Chapter Title (ACSA will complete) 1

Digital Matatus: Using Mobile Technology to Visualize Informality

INTRODUCTIONAccess to data is essential for generating urban plans for rapidly developing cities in the global south, however it is often challenging to acquire. This is because it either does not exist or is tightly controlled by those in power.[1] However, data does exist—locked away in records of everyday interactions such as cell phones and credit card transactions. My research team, Digital Matatus, set out to test whether it is possible to capture this data by leveraging the ubiquitous nature of mobile technologies in developing countries, to collect data on an essential infra-structure, and open that data up for anyone to use. The results show that devel-oping strategies that allow the public to interact with data through the creation of civic tools and visualizations helps to improve civic life while also helping to generate new planning strategies.

Focusing on Nairobi’s semi-formal bus system, or Matatus as they are com-monly known, the project was not only successful in generating data but also in transforming it into tools that could be acquired by civic actors to develop urban change. Distributed using the open data standard GTFS, the local technology community used the data to develop mobile phone routing applications. Non-governmental Organizations (NGOs) used it to develop Bus Rapid Transit Plans (BRT) plans. The downloadable paper transit map allowed Nairobi’s residents and the government to visualize, for the first time, the comprehensive system that serves their city, which generated a debate about its future amongst the vari-ous stakeholders. The implications of the research go beyond Kenya as it shows how we can leverage the power of “Big Data” to generate urban change through active data collection, data sharing and the development of visualizations and civic tools with that data.

SARAH WILLIAMS

MIT

JACQUELINE KLOPP

Columbia University

DANIEL ORWA & PETER WAIGANJO

University of Nairobi

ADAM WHITE

GROUPSHOT

“ B i g d a t a ” w i l l n o t t r a n s f o r m t h e w o r l d u n l e s s i t i s c o l -

l a t e d a n d s y n t h e s i z e d i n t o t o o l s t h a t c a n a c t i v a t e c h a n g e .

Digital Matatus

2 The Expanding Periphery and the Migrating Center

THE NEEDS FOR DATA IN DEVELOPING COUNTRIESData is essential for developing infrastructure models, and one of the biggest challenges for transportation systems is acquiring data that accurately represents these systems. [2] Data for para-transit or semi-formal bus systems rarely exist. [3] This is particularly true in developing countries and makes it difficult for resi-dents to know how to navigate their city, and, more importantly, makes it difficult for cities to know how to route and manage traffic flows.

Semi-formal transit, including Nairobi’s Matatus, often constitute the backbone of mass transit for the majority of citizens in the rapidly growing cities of Africa, Asia, and Latin America. [4] This is true in Nairobi, where the Matatus system is the main form of transit [5] serving over 3.1 million people in Nairobi metro-politan area on 130 routes. The Matatu system is run by hundreds of privately-owned companies licensed by Nairobi’s City Government. Matatus range from small vans that can hold up to 9-14 passengers to 32-seat vehicles. [6] Matatus often clog the already overstrained roadways and knowledge of the routes and schedules are essential for developing models of traffic flows as they represent a significant portion of the vehicles on the roadway.

Collecting data on semi-formal bus systems is challenging because hundreds of different operators individually set route and schedules. Governments are often unwilling to collect data on these systems as they find them too “chaotic” or do not want to sanction their existence by providing data. [7] Operators of semi-for-mal systems often collect data about the individual routes they provide, but the data is rarely shared with others since the information is primarily used to opti-mize their own business operations.[8] Often in analog form, these data sets are typically incomplete and hard to use for planning proposes.

At the onset of the project, data on Nairobi’s Matatus system was incomplete and difficult to use for planning purposes. The City Council records only identified 90 of the 130 routes. These government files, only noted the start and end loca-tion of the Matatus. Which made it difficult to know how the Matatu navigates the city and where a passenger might be able to find a stop. The research team found a stylized paper based map created in 2010 by Kenya Buzz, a Nairobi-based media company. [9] While informative, the map was no longer publicly available, and the data used to create it was never released. While performing our work we found several incomplete sources of the data collected by private actors; how-ever, none were usable for developing a comprehensive data set. [10]

USING MOBILE TECHNOLOGY TO COLLECT TRANSPORT DATAResearch studies in which cell phone users actively collect data on their mobile devices have greater potential for use in planning urban infrastructure as the collector owns the data and can decide how it will be distributed. Investigations into the geo-locative capabilities of cell phone records to develop data on tran-sit flows have been successful, but acquiring this data from telecommunication companies is extremely difficult because of privacy concerns. [11] Other research projects have used the geo-registration provided by mobile devices to capture transit data on formal systems, but the majority of these projects look at capacity and transit flows. [12] Often used to evaluate existing routes or identify unserved communities they rarely identify the location of transit infrastructure

Few projects use cell phones to capture the route location of the transit infra-structure by asking study participants to crowd source the location of vehicles in real-time on formal transit systems. [13] Some formal transit agencies have

Figure 1: Matatu bus in Nairobi. Matatu operators

develop elaborate designs on their vehicles to

attract customers. Image Credit: Sarah Williams

1

Chapter Title (ACSA will complete) 3

modeled routes by attaching GPS devices to transit vehicles.[14] Several Matatu operators in Nairobi have tested this method. However, the data is limited to the routes they operate, is not consistent, and is rarely shared. [15] Active collec-tion projects have been more successful for planning purposes but have not been applied to semi-formal systems in a comprehensive way, our research project set out to test the ability to collect data for an entirely unmapped system.

THE UBIQUITOUS USE OF CELL PHONES IN NAIROBICell phones are widely used in Nairobi. They provide essential services such as buying products, transferring funds, and providing health education. The devel-opment of these kinds of cell phone applications has exploded in Africa, and Kenya has become a center for some of these developments. [16] One of the biggest exports is the M-Pesa application, which allows users to send cell phone minutes for payment of everything from a cup of coffee to tickets to a movie. M-Pesa was started in 2007 by Safaricom, the local telecommunications network. After two years of operation, M-Pesa had adopted 8.5 million Kenyan users and transferred US $3.7 billion in funds (equivalent to 10 percent of Kenyan’s GDP) through the system. [17] In sub-Saharan Africa, there are ten times as mobile phones as mobile phones and between 2002 and 2007 cell phone subscriptions have increased by 49 percent annually while in Europe the increase has only been 17 percent. [18] Our research project sought to test whether we could leverage the widespread use of cell phones in cities like Nairobi to collect essential data for planning.

COLLECTING THE DATA ON THE MATATU SYSTEM While data collection is still ongoing, the initial process was performed dur-ing 2012-2013 in coordination with our research team, which was comprised of a partnership between the University of Nairobi School of Computing and Informatics, Columbia Universities Earth Institute, and Groupshot. The pro-cess involved identifying the current routes; developing technology to collect

Figure 2: Image of Universtiy of Nairobi student

collecting data on a Matatu.

Image Credit : Adam White

Digital Matatus

2

4 The Expanding Periphery and the Migrating Center

geo-registered data on routes and stops; translating the data in the General Transit Feed Specification (GTFS); developing unique numeric identifiers neces-sary for the specification; validating the data with the local transit community; and ultimately openly releasing the data by posting it on the GTFS Exchange.

Working with our team at the University of Nairobi, we tested and modified various mobile applications for data collection. Tool development was based in Nairobi to ensure that the technology would work on local mobile devices and cell networks and would remain a local resource. The team modified an open source mobile application to perform the data collection, developing a unique interface. [19] The application allowed University of Nairobi students riding the Matatus to collect latitude and longitude data trails of the Matatu routes. When the Matatus made a stop, they used the mobile tool to mark the stop and gener-ate a unique identifier for use in the GTFS data protocol. Students collected the data by riding the Matatus, but in some of the more dangerous areas in Nairobi they followed the Matatus in a private car.

Once collected, the route data was translated into the GTFS data specification. This specification was chosen to help facilitate the development of mobile appli-cations since many open-source transit applications already use this data speci-fication as their base. The GTFS data specification was developed in 2005 by Portland’s Tri-Met transit agency in partnership with Google. [20] This format is now widely used by transit agencies across the world to openly share their data and is the basis for transit routing directions provided by Google Maps. The open-ness with which GTFS data is distributed has encouraged its use for transit appli-cations beyond trip planning by creating tools to improve transit operations and planning overall. [21] Once transformed into this common standard, the data was

MOMBASA ROAD

THIKA R

OAD

LANGATA R

OAD

BOMA ROAD

KIA

MB

U R

OA

D

WAIYAKI WAY

OUTER RING ROAD

KANGUNDO ROAD

MOMBASA ROAD

NGONG ROAD

JUJA ROAD

JOGOO ROAD

ARWINGS KODHEK RD

EAST

ERN

BYP

ASS

NORTHERN BYPASS

LIMURU ROAD

KIAMBU ROAD

RUIRU KAMITI ROAD

LOWER KABETE ROAD

DAGORETTI ROAD NGONG ROAD

SOUTHERN BYPASS

MAGADI R

OAD

KIAMBU ROAD

20

118

11910

3

107

106

116

116

11F

11A

116

108

100

120

121

121

120

44

4549

44

25

43

115

25A

25A

49

145

45

42

42

31

103

38/39

14

45

4344

4618

C

14

3218C

106107

116

32146

18C

17A

3132

31 18C

3132

36

23

10

5836

6

2336 58

69

19C

19C218

C

19C

34B

34B

17A

33

33

33PM33F

34

33F

3333F

33F

70/71 3433F

34

11

33

12C

33

12D

12C

12D

33F

33PM

405

14B

16

1512

6

125

14A16

3424C

2433

328

7C32

15 34

125

126

126

2411

1

24C

24C

111

126

4W

2102

2

1

102

105

115

105

30 115105 1053023 48

46

48 48B

48A

4848

B

48B

48A

115

119

118

48C

20

239

129

119

118

11B

11

120

103

17A

33F 33PJ 33PM70/71

33PJ33PM

33PM

23

26

26

3433F

33PJ 33PM

33

3914

539

145

33

107D

107D107D

25A25

2222

4514

5

145

145

49

36

10

36

26

145

17B

24

2020

22

17B

17B 17B

17B

44

1960

110KIT110ATH

110KIT110ATH

KARURA FOREST

NGONG ROAD FOREST

NAIROBI NATIONAL PARK

CITY PARK

GIGIRI FOREST

BOMAS OFKENYA

JKIA

WILSONAIRPORT

UHURUPARK

MUTHAIGAHURUMA

KIAMBU

BLUE SKY/T-MALL

NAKUMATTJUNCTION

KITENGELA

ATHIRIVER

KANGEMI

BEE CENTRE

KOMAROCKSROUNDABOUT

SAIKA

DONHOLM

KAYOLE

JACARANDA

TOTAL PETROL

BURUBURUMARINGO

KARIOBANGI SOUTH

GENERALMOTORS

CALTEX

KARIOBANGI N

FEDHA ESTATE GATE A / EMBAKASI

PIPELINE

AIRPORT

KISERIAN

GITHUNGURI

BOMAS INTERCHANGE/NAKUMATT GALLERIA

CITY STADIUM

CABANAS

IMARA DAIMA

HAZINA

SOUTH B/MCHUMBI RDNYAYO

NAIROBI HOSPITAL

NGUMO STAGE

NGUMO/KENYATTA MKT

STRATHMORE

NGONG

KAREN

GASOLINA/GICHAGI

KARURA

BANANA TERMINUS

HIGH RIDGE

REDHILL

KWAHERI

NDUMBERI

RUIRU TOWN

GITHURAI

KENYATTA UNIVERSITY

CAR WASH/ZIMMERMAN ESTATE

KAHAWA WEST

SUNTON

MWIKI

NGUMBA

KARIOBANGIROUNDABOUT

BABA NDOGO

LUCKY SUMMER

ALLSOPSDANDORA

WESTLANDS/SANKARA

WANGIGE

KWA NGWACII

KIKUYU

DAGORETTIKABERIA

KREP EQUITY

KAWANGWARE

UTHIRU

ABC PLACE

STRATHMORE SCHOOL

CHIROMO

VALLEY ARCADE

LAVINGTON

GITANGA

AYANI / KIBERIA

ST MARY’S HOSPITAL

HIGHWAY ESTATE

SOUTH C

MATHARE NORTH

CALTEXJOSTER

RIKANA

EASTLEIGH

NAKUMATTPRESTIGE

THIKA TOWN

YAYACENTRE

ONGATARONGAI

LIMURU

UTAWALAJUNCTION

RUAI

RUAKA

BYPASS

UMOJA

NJIRU

MASIMBA

KWA CHIEF

CIVIL SERVANTS

GIRAFFE CENTER

KARENBLIXENMUSEUM

UoN

KNH

UN

REDHILL ROAD

MATER HOSPITAL

UTAWALA

RUAIBYPASS

NDUMBUINI

ROYSAMBU/KASARANI

DONHOLM

MAMA LUCYCITY

CENTER

Route

Stop Terminus

Outbound From City Center

Inbound To City Center

Junction

46

Kawangware

46

Yaya

19C

Komarocks

34B

Jacaranda

1960

Kayole

1961C

Kayole

1961CK

KayoleCaltex

12C

South C

12D

KPA

33

Utawala**

110ATH

Athiriver

110KIT

Kitengela

7C

KNH

8

Kiberia

24

Karen

24C

Hardy

32

Ayani

33

Ngumo

111

Ngong

4W

Kaberia2

Dagoretti

102

Kikuyu

25A

Lucky Summer

43

Ngumba

44

Kahawa West

145

Ruiru Town

45

Githurai

49

Sunton

29/30

Mathare N

25

Baba Ndogo

119

Wangige

103

WangigeNakumatt

11B

Banana

118

Wangige

**33 lines also to Fedha, Imara Daima, Highway, South

37383738

33PJ

Pipeline Jogoo

34

JKIA

70/71

Lunga Lunga

33PM

Pipeline Mombasa

33F

Fedha Estate

3738

11F

Ruaka

106

Banana

107

Ndenderu

116

Limuru

11A

Highridge

120

Githunuri

121

Ndumberi

100

Kiambu

14

Kariobangi N

17A

KayoleRoundabout

17AKY

KayoleEastleigh

18C

EastleighKomarocks Roundabout

31

DandoraGikomba

32

Dandora

46

Huruma***

3738

RuaiRoundabout

6

Eastleigh

F lin

eH

line

J lin

eG

line

I lin

e

14B

Deep West

15

Langata

16

Highrise

34

KenyattaNairobi Hospital

125

Rongai

126

Kiserian

14A

Strathmore

48A

Lavington

48B

Othaya

48

Othaya

33

Ngumo South B

Z lin

e

1

DagorettiKaren

MwikiMama Lucy

20

6

JosterRikana

42

DandoraAllsops

405

NyayoCity Stadium

48

KawangareWestlands

69

GMCity Stadium48C

YayaWestlands

24

Bomas Dagoretti

129

KikuyuWangige

108

GasolinaRedhill

11

Hazina

239

ThikaRuiru

110AK

KitengelaAthiriver

36 33 48B48B

L lin

eK

line

O li

neP

line

M li

neN

line

115

Limuru105

Kikuyu30

Uthiru23

Kangemi22

UthiruChiromo

10

Maringo

23

Kariobangi S

36

DandoraCity Stadium

58

Buruburu

26

Kariobangi N

B li

neA

line

C li

neD

line

E lin

e

Umoja

35/60

38/39

Ruai

38/39KomarocksUmoja/Donholm

19C2

17B

MwikiMuthurwa

107D

MaruruiRuaka

*44 lines also to Mwiki, Sunton,Zimmerman,and Githurai***36 line also to Kariobangi Roundabout

17B

Mwiki

Terminus City Center

Data developed through a research collaboration between the Civic Data Design Lab, MIT; Center for Sustainable Urban Development, Columbia University; School of Computing and Informatics, University of Nairobi; Groupshot. Research funded by the Rockefeller Foundation.

NAIROBI MATATU ROUTES

Data Collected by:University of Nairobi School of Computing and InformaticsPaper Maps Illustrated by:Wenfei Xu and Sarah Williams (Civic Data Design Lab)

24

29/3029/30

17Aky

17A

ky

17Aky

1960

1960

1961

K

1961

CK

1960

1961

K

17Aky

35/60

38/39

19C2

38/39

110ATH110KIT11

0AK

17Aky

38/3

919

C2

FOR FEEDBACK & MORE INFORMATION:http://www.digitalmatatus.comBeta Release : Created January 23, 20143

Figure 3: Front of the official Matatu map of Nairobi

developed by the project team.

Image Credit : Sarah Williams and Wenfei Xu

Chapter Title (ACSA will complete) 5

uploaded to the GTFS exchange, a common repository for openly sharing GTFS data; anyone can use the site to download the data our team created for Nairobi.[22]

One of the biggest challenges of the project was modifying the GTFS specifica-tion to semi-formal transit since it was initially developed for more formalized systems with less chaotic scheduling and fare changes. However, storing the data in GTFS proved beneficial as it allowed the local technology community to eas-ily generate mobile applications as detailed information was provided for the protocol.

VISUALIZING THE DATA: PUTTING THE MATATU SYSTEM ON THE MAPThe GTFS data, when plotted in the map in its raw form, was too complex to read. The mess of latitude and longitude points representing overlapping routes and stops was hard to comprehend. Order was given to this complex network using the visual language of traditional subway maps seen in Paris, London, and New York. The map structured the Matatu data, and the visual cues provided by the style of organized transit systems made map readers contextualize Nairobi’s Matatus as one organized system.

Maps are powerful as they can expose a system or idea by clearly representing it, thereby giving it a form of legitimacy .[23] Nairobi’s Matatus mostly follow the few main road corridors heading out of the city’s center with routes splitting off onto smaller tributaries as they move away from the city. In the map, each cor-ridor’s routes were grouped and color coded, as seen on more formal transit sys-tem. Identifying major stops and points of interest including parks, airports, and landmarks allowed users to position themselves on the visualization.

4649

46

19C19C

58

35/6019C

7 33PM

33F19

C58

44

7464

6

33

33PM

11

33PJ

1023

26

49

25A

25

45

44

4349

48A

46

48B

48

107

34

106

106 10711F11

6

19C34B

100

120

121

466

11F

11A

116

11B

118

116 25

A

256

46

119

118B1

130

115

1516

32

100

120

121

33

4544

43

24

14

8

102

4W

2

111

126

125

24C

14B

14A

12D

12C

3214

105

23

35/60

38/39

19601961K

29/30

110K

IT

1110

ATH

29/30

TUK

Uon Main Campus

NairobiRailway

UhuruPark

Wakulima Market

Pangani PoliceStation

STIM

A

GIK

OM

BA

MURANG'A ROAD

KENYATTA

AVE

MOI

AVE

HAILE SELASSIE AVE

UHURU HIGHWAY

Parliament

Holy FamilyBasilica

KICC

Supreme CourtCity

Hall

Way

CityHall

Jogoo House

Ministry of ForeignAffairs

Office of the President

SheriaHouse

Intercontinental

32

7C

24C

4646

Harambee

Ave

Lifestyle

33

33

33

Wakulima Rd

LANDHIES ROAD

DR

GNI

R 100

120

121

4544

43

1960

1961

K

3214

DR ES

RU

OCE

CA

R

PANGANI

KARIOKORROUNDABOUT

KOJA

OTC

RAILWAYS

ODEON

BUS STATION

33F

1069

33PJ70/71

MUTHURWAFACTORY ST

NGARA

119

LIM

UR

U R

OA

D

11ATERMINAL

14

43

44

45

TOM MBOYA ST

33F

33PM

MAIN STAGE

7

7

46

KINYANJUI ST

PUMWANI RD

KENCOM

AMBASSADEUR

625

A52

03/92

29/30

191

FORREST ROAD

NairobiGymkhana

JamuhuriHigh School

Pangani GirlsSecondary School

New Ngara Hawkers Market

KahamaHotelNairobi

Murang’a RdPrimarySchool

River BankPrimary School

Ngara SecondarySchool

Ngara RdRing Rd

Park

Rd

MuslimAcademy

Nairobi TechnicalTrainingInstitute

JUJA RD

Hombe Rd

46

17B

145

145

58

36

STIM

A

17B

17B

17B

17B

MATATU STATION LIST CITY CENTRE

110ATH

38/39

14 32

25A2532

35/605836

31

23

19C 34B48

18C17Aky633F33

10869

48A

33PM33F110KIT110ATH

48B

3331

46

1960

1961CK1961K 33F 33PJ 33PM 34 70/71

29/3034

110ATH

15 34 125 126

43

2433 33D 33F 33PM

44

1961K19C 34B 1960

26 3658

23 105 11530111

231034 70/71

24C241024W211610710611F11A

34

14

6 46

36

125 126 2433 14B 15 16 34 125 126 24 24C

1062325 25A32

116

23

44

10

38/39

45

120100116107106

34

424945

11F

110KIT

118

49

11B

125 12658

19C

14B 15 16 125 126

110KIT110ATH12C 12D 33

37/38

2

33F 33PM5849 463346

4W

110ATH

196019C 1961CK

1961CK

34B 1961K

196019C 34B 1961K

45

102 8 24 24C

44

32 111

31 3236

49

422 1

11610710611F11A6 46

44

4614

4325A

35/60 58 19C 33F33PM

25

110KIT

44 45 14 3243121120100

19C 18C

46

19C2

37/38313623 37/38323118C 42

145

32 4618C17Aky17A26

49454443196034B19C38/3935/605836

145

3433PM33PJ33F11

12D494544

46

19C

46323118C17Aky

11

17A141113224C2481024W2

331961K196034B19C

64W

43 441113224C248

116105

2 4W 102

304544

115105302318C17Aky17A1960

24111124C24421426

1961CK

48A

19C2 38/39

25A115116

125 126 24C

24C1261253415

48 48A 48B 23 30 105 115

18C1022103

11A103129115105121120100116

24126

7

1253415

11B119 11A 25A25 49

126126

145

105 129 102

120 1214516 34 24 24C 3325 25A 29/30 43 44 45 49

1960 1961K 17A 17Aky

46 4244 49

118

1267 33F 33PM 58

33

32

25A

17Aky

25A

26

4929/3025A25

25121120100116

126110ATH33

125

33

12D

16

24C

15

38/39

33F14A

37/38

16

12611143

33243433

2014B 40533PM 110KIT12C

11911811B

1960 1961K

10

24C29/3019C2

11911811B11610710611F

25A4443 4945

70/7133PJ69102644

111333224C248 7C16119

11124C241024W2103

1113224C2481024W2120

494544121

33PM

29/30

33PJ33F33D

2017Aky17A34B19C1124 111

145

116 100 120 121

44 49

14537/38

116107106

4544 49

6120 121100116

33F 33PM33PJ

33PM

33F 33PJ

7 4544

145

1961K

1960

11818C

48A

12C

118

14A

45

1146

19C

44

34B

118

36

15

10

44

25

34

119

12D337C49

25A

11937/38

30

17A120

125

49

48

119

3629/30

119

32

38/39

17Aky

121

105

126

5843

24C

35/60

4438/39

115 48

4519C

23

1960

48B4846

11911811B48126

129

115125

119

10534

11811930

48C

15

46

23

46

118103

2317A 17Aky 19C2 38/39 20

38/3935/6019C21961K196034B19C

11F 106 107 116

2

3325

48

3033

25A4W48A

10539

29/3010248B

1151454324

23

103

4424C

30

45111

105

49

115

34B

38/39 39 14533

19C 34B 1960 1961K

35/60 38/39

4345

44100 120 121

494544

7

145 237

110AK

22

145

4945444329/3025A25 6

24 24C

110KIT110ATH

24C23 105 1153022

69 405

23 105 1153022

145

48

1961K

145

145145

107D

22

2219C

4242

17B 17A145

33

33

24C

24

48

145145145

17B

17B

341024W2

145145145

145145145

145145145

110ATH

26 36

121145145145

46321433 33D 33F 33PM33PJ33

14A

10811611F 107106116 120100 121

116 120100 121106

14522

30 105 115 48232211610710611F11A29/30

2623

14526

1961K

24

33

35/60

22

145

22

11F 106 107 116

24C2414B 16

Donholm / Caltex

GitangaGikombaGeneral Motors

EastleighDrive In

Deloitte

DaystarDandora

DagorettiCorner Stone AcademyChoaka

Clayworks

City StadiumChiromoCarnivoreCarbanasCar Wash / Zimmerman

Fedha Estate Gate A

BuruburuBrookhouse School

Bomas Interchange/Blue Sky / T-MallBanana TerminusBahati

Baba Dogo

Bus Station / Railways

ABCAdamsAga Khan HospitalAirportAirport JunctionAllsops

AyanyAmbassadeur / Kencom

Athiriver

GithuraiGreenfield EstateGreenspan Mall

EastleighGasolina / Gichagi

Nakumatt Galleria

Kariobangi S

Kariokor Roundabout

Komarocks Roundabout

Kariobangi RoundaboutKariobangi N

GSUHamza

HazinaHighway Estate

HomelandHurlingham

HurumaUchumi - Hyper / iHubImara DaimaJacarandaJosterKaberiaKabeteKahawa WendaniKahawa WestKangemi

KanisaniKaren

Terminal / Main Stage

Traffic Police HQTotal PetrolThindigwa

Uchumi Hyper Langata Rd

Uchumi

St MarksStimaStrathmoreStrathmore School

SuntonSukari / Kwale

Runda

Spring Valley Police StaSpring ValleySouth CSouth B/Mchumbi RdSerenaSeasonsSaritSaikaSafaricomSafari Park

Ruiru Town

Uchumi / Nairobi West

UmojaUhuru

Umoja / Donholm

UoN Chiromo CampusUoN Science CampusUtaliiUtawalaUthiru

Yaya Centre

Wilson AirportWestlandsWestgateWangigeVillage MarketValley Road

UN

RuaiRuaka

Roysambu / KasaraniRock City

Rikana

Valley Estate

Kitengela

Lavington

Lucky SummerMama Lucy

Limuru

Langata Cemetary /Kenya Wildlife Service

Langata Police Station

Langata BarracksKwaheri

Kwa NgwaciiHigh RidgeKrep EquityKomarocks Roundabout

Koja / Odeon

Kiserain

Kasarani PoliceKawangwareKayole

KCAKNHKenyatta University

KiambuKikuyu

Ongata RongaiOTC

Muthaiga (2)

NjiruNgumo/Kenyatta MktNgumo Stage

NgumbaNgong

Nyayo

Ngara

Masimba / Bee CentreMaringo

Mater HospitalMeteorological HQ

Mathare NorthMowlemMuseumMuthaiga (1)

Muthaiga Police Station

MuthurwaMutindwa

MwikiMwimutoNairobi Hospital

NaivasNakumatt JunctionNakumatt Prestige

NdumberiNew Naivas

Pangani

PipelineRedhillRidgeways

Ridgeways

Umoja

20

20

20

4

Figure 4: Back of the official Matatu map of Nairobi

developed by the project team. This image shows

the location matrix which allowed users to more

easily find the route they needed.

Image Credit : Sarah Williams and Wenfei Xu

Digital Matatus

6 The Expanding Periphery and the Migrating Center

A legend, inset map, and matrix that matched routes with landmarks was included in the stylized map. The legend helped to associate colors with route numbers and to indicate the start and end location of all routes. The complexity of the Matatu stations in the city center made an inset map of this area essential. The matrix became the most useful addition to the composition as it mimicked the way people navigate the city—using landmarks instead of numeric addresses. The Matrix was constructed by listing the name of every landmark in the city and the Matatu routes that went to them. This allowed users to find their route in much the same way they would ask for directions.

OPERATIONALIZING THE DATA: Transforming data into a meaningful tool for urban development needs civic actors that trust the data and can see potential for its use. Data development projects are useful, but it is not until the data is leveraged for change that it gen-erates value. Throughout the data collection process, the research team engaged the local transit community through a series of workshops. These outreach activi-ties sought advice from the various organizations, allowed them to participate in the data collection, provided information on how the data would be released, and ultimately created trust in the data and its creation so that it could be readily adopted.

Workshops were held in coordination with the Kenya Institute for Public Policy Analysis (KIPPRA), Kenya’s primary government think tank tasked with transport modeling. Workshop participants included members of the government, aca-demia, the head of the Matatus Owners Association, Matatu operators and driv-ers, NGOs, and the local technology community. The University of Nairobi also held a “hackathon” using the data to encourage the development of civic appli-cation by the private sector technology community. [24] This collaboration was successful in gaining support from the government, the Matatus owners and, perhaps more importantly, for engaging the technology community in developing applications that use the data.

One of the most powerful moments of the stakeholder collaboration was our team’s presentation of the stylized transit map. The transit community was able to see the data we collected with their own eyes, and the visualization instantly generated a debate about its use. The Matatus operators used the maps as a planning tool. Conversations with them helped us to identify routes that we missed in the collection process. However, more time was spent on discussing under served neighborhoods, which were clearly identified in the visualization, and the potential for developing new routes in the areas.

When the Kenyan government saw the data visualized, they began to think of themselves as co-owners and made the visualization the official Matatu map of the city. Until the maps were presented, government officials appeared disinter-ested in the data collection process and primarily played a listeners’ role in the workshops. The visualization allowed them to see the potential of the data to cre-ate positive interaction with the public. The Ministry of Transport held a press conference releasing the visualization as the official Matatu map of the city. The forum created a debate between the government, the transport community, and the public about the future of the Matatus system. Participants questioned the government about how they would respond to necessary services changes using the map as a tool. After the press event, the downloadable maps went viral on social media and large versions were printed in local newspapers allowing

Chapter Title (ACSA will complete) 7

everyone to access the data we created.

The local technology community saw the potential of the data before it was visu-alized and developed two mobile applications using the data before the map was officially released. Laban Okune developed Ma3Route, a mobile app that pro-vided routing information using the data as a base. The tool allows users to share real-time, crowd-sourced data about Matatu route changes, accidents, and traffic congestion. Ma3Route is widely used in Nairobi and won the Kenya’s Vision 2030 ICT Innovation award in April of 2014. Sonar is another routing application devel-oped by Jeremy Gordon of Flashcast.

UN-Habitat/Institute for Transportation and Development Policy (ITDP) used the data as the basis for creating the Bus Rapid Transit (BRT) Service Plans for the city. The data allowed them to identify the location of existing routes along potential BRT service lines proposed for the Thika highway corridor. It also allowed them to identify the location of overlapping routes, showing the potential for other BRT lines in the city.

The conversations generated during the engagement process allowed stakehold-ers to come together, edit the data, and generate a debate on how the data could be leveraged to create change in the city. These conversations were essential for operationalizing the data to develop important changes to an infrastructure most citizens depend upon.

GIVING MEANING TO DATALeveraging “big data” to develop informed urban strategies often needs active data collection, data sharing, and ultimately the synthesis of data into visualiza-tions and civic tools. The Digital Matatus project shows that anyone can leverage mobile technology to capture hard-to-find data, transform that data into repre-sentations that can expose informal systems, and use the data and its visualiza-tions to help develop conversations for the future of infrastructure development

5

Figure 5: Identify locations for new Matatu routes

with transit stakeholders in Nairobi.

Image Credit : Adam White

Digital Matatus

8 The Expanding Periphery and the Migrating Center

in developing countries.

Collecting data outside formal channels is often essential for developing strate-gies for the creation of urban infrastructure. While data may exist in the hands of the government, cell phone operators, or private organizations, it is often hard to acquire this information. This is particularly true for informal systems. Semi-formal transit systems are the primary form of transit in most cities in the devel-oping world, and the tools and methodologies developed through this research project can be used to capture data anywhere these systems exist.

Data visualizations are powerful vehicles for generating debate and evidence for planning strategies. The stylized transit maps developed by the project allowed the government to engage in conversations with the public. Nairobi’s Matatu operators, who are currently the de-facto planners of the systems, used the map to identify and develop new routes for the system. Most importantly, the citizens of Nairobi now have vital information for navigating their city.

Development and dissemination of data needs to be inclusive and open in order to allow multiple civic actors to trust the data for their research. By engaging the transit community and opening the data up for anyone to use, the project was able to generate new products and tools for planning, including the development of mobile applications for citizens and the development of new transportation projects in the form of BRT lines.

This unprecedented growth of data has generated excitement about using it to reshape the way we live. [25] However, “Big Data” will not change the world unless it can be acquired and transformed into visualizations and civic tools that can be operationalized by governments, designers, and other civic actors to advocate and develop urban change. The Digital Matatus project provides an example of how we can use big data to developed necessary changes to the urban environment.

ACKNOWLEDGEMENTS:Digital Matatus is a collaborative project. Jacqueline Klopp from Columbia University’s Center for Sustainable Urban Development brought the team together. Adam White, of Groupshot, helped context the team to Nairobi’s local technology community. Peter Waiganjo and Daniel Orwa Professors of the University of Nairobi’s School of Computing and Informatics and Researcher at C4DLab directed the project in Nairobi. This work could not have been per-formed without the students who worked under their direction. Special thanks to student coordinators in Nairobi including Peterson Kamiri, Samuel Kariu, Jack Mutua. Researchers at MIT’s Civic Data Design Lab also contributed significantly to the work including; Jonathan Andrew Campbell, Emily Eros, Alexis Howland, Lindiwe Rennert, Alicia Roualt, Christopher Van Alstyne, Catherine Vanderwaart. Special thanks to Wenfei Xu from the Civic Data Design lab who was instrumental in the development of the Matatu Map. Finally, I would like to thank Benjamin de la Peña, who was our project officer at the Rockefeller Foundation and supported our work finically but also intellectually. This project would not have been pos-sible without the funding provided by the Rockefeller Foundation.

ENDNOTES

1. Williams, S., Marcello, E., & Klopp, J. M. (2014). Toward Open Source Kenya: Creating and Sharing a GIS Database of Nairobi. Annals of the Association of American Geographers, 104(1), 114-130.

2 Herrera Juan, Daniel B. Work, Ryan Herring. Xuegang (Jeff) Ban, Quinn Jacobson and Alexandre Bayen. “Evaluation of traffic data obtained via GPS-enabled mobile phones: The Mobile century field experiment” Transportation Research Part C 18 (2010): 568-583.

3 Ching, Albert, et al. “A User-Flocksourced Bus Experiment in Dhaka: New Data Collection Technique with Smartphones.” Transportation Research Record: Journal of the Transportation Research Board (2013).

4 Cervero, Robert, and Aaron Golub. “Informal Transport: A Global Perspective.” Transport Policy 14.6 (2007): 445-57.

5 Salon, Deborah and Eric Aligula. “Urban Travel in Nairobi, Kenya: Analysis, Insights, and Opportunities.” Journal of Transport Geography 22 (2012): 65–76.

6 Mutongi, K. “Thugs or entrepreneurs: Perceptions of ‘Matatu’ operators in Nairobi, 1970 to the present.” Africa: Journal of the International African Institute 76.4 (2006): 549-568.

7 Cervero, Robert, and Aaron Golub. “Informal Transport: A Global Perspective.” Transport Policy 14.6 (2007): 445-57.

Chapter Title (ACSA will complete) 9

Kemei, K. (2014) ‘Move on night transport won’t end accidents, Says Former TLB Chairman

Hassan Kamwaro’ East African Standard January 3, 2014. Retrieved from: <http://www.standardmedia.co.ke/?articleID=2000101376&story_title=kamwaro-move-on-night-transport-won-t-end-accidents>

8 Eros, Emily Jean. Transportation data as disruptive innovation in Mexico City. Diss. Massachusetts Institute of Technology, 2014.

9 Last Accessed 10/9/2014 (http://www.globalpost.com/dispatch/kenya/100127/kenyas-wild-matatus-captured-map)

10 The “Living in Nairobi” blog published a highly stylized route map in 2012 after we had begun our work, but they also did not publish any of the data collected to create the map, and have not maintained the map. (Last Accessed 10/9/2014 (http://www.jambonairobi.co.ke/services/public-transport/nairobi-route-maps/nairobi-matatu-route-map-eastlands/)

Panga Safari, formally Matatus Online, developed a private matatu route database that covered parts of the city, but did not include standardized routing information or consistent stop documentation, making it difficult to upgrade this data to a stan-dardized format such as GTFS. The database has been expanded, and can be searched through a web interface, but the back-end data was never made public. Last Accessed 10/9/2014 (http://www.matatuonline.com/)

11 Ratti, Carlo, Dennis Frenchman, Riccardo Maria Pulselli, and Sarah Williams. “Mobile Landscapes: Using Location Data from Cell Phones for Urban Analysis” Environment and Planning B 33.5 (2006): 727-748.

Gonzalez, Marta C., Cesar A. Hidalgo, and Albert-Laszlo Barabasi. “Understanding individual human mobility patterns.” Nature 453.7196 (2008): 779-782.

Caceres, Noelia, et al. “Traffic flow estimation models using cellular phone data.” Intelligent Transportation Systems, IEEE Transactions on 13.3 (2012): 1430-1441.

Talbot, David. “African Bus Routes Redrawn Using Cell-Phone Data” MIT Technology Review 30 April (2013) Web. 6 October 6 2014. <http://www.technologyreview.com/news/514211/african-bus-routes-redrawn-using-cell-phone-data/>

Wakefield, Jane. “Mobile Phone Data Redraws Bus Routes in Africa.” BBC 30 Apr. 2013. Web. 30 Sept. 2014. <http://www.bbc.com/news/technology-22357748>.

12 Caceres, Noelia, et al. “Traffic flow estimation models using cellular phone data.” Intelligent Transportation Systems, IEEE Transactions on 13.3 (2012): 1430-1441.

Herrera Juan, Daniel B. Work, Ryan Herring. Xuegang (Jeff) Ban, Quinn Jacobson and Alexandre Bayen. “Evaluation of traffic data obtained via GPS-enabled mobile phones: The Mobile century field experiment” Transportation Research Part C 18 (2010): 568-583.

Wang F. Calabrese, G. Di Lorenzo, and C. Ratti. “Transportation mode inference from anonymized and aggregated mobile phone call detail records”. In Proc. 13th Int. IEEE Conf. Intelligent Transportation Syst. (ITSC 2010), Madeira Island, Portugal, Sep 19-22, (2010): 318-323.

13 Thiagarajan, Arvind, et al. “Cooperative transit tracking using smart-phones.” Proceedings of the 8th ACM Conference on Embedded Networked Sensor Systems. ACM (2010).

14 Farzin, Janine M. “Constructing an automated bus origin-desti-nation matrix using farecard and global positioning system data in Sao Paulo, Brazil.” Transportation Research Record: Journal of the Transportation Research Board 2072.1 (2008): 30-37.

15 Interviews with Matau drivers.

16 Aker, Jenny, and Isaac Mbiti. “Mobile Phones and Economic Development in Africa.” Journal of Economic Perspectives 24.3 (2010): 207-232.

17 Safaricom. Industry Update.12 March 2009. Web. 5 Oct. 2014. http://www.safaricom.co.ke/fileadmin/template/main/down-loads/investor_relations_pdf/Industry%20Update%20120309.pdf

Mbiti, Isaac, and David N. Weil. “Mobile banking: The impact of M-Pesa in Kenya.” National Bureau of Economic Research No. 17129 (2011).

18 UITP (International Association of Public Transport). Public Transport in Sub-Saharan Africa. Trans-Africa Consortium. (2010)

19 We tested the accuracy of this tool with standard GPS units. Towards the end of the study, we tested the application Transit Wand, which had greater accuracy but the data was harder to work with after it was collected. Modifications to Transit Wand, might make it a better resources for the project. We are currently working toward creating a new tool for data collecting that modifies this open source resource.

20 McHugh, Bibiana. “Pioneering Open Data Standards: The GTFS Story.” in Beyond Transparency: Open Data and the Future of Civic Innovation Edited by Brett Goldstein with Lauren Dyson: (2013).

21 Catala, Martin. 2011. Expanding the Google Transit Feed Specification to Support Operations and Planning. Retrieved from: <http://www.dot.state.fl.us/research-center/Completed_Proj/Summary_PTO/FDOT_BDK85_977-15_rpt.pdf>

Lee-Gosselin, Martin and Ron N. Buliung. “The role of ICTs in the transformation and the experience of travel” Transportation 39 (2012): 873–876.

22 We developed several tools to transform the data into GTFS. They are not detailed here as they are very technical in nature including information about the use of a unique coding struc-ture we developed for the data itself.

23 Kitchin, Rob, and Martin Dodge. “Rethinking maps.” Progress in Human Geography 31.3 (2007): 331-344.

24 These workshops should not be confused with collaborative design workshops, rather they asked the key stakeholder to be active participants in the data collection process.

25 Boyd, D., & Crawford, K. (2012). Critical questions for big data: Provocations for a cultural, technological, and scholarly phenomenon. Information, Communication & Society, 15(5), 662-679.

Digital Matatus