data in line with logged data (not

Italian LFOT)oSI/SA: questionnaire data in line with logged datao40% Spanish LFOT1 assessed decrease in duration due to NAVoWhere logged data showed sig differences due to treatment,

questionnaires in line with logged data (not Sweden LFOT2) Travel Diary (holiday periods filtered out)

o GD showed decrease in duration (Finnish LFOT, contradicts logged data)

o NAV + TI helped decrease duration of journeys (Greek LFOT)o Longer commutes used SI/SA (Italian LFOT)o No sig difference in duration (Spanish LFOT1)

No sig found related to age (Finnish LFOT)

No sig found related to experience driving

Previous traffic information experience significant impact

Those who drive most with city traffic, rural roads or motorways were more likely to have an increase in distance than those who drive less.

2013 of 160

VTT


Contract N 224067

MRQ3 - Mode of TravelTable A11 – Mode of Travel


Data analysed



Comments

M-RQ3.1 Is there a change in commuting mode of travel?

All sites tested Travel Diary

Questionnaire

ANOVA` Significant increase in car use for commuting (NAV + SI/SA) – Spanish LFOT1

Significant decrease in car use for commuting (NAV + SI/SA) – UK LFOT

Questionnaires indicate no change in car use (all journeys)

UK car use change likely due to economic downturn in the region

Decrease in number of journeys in Spain related to economic downturn in the region

Effect of TELEFOT functions on drivers was low

Effect of TELEFOT functions on drivers was low

Hard to conclude that change in travel behaviours is due to TELEFOT treatments

2013 of 160

VTT


Contract N 224067

MRQ4 - Route of ChoiceTable A12 – Route of Choice

2013 of 160

VTT


Contract N 224067


Data analysed



M-RQ4.1 Is there a change in route choice in commuting?

All test sites Logged data Chi-squared Significant changes in types of roads used (Spanish LFOT1), possibly due to navigation finding shortcuts on routes through the city centre but not certain

Increased use of main roads (Finnish LFOT1, GD, SI/SA + TI), green driving making highways more attractive to use

TI of little use due to low reporting ability (Spanish LFOT2)

No change in road type distribution or use of new roads.

No impact on traffic situation or duration

Functions tested may change people’s route choice, Navigation having the largest impact, followed by traffic information. Green driving may be useful.

2013 of 160

VTT


Contract N 224067

MRQ6 - Departure TimeTable A13 - Departure Time


Data analysed



Comments

M-RQ6.1 Is there a change in the departure time of a commuting journey?

All test sites Travel Diaries

Logged Data

Two-sided paired-sample t-test

No significant differences in departure times

Travel diary timings appear rounded

Common journeys very small compared to total journeys

Influence of weather hard to determine

Telefot Functions best for unknown routes or in unexpected delays

2013 of 160

VTT


Contract N 224067

MRQ6 - Travelling in Adverse ConditionsTable A14 – Travelling in Adverse Conditions


Data analysed



Comments

M-RQ6.3: Is there a change in travelling in adverse conditions (dark, fog, slippery road, etc.)?

All test sites Questionnaires Friedman’s test No functions changed amount of journeys in the dark or adverse weather conditions.

Functions do not influence driving in the dark or in adverse weather conditions.

2013 of 160

VTT


Contract N 224067

MRQ10 - Stress and User UncertaintyTable A15 – Stress and Uncertainty

Research Question

Data analysed



M-RQ10.1 Is there a change in user stress?

M-RQ10.2 Is there a change in user uncertainty?

All test sites Questionnaire Friedman’s Test Decrease in stressoNAV 24-50%oSI/SA 13-38%oTI 18-40%oGD 10-18%

Decrease in uncertaintyoNAV 41-66%oSI/SA 23-39%oTI 14-43%oGD 4-19%

Older participants decreased stress due to use of functions than younger (Swedish LFOT2 and Italian LFOT)

No significant influence of gender on perception.

2013 of 160

VTT


Contract N 224067

MRQ11 - Feeling of Safety and ComfortTable A16 – Feeling of Safety and Comfort


Data analysed



Comments

M-RQ11.1 Is there a change in feeling of subjective safety?

M-RQ12.1 Is there a change in feeling of comfort?

All test sites User uptake questionnaire

Non-parametric statistical tests (Friedman)

Chi-squared

Participants expected no, or slight increase in safety or comfort

No overall perceived increase in safety (NAV – Swedish LFOT2, UK LFOT; TI – Greek LFOT, Swedish LFOT2, Finnish LFOT; SI/SA – UK LFOT, Finnish LFOT; GD – Finnish LFOT, Swedish LFOT2)

Slight increase of perceived safety at the end of the trial (NAV – Italian LFOT, Spanish LFOT1, Greek LFOT, UK LFOT; SI/SA – Italian LFOT, Spanish LFOT1, UK LFOT).

Very few reports of radical decrease in feelings of safety or comfort.

Women reported an increase in feeling comfort less often with NAV than men (only Spanish LFOT1)

Participants between 30-45 more likely to feel increased comfort with NAV than older participants (only Swedish LFOT2)

Visual or hearing aids, Annual distance travelled had no impact on comfort

Previous experience with functions increased reports of comfort.

2013 of 160

VTT


Contract N 224067

EFFRQ1 - Is the travel time from origin to destination affected?Table A17 – Effect on the travel time from origin to destination


Data analysed



Comments

EFF-H1.1 Travel times are likely to increase/decrease 9when device is used compared to when it is not used)

Finland LFOTItaly LFOTSpain (Valladolid) LFOTSpain (Madrid) LFOTSweden LFOT2Sweden LFOT4UK LFOTGreece LFOT1-3

Logged data Paired sample t-test Duration was 13.3% longer with green driving than without

Duration was 18% longer with green driving compared to speed info/alerts

Duration was 10% shorter with GD, TI and Nav than without

Duration was 9.7% shorter with Nav and IT than only Nav

Green driving and bundle navigation and speed information/alert may affect duration

EFFRQ2 - Are there any delays avoided?Table A18 – Effect on delays

2013 of 160

VTT


Contract N 224067

Research Question

Data analysed



EFF-H2.1 Travel times are likely to increase/decrease (when device is used compared to when device is not used)

GreeceGreece LFOT3Italy LFOTSpain LFOTSweden LFOT1Sweden LFOT4Sweden LFOT2UK LFOTFinland LFOT

Questionnaire Chi-Squared test for equal proportions and binomial test for significance

Further tests with Fisher and Chi-Squared

ANOVA evaluated influence of type of roads

Majority of users thought Nav support made no change in perceived delays

Positive change in encountered delays prevalent in Greece but not UK (geographically related)

Majority of users reported Traffic Information making no change in perceived delays

EFFRQ3 - Are the vehicles speeds in the network reduced or increased?Table A19 – Effect on vehicle speeds in the network

2013 of 160

VTT


Contract N 224067


Data analysed



Comments

EFF-H3.1 The Vehicle speeds in the network are likely to increase/decreased (when the device is used compared to when device is not used)

German DFOT Micro-simulation tool PELOPS

Traffic jams were predicted to clear 50 seconds faster with use of device

Traffic flow simulation was necessary, yet data was not complete enough to allow this in full and allow conclusive outcomes.

EFFRQ4 and 6 - Is the time headway between the vehicles increased or decreased?Table A20 – Effect on Speed

2013 of 160

VTT


Contract N 224067


Data analysed



EFF-H4.1 Headways are likely to increase/decrease (when device is used compared to when device is not used)

Greek DFOT

UK DFOT

German DFOT

Non-parametric statistics

Paired t-tests ANOVA (rare use)

Addition of nomadic device increased headway, but not significantly

Addition of nomadic device (Foot-LITE system) significantly increased time headway

In German DFOT, no significant changes found (influence is geographical)

Significant decrease in headway (DFOT2) relating to baseline and DFOT1

Time headway increased with Foot-LITE significantly (reason for increase not possible to determine)

No statically significant decrease in minimum headways

Navigation systems are more useful and better at decreasing minimal headways in unfamiliar roads, significant decrease on motorways

Speed alert sound not intuitive

2013 of 160

VTT


Contract N 224067


Data analysed



EFF-H6.1 Distance from the preceding vehicle is likely to increase/ decrease (when device is used compared to when device is not used)

Greek DFOT

UK DFOT

German DFOT

Non-parametric statistics

Paired t-tests ANOVA (rare use)

Significant decrease found in DFOT1 compared to baseline 2

Mean decrease in distance headway in Foot-LITE condition compared to control was not significant

No statistically significant differences in mean distance were found for all speed categories in UK and Greek DFOTs

2013 of 160

VTT


Contract N 224067

EFFRQ5 - Are traffic jams avoided?Table A21 – Effect on Traffic Jams

Research Question

Data analysed



EFF-H5.1 it is likely that traffic jams are avoided (when device is used compared to when device is not used)

Finish LFOT

Swedish LFOT2

Swedish LFOT4

Greek LFOT3

Spanish (Madrid) LFOT2

Questionnaire Majority of participants in Sweden LFOT2 and thought no change in traffic jams encountered

Majority of participants reported no change in getting stuck in traffic (Finnish LFOT, Spanish LFOT2)

Over half of Greek LFOT3 through slight decrease in perceived traffic jam avoidance

No statistical influence from millage, road types

Familiarity with traffic information was seen as important

2013 of 160

VTT


Contract N 224067

Table A22 – Effect on Environment – all RQs and Hypotheses.


Data analysed

Devices tested

Summary of analysis techniques used


ENRQ1: Is average speed affected?

UK DFOT Studies 2 & 3

FIN DFOT2

SWE LFOT2

FIN LFOT

GER DFOT1

Logger Paired sample t-test

No difference in average speed observed (UK DFOT 2). A positive outcome

Eco-driving increases energy efficiency with no impact on journey time or average speed

Increase in average speed (SWE LFOT 2 and FIN LFOT 2) utilising GDS

No difference in average speed for the journey observed when using ADAS

No change in average speed when using SI/SA (GER DFOT1, FINLFOT2)

Average speed increased with TI (SWE LFOT2), no effect measured with FIN LFOT 2

Average speed increased when using ND, but only in urban environments (SWE LFOT 4)

Two FOTs revealed sig average speed (SWE LFOT 2 and 4), DFOTs revealed no significant result with ND used.

Combined functions used in tests, so drawing conclusions difficult

Giving speed related feedback did not lead to a change in average speed

2013 of 160

VTT


Contract N 224067


Data analysed

Devices tested



ENRQ2: Is speed homogeneity affected?

UK DFOT2

ES DFOT1

Logged Non Parametric (Mann Whitney)

Both UK and ES data distributions have the same mean speed (mph), but different standard divisions, making them not homogeneous

2013 of 160

VTT


Contract N 224067


Data analysed

Devices tested



ENRQ3: Is speed distribution affected?

UK DFOT2

ES DFOT1

SWE LFOT2

FIN LFOT2

IT LFOT

Logged Paired Sample t-tests

No significant change in speed distributiono ES LFOT1 (SI/SA)o IT LFOT (SI/SA)o FIN LFOT (TI, SI/SA)o ES DFOT1 (SI.SA)

Change in distribution, variance and speed greatero SWE LFOT2 (NAV, TI, GD)

Change in distribution, variance and speed smallero UK DFOT2 (GD)

Participants perceived eco-driving lead to increased journey time as a result of slower driving.

If speed distribution is reduced, fuel can be saved because a smoother velocity profile uses less acceleration

2013 of 160

VTT


Contract N 224067


Data analysed

Devices tested



ENRQ4: Is number of journeys affected?

FIN LFOT

GR LFOT

IT LFOT

ES LFOT1

SWE LFOT2

UK LFOT

Travel diaries No information available

No significant influence of seasonal weather

No significant change in number of journeys due to:o GDo SI/SAo GD + SI/SAo NAV + SI/SAo NAV + GD + TI

No significance from background variables.

2013 of 160

VTT


Contract N 224067


Data analysed

Devices tested



ENRQ5: is the distance travelled affected?

FIN LFOT

IT LFOT

ES LFOT1

ES LFOT2

SWE LFOT2

SWE LFTO4

UK LFOT

GR LFOT

Logged No information available

Significant difference for TI and GD 9.6% longer than TI and SI/SA (Finland LFOT)

Significant difference for NAV and SI/SA 2.5% shorter than no functions (ES-LFOT1)

2013 of 160

VTT


Contract N 224067


Data analysed

Devices tested



ENRQ6: Is road type and choice of routes affected?

SWE LFOT2

FIN LFOT2

FIN LFOT

IT LFOT

UK LFOT

ES LFOT1

ES LFOT2

SWE LFOT4

SWE LFOT1

GR LFOT

Logged

Questionnaire

Focus Group

Paired-sample t-test

Significant changes in road types used, although hard to determine which functions caused the change, most probably NAV (ES-LFOT1)

2013 of 160

VTT


Contract N 224067


Data analysed

Devices tested



ENRQ7: Is transport mode affected?

UK LFOT

ES LFOT1

FIN LFOT

SWE LFOT2

IT LFOT

GR LFOT1-4

Travel Diary

Questionnaire

ANOVA SI/SA + NAV sig decrease in car use (UK FOT), sig increase in car use (ES FOT1). No explanation why

Post uptake survey does not indicate change of travel mode.

Effect of introducing functions is low, with economic downturn in the regions likely to be a large factor in observed behaviour change.

2013 of 160

VTT


Contract N 224067


Data analysed

Devices tested



ENRQ8: Is total fuel consumption affected?

FIN DFOT

UK DFOT2

IT DFOT

GER DFOT

Logger T-test (FIN, UK, IT)

ANOVA (GER)

GD reduces fuel consumption significantly (FIN DFOT, UK DFOT2, IT DFOT)

ENRQ9: Is mount of CO2 emissions affected?

N/A N/A N/A N/A CO2 emissions could not be determined.

Green Driving function contributes to a reduction in CO2

2013 of 160

VTT


Contract N 224067

User Uptake Research questions and HypothesesTable A23 – Knowledge, Awareness and Behaviour


Data analysed



UURQ1: Will problem awareness/ problem perception change?

Greek LFOTs

Finnish LFOT

Italian LFOT

Spanish LFOT2

Swedish LFOT2

UK LFOT

Questionnaire No information available

o No significant change over time (Greek LFOT)

o Agreed less “private cars contribute to

global warming” (Finnish LFOT)

“would reduce car use if traffic increased” (Finnish LFOT)

“traffic is a serious problem for environment” (Italian L-FOT, UK FOT)

Technical development will solve driving risks” (Italian L-FOT, Swedish LFOT2)

Technology plays a role in reducing traffic (Swedish LFOT-2)

“would reduce car use if

2013 of 160

VTT

Katia Pagle, 15/03/13,

Different format for the UU RQs


Contract N 224067


Data analysed



cost increased” (UK FOT)o Agreed more

“new technology plays a part in environment future” (Finnish LFOT)

“would reduce car use if traffic increased” (Italian LFOT, Spanish L-FOT2)

“would reduce car use if cost increased” (Italian LFOT, Spanish LFOT2, Swedish LFOT2)

“would reduce car use if public transport was cheaper” (Swedish LFOT2)

2013 of 160

VTT


Contract N 224067


Data analysed



UURQ2: Is there a change in the participant’s compliance with speed regulations?

Finnish LFOT

Swedish LFOT2

Italian LFOT

Greek LFOT

UK LFOT

Questionnaire Friedman Test Kruskall Wallis (Finnis

LFOT)

o Green Driving No significant change (Finnish

LFOT) Sig Decrease in perceived

effect on speed compliance (Swedish LFOT2)

o Navigation No significant change (Finnish

LFOT, Spanish LFOT) Sig increase in NAV would

help comply with speed (Italian LFOT)

o Speed Alert Information Sig increase in compliance

with speed (Greek LFOT, Spanish LFOT, Italian LFOT)

No change (UK LFOT)o Traffic Information

Slight decrease in number of participants who believed TI has an effect (not Swedish LFOT2)

2013 of 160

VTT


Contract N 224067

Attitudes and AssessmentsTable A24 – Attitudes and Assessments


Data analysed



UURQ3: What are the participants’ attitudes towards the devices and functions

UURQ4: Do the attitudes change over time?

Finnish LFOT

Swedish LFOT2

Greek LFOT

Spanish LFOT

Italian LFOT

UK LFOT

Questionnaire Chi Square Fisher’s exact Kolmogorov-Smirnov

2-sample test

o Green Driving Began with neutral

attitude (Finnish LFOT, Swedish LFOT2)

Sig negative change in assessment (Finnish LFOT, Swedish LFOT2)

o Navigation Began with positive

assessment (all FOTs) Sign negative change

in assessment (all, but not Swedish LFOT2, which was increasingly positive)

2013 of 160

VTT

Katia Pagle, 15/03/13,

Different format for the UU RQs


Contract N 224067


Data analysed



o Speed Alert Information Began with fairly

positive Attitudes remained

positive (Greek LFOT, Spanish LFOT)

Attitudes became more negative (Finnish LFOT, Italian LFOT, UK LFOT)

o Traffic Information Began with positive

assessment (Greek LFOT)

Began with cautious assessment (Finnish LFOT)

Sig negative change in assessment over time (All LFOTS)

2013 of 160

VTT


Contract N 224067


Data analysed



UURQ5: How do the participants rate the benefits of having access to the devices and functions?

UURQ6: What type of benefit do the participants notice?

Italian LFOT

Finnish LFOT

Swedish LFOT2

UK LFOT

Greek LFOT2

Greek LFOT4


2-sample test

o Green Driving Began with small

benefit assessment (Finnish LFOT) or moderate assessment (all other FOTS)

Sig decrease in assessment over trial

No change over time (Finnish LFOT, Swedish LFOT2)

More participants became more positive over time than negative (Finnish LFOT, Swedish LFOT2)

Perceived benefits centred on economic (Finnish LFOT) and Environmental

2013 of 160

VTT


Contract N 224067


Data analysed



(Swedish LFOT2) issues

o Navigation Most participants

expected moderate – v. large benefit (all LFOTs)

Sig decrease in assessment over time (all LFOTS except Sweden)

High use participants reported higher benefit than low use participants (Spanish, Swedish, Greek)

Assessment focused on producing confidence and comfort while travelling

2013 of 160

VTT


Contract N 224067


Data analysed



o Speed Alert/Information Began with moderate

assessment Sig decrease in

benefit assessment over trial (Finnish, Greek, Italian, UK LFOTs)

Benefits focused on lower speed for safety (Greece LFOT2 & 4, Italian LFOT, Spanish LFOT, UK LOT)

o Traffic Information Anticipated large or

very large benefits (Greece, Sweden) or moderate benefits (other LFOTs)

2013 of 160

VTT


Contract N 224067


Data analysed



Sig decrease in assessment benefits over time (all LFOTs)

Benefits focused on convenience (GR LFOT2, SE LFOT2)

UURQ7: To what degree do the participants trust the information provided?

UURQ8: Does their trust in the information provided change over

Finnish LFOT

Swedish LFOT

Questionnaire Friedman Chi-Square

o Green Driving No statistically

significant change in trust (Finnish LFOT)

Sig decrease in trust during trial (Swedish LFOT)

o Navigation Sig but slight

decrease in total trust over time

o Speed Alert Information No statistical change

2013 of 160

VTT


Contract N 224067


Data analysed



time? in trust perceptions (all LFOTs)

Italian LFOT and Spanish LFOT trusted function the most

UK trusted info moderately

o Traffic Information High initial trust in

information (Greek LFOT, Finnish LFOT)

Sig decrease in trust in information (Swedish LFOT)

2013 of 160

VTT


Contract N 224067

Acceptance and User UptakeTable A25 – Acceptance and User Uptake


Data analysed



Comments

UURQ9: To What degree are the participants willing to keep the devices and functions for future usage

UURQ10: Is there a change in users acceptance over time?

Finland LFOT

Sweden LFOT

Questionnaire Wilcoxon test Chi Square Fishers exact Kolmogorov-Smirnov

2-sample tests

Majority of participants wanted to keep the devices and functions after the trial.o GD: 41-65% wanted to

keepo NAV: 48-93% want to

keepo TI: 12-88% wanted to

keepo SA/SI: 35-76% wanted

to keepUK and Finland least willing to keep functions

Participants become less willing to keep functions over time

2013 of 160

VTT


Contract N 224067


Data analysed



Comments

UURQ11: What is the participants’ willingness to pay?

UURQ12: Is there a change in perceived affordability over time?

Finnish LFOT

Swedish LFOT


2-sample test

Participants differed in willingness to pay for access to device and functionso GD: 4% Finnish and

17% Swedish willing to pay

o NAV: 7%-44% willing to pay

o TI: 3%-23% willing to pay

o SA/SI: 2-29% willing to pay

If willing to pay, consensus was 1-10 EURO

Decrease in willing to pay over time

Low use of functions related to low desire to pay

2013 of 160

VTT


Contract N 224067


Data analysed



Comments

UURQ13: Is user acceptance influences by perceived benefit?

All FOTs Questionnaire Spearman rank correlation

Users acceptance influenced by perceived benefit of having device/functions

Support is stronger for the relation between perceived benefit and willingness to than for relation between perceived benefit and willingness to pay

Ratings of benefit explained by 38% of variation regarding willingness to keep and 24% of variation regarding willingness to pay

No impact of age or gender

2013 of 160

VTT


Contract N 224067


Data analysed



Comments

UUQ14: Does the (physical) design of the device affect user’s acceptance of function/device?

All FOTs Questionnaire Multiple logistic regression analysis

Week support for user acceptance influenced by the nomadic device

Design of nomadic device accounted for less than 10% variation in willingness to keep and pay for item

Familiarity with NAV and SA/SI influenced predictive power of physical design uptake

Physical design of device had greater influence on younger users on willingness to pay for it

Males less influenced by physical design of device than females

Less experienced drivers more influenced by physical design than experienced

No identified effect of effect of mobile device familiarity

2013 of 160

VTT


Contract N 224067


Data analysed



Comments

UURQ15: Does the design of the user interface affect user’s acceptance of function/device?

All FOTs Questionnaire Multinomial logistic regression analysis

No strong correlation between design of interface and willingness to pay

Greater willingness to pay in younger than older participants, greater influence of design.

Driving experience slight predictor for willingness to pay

2013 of 160

VTT


Contract N 224067


Data analysed



Comments

UURQ16: Is user acceptance influences by perceived trust in function/device?

All FOTs Questionnaire Linear regression analysis

Strong correlation between trust in information and willingness to keep the device

No effect of age or gender on relationship

2013 of 160

VTT


Contract N 224067

Results from eCall Field Operational TestsTable A26 – Results from eCall Analysis

Research Question

Data analysed



eCRQ13: Is driver’ age, experience and gender may have any influence on using eCall functions?

FRA LFOT eCall Analysis of subjective data only

Women issued 10% less alerts than men

As age increases, the tendency to issue alerts increases

Usage of eCall/Alert functions were significantly dependant on driving experience

eCRQ12: Is there a change in willingness to pay over time?


Most participants showed no willingness to pay for having access to eCall/Alert

2013 of 160

VTT


Contract N 224067

Research Question

Data analysed



eCRQ11: Is there a change in users’ acceptance over time?


Assessment of advantage of eCall/Alert significantly correlated to usage over time

Significant decrease in function use over time due to functions having no perceived usefulness

Impression of eCall/Alert significantly correlated to usage over time.

Decrease in very positive and positive impressions over time.

2013 of 160

VTT


Contract N 224067

Research Question

Data analysed



eCRQ9: Does the (physical) design of the devices affect user’s acceptance of function/device?


Assessment of physical design significant correlated to usage over time

Increasing at first (summer/ autumn months) then decreasing over time (winter)

eCRQ11: Are the functions/ devices being used more or less over time?


Usage of functions significantly correlated to their usefulness over time

Significant increase in assessment of usefulness and satisfaction over time

2013 of 160

VTT

Document title - cordis.europa.eu€¦ · Web viewGreen driving advisory systems were found to...

Documents

Transcript of Document title - cordis.europa.eu€¦ · Web viewGreen driving advisory systems were found to...