SEPTEMBER2015 asset management - IWA Publishing€¦ · SEPTEMBER2015 ISSUE3 VOLUME11 PAPERS 3...

waterassetmanagementI N T E R N A T I O N A L

SEPTEMBER 2015ISSUE 3 � VOLUME 11

PAPERS3 The use of the Infrastructure

Value Index to communicate andquantify the need for renovationof urban water systemsHelena Alegre, Diogo Vitorino andSérgio Coelho

8 Energy efficient water supplysystems in Albania: moving fromenergy scans and cost-benefitanalysis to the planning,implementation and managementof energy efficiency waterinfrastructure in AlbaniaFridtjof Behnsen, Pirro Cenko,Kastriot Shehu and Piro Ndreu

13 Analysis of three efficiencymodels for water supply andsewerage companies: the caseof AlbaniaEvis Gjebrea

20 Milestones and results for theBudapest Waterworks watersafety systemGéza Csornyei and Genovéva Frank

ADB to support Uzbek city of DjizzakThe Asian Development Bank (ADB) has

approved a $81 million loan to supportupgrade of the urban sewage and wastewatersystem in the city of Djizzak, Uzbekistan, in amove it says will help raise living standards,improve the environment and boost public health.‘Djizzak is a key driver of regional economic

growth,’ said Hao Zhang, Principal UrbanDevelopment Specialist with ADB’s Centraland West Asia Department. ‘This project willgive at least 85,000 residents and over 350businesses access to clean water, and reliablesewage collection and wastewater managementservices.’Djizzak is capital of the Djizzak Province and

is a former Silk Road junction connectingSamarkand with the Fergana Valley in easternUzbekistan. According to ADB, the governmentrecently approved a long-term development planthat will accelerate urbanization, expand someindustries and relocate others, and attractforeign investment by establishing a special

industrial zone.To cope with an expected dramatic increase in

the city’s population, the project will carry out amassive rehabilitation of Djizzak’s deterioratedwastewater and sewage system, whichaccording to ADB was constructed in the 1970sand has been barely maintained since. This hasresulted in serious environ-mental damage andpublic health threats, such as an alarmingrise in acute intestinal infections and viralhepatitis.Along with a new wastewater treatment plant,

the project will construct or rehabilitate over 62kilometres of trunk sewers and four pumpingstations and provide support to operate andmaintain the new facilities. By the time theproject is completed in 2020, the system willbe able collect and treat up to 30,000 cubicmetres of raw sewage per day, and the pumpingcapacity will have been increased to more than15,000 cubic metres per day from around 9000cubic metres per day in 2013. �

Recommendations for reforms to improveperformance of Nigeria’s State Water

Agencies have been set out in a report releasedby the World Bank.With the country’s largest water suppliers

struggling to improve service provision in theface of rapid urbanisation, the report, ‘Statewater agencies in Nigeria: a performanceassessment,’ by Berta Macheve et al, surveyed35 of the 37 State Water Agencies. It foundthat many utilities are able to collect and reportperformance information, but that suchreports need to be analysed at the federal level.‘It is necessary to institutionalize a federalperformance benchmarking system,’ thereport states.According to the report, the 35 State Water

Agencies were able to connect some 1 millionnew customers between 2011 and 2013, but thisfigure was dwarfed by the approximately 9million new urban dwellers during this time,meaning urban water coverage dropped byalmost 3 percent, to less than 40%. ‘At thecurrent rate, within 10 years water coveragemay drop below 30% and only 20% of urbanresidents will have a direct water connection,’

the report states.Around $6 billion needs to be invested over

the next ten years to achieve universal watersupply, or some $2.9 billion to connect allcurrent municipal residents, yet 18 of the StateWater Agencies have never had an investmentproject valued at more than $10 million. Thatthere is the potential for users to contribute tothis investment is supported by the fact that thenational cost of water from alternative waterproviders is estimated at $650-700 milliona year, which is four times more than thecombined revenue of all the 35 State WaterAgencies surveyed, notes the report.The report’s other recommendations include:

closing the current information gap as a firstpriority; clarifying the corporate status of StateWater Agencies, especially achieving trueseparation from State Governments; improvingnational and state tariff policy guidelines andregimes; reviewing tariffs for cost recovery andaffordability; institutionalising metering of waterproduction and consumption; preparing for thefinancing of future projects with input from StateGovernments; and for the possibility of a nationalwater fund to be explored. �

Report sets out reformopportunities for Nigeria’sState Water Agencies

WATER ASSET MANAGEMENT INTERNATIONAL • 11.3 - SEPTEMBER 2015 • 2

EDITORIAL

Editors

Dr John [email protected]

Professor Stewart [email protected]

Mr Scott [email protected]

Dr Shiv [email protected]

Water Asset Management Internationalis an international newsletter on assetmanagement in water and wastewaterutilities. The focus of the newsletter ison the strategic aspects of this developingfield, providing utilities with internationalperspectives on infrastructure planningand maintenance as they seek to delivercost-effective services to their customers.

Instructions for authors are available at:www.iwaponline.com/wami/default.htm

PUBLISHING

PublisherMichael Dunn

Water Asset Management Internationalis published four times a year (March,June, April, December) by IWAPublishing. Statements made do notrepresent the views of the InternationalWater Association or its Governing Board.

IWA PublishingAlliance House,12, Caxton Street,London SW1H 0QS, UKTel: +44 (0)20 7654 5500Fax: +44 (0)20 7654 5555Email: [email protected]: www.iwapublishing.com

SUBSCRIPTIONS

Water Asset Management Internationalis available as either a print or anonline subscription.

2015 price (4 issues):£299 / €451 / $596(IWA members: £228 / €342 / $434)

ContactPortland Customer ServicesCharles Darwin House,12 Roger Street,London WC1N 2JU, UKEmail: [email protected]

Or visit: www.iwaponline.com/wami/subscriptions.htm

Design & printDesigner: David ParryPrinted by Hobbs The Printers, UK

ISSN (print): 1814-5434ISSN (online): 1814-5442

© IWA Publishing 2015

waterassetmanagementI N T E R N A T I O N A L

NEWS

European Parliament votes for furtherCommission action on Right2Water

i2O recruits for pressure expansion

The European Parliament voted on 8 Septemberin favour of a resolution urging action by the

European Commission to take forward measurescalled for by the Right2Water citizens’ initiative.The European Commission is the body responsi-

ble for drafting EU legislation for possible adoptionby the European Parliament and Member States.The European Citizens’ Initiative is a means by whichEU citizens can call directly on the Commission todraft proposed legislation if a minimum of one millionsignatures can be gathered in support of an issue.Right2Water had been the first such initiative to

gather the number of signatures needed – in factattracting almost 1,900,000 signatures. TheCommission set out its response to the Right2Waterproposals in a 13 page communication issued onMarch last year. At the time, Right2Water’s vice-president Jan Willem Goudriaan comment: ‘Thereaction of the European Commission lacks anyreal ambition to respond appropriately to theexpectation of 1.9 million people. I regret that thereis no proposal for legislation recognising the humanright to water.’The European Parliament’s response levels a

number of criticisms at the Commission and calls for

a range of actions. The European Parliament saysthe Commission’s ‘alleged neutrality’ regarding waterownership is in contradiction with the privatisationsimposed on some Member States by theCommission, European Central Bank and theInternational Monetary Fund. It says that theCommission ‘should remain neutral’ and notpromote privatisation of water services in any way. Tothis end, the European Parliament document addsthat production, distribution and treatment of waterand sanitation services should remain excluded fromthe Concessions Directive.Among the many other proposals in the European

Parliament’s response, it urges the Commissionto ensure quantitative assessments of water afford-ability become mandatory in any revision of the EUWater Framework Directive.Private water operator federation AquaFed

commented: ‘The choice of “technical” operatormust be made by the public authority that remainsin control…While some intended to exclude thechoice of the private sector option from the rangeof solutions, the Parliament recognised thepositive contribution of all operators including theprivate sector. �

Suez marks name change withstrong half-year resultsFrench group Suez released its half year

results towards the end of July, reporting anincrease in group revenues to the end of June of€404 million, up by 5.9% to€7295 millioncompared to the same point last year.The Water Europe division saw revenues rise by

€56 million at constant scope and exchange rate,and the International division saw revenues rise by€91 million. A favourable exchange rate impactaccounted for much of the increase in the overallincrease in revenue.CEO Louis Chaussade commented after the

release of the results that the board had approvedthe change of the brand name to Suez, ‘a short,strong name and full of history’. ‘This agreementcompletes the launch of our single worldwide brand.Since March, the 40 Group’s trademarks are

federated under a single brand positioned in thesustainable resource management, which is alreadybearing fruits,’ he said.Chaussade added: ‘The “Water Europe” division

reported strong performance, driven by highervolumes, prices and new services. The“International” division benefitted from sustainedgrowth in almost all geographical regions and in all isbusinesses.’In early September, Suez also announced that it

had successfully completed placement of€500million in bonds to mature in 2025. According to thecompany, demand for the bond had been six timesgreater than the offering, allowing the company toobtain its lowest 10-year coupon, at 1.75%. Thecompany also noted that the placement was in linewith its policy to refinance and extend its debt. �

Uk-based smart pressure management technologycompany i2O Water has announced the

appointment of Keith Hilson as its new Head ofCustomer Solutions.Hilson will work with the company’s water utility

customers around the world to help themmeet theirleakage, burst reduction and customer servicetargets. The company announced that he willoversee development of i2O’s new products andservice delivery models as well as leading itsinternational sales and marketing teams.Hilson was previously with metering company

Sensus, where he was responsible for the company’spre-sales work. Prior to that he was Head ofLeakage at water utility South East Water, where histeammade wide use of data and pressure manage-ment technologies to reduce leakage and helpdeliver significant operational and environmentalsavings. This work included specifying use of thei2O system.The i2O smart pressure management technology

is currently used by 66 water utilities around theworld, and the company recently secured £8 millionin funding to further develop its solutions. �


Urban water systems continueto accumulate renewal

deficits, in spite of the effortsmade in recent years in termsof awareness, scientific and tech-nological developments, capacity,and renovation itself.Ample evidence has been provided

by reports published in both developedand developing regions. In the USA,for example, theASCE 2013 ReportCard ForAmerica’s Infrastructure gaveAmerica’s drinking water and waste-water public infrastructure a D (poor)in anA-to-F grading scheme.Evencountries where the quality of theburied networks is a referenceworldwide, such as Germany, facedifficulties and pressure from themedia.“Germany is collapsing”wasthe headline of a recent, admittedlyprovocative article by Bartek Starodajin German news magazine DerSpiegel that criticized“(…) the sorrystate of Germany’s infrastructure”(www.policyreview.eu/germany-and-the-usa-are-falling-short-on-infrastructure-investment).When looking for an explanation,

economic constraints provide the easyanswer.Policy makers and utilityCEOs argue that budgets are limitedand that they cannot afford the invest-ments needed to compensate for theexisting renovation gaps.However,whenever a budget is limited, choiceshave to be made.Out-of-sight, taken-for-granted, long-lasting infrastruc-tures,when not failing at pace or withconsequences that attracts mediaattention, are at the bottom of society’spriority list, and consequently of

central and local governments.Therefore, the issue is to understandand communicate the severeconsequences of the current policies,in order to create awareness andchange priorities.

Internal and external communicationOne of the key bottlenecks for achange in policy,mindset and practiceis communication, internally to theorganizations and among stakeholders.This has been recognized in recentevents on infrastructure asset manage-ment (IAM), such as in the closingsession of LESAM 2013 (IWA &AWA,Sydney) and in the IAM Forum heldduring theWorldWater Congress 2014(IWA,Lisbon); as well as in publications

such asAMQI (2014).Internal communication challenges

increase with organizational complexi-ty.The high degree of technical special-ization and the busy schedulesof most professionals may contributeto a more difficult and less frequentdialogue among colleagues. Silos arenaturally created between managementprocesses and between utility depart-ments. Spiraling miscommunication isnaturally created: the less interactionamong professionals, the less they knowabout each other’s activities andunderstand each other’s professionallanguage, and the less they interact.Asimilar problem occurs betweendecisional levels.Mid-level technicalmanagers often complain about the

The use of the Infrastructure Value Index tocommunicate and quantify the need for renovationof urban water systemsUrban water systems continue to accumulate renewal deficits. One of the key

bottlenecks for a change in policy, mindset and practice is communication.

Communication difficulties exist within utilities and between stakeholders. There is a

need to convey that if considerable investments in renovation are not made at the

present time, and consistently over the years to come, we are preventing our children

and grandchildren from full access to the services that current modern societies take for

granted. The TRUST project provides paths forward, and one of them is the focus of this

paper – the concept of IVI (Infrastructure Value Index) and a software tool to assist in

awareness creation, assessing the long-term impact of renovation policies and assist in

stakeholders’ negotiations. By Helena Alegre, Diogo Vitorino and Sérgio Coelho.

USE OF THE IVI TO COMMUNICATE RENOVATION NEEDS

Helena Alegre,Senior Researcher, LNEC, Lisbon,Portugal. Email: [email protected]

Diogo Vitorino,Addition, Portugal.Email: [email protected]

Sérgio Coelho,Baseform, Portugal.Email: [email protected]


Annual capital gap for water infrastructure in the USA(billions of 2010 US dollars)

Year Spending Need Gap2010 36.4 91.2 54.82020 41.5 125.9 84.42040 51.7 195.4 143.7Source: ASCE (2011)

At the dawn of the 21st century, much of our drinking water infrastructure is nearing the endof its useful life. There are an estimated 240,000 water main breaks per year in the UnitedStates. Assuming every pipe would need to be replaced, the cost over the coming decadescould reachmore than $1 trillion, according to the AmericanWater Works Association(AWWA). The quality of drinking water in the United States remains universally high,however. Even though pipes andmains are frequently more than 100 years old and in needof replacement, outbreaks of disease attributable to drinking water are rare.

2013 Report Card For America’s InfrastructureASCE www.infrastructurereportcard.org


lack of understanding that their topmanagers – often elected politicians ormanagerial executives without abackground in urban water services –have of the existing problems and ofsolutions to solve them.On the otherhand, top managers are often facedwith investment proposals, presented ascrucial by their staff, but without aconvincing rationale and expressed interms that non-experts cannot fathom.Accountability and justificationperceivable by users, shareholdersand other stakeholders is almostalways lacking.Poor external communication is also

a major obstacle to good governance.Expert jargons are often used inmeetings and reports,which are notperceived by people with differentbackgrounds.How to convey theneed for …… investing in renovation of existinginfrastructures at an adequate pace,when the lack of renovation does nothave a short-term impact on thequality of service to users?… rejecting investments with anobvious short-term impact on thepopulation in favor of investing insomething that not visible and has noperceivable short-term benefit?How to transmit that if considerableinvestments in renovation are not madeat present and years to come,we arepreventing our children and grand-children to have access to the services

that current modern societies take forgranted? Media initiatives such as LastWeekTonight with John Oliver:Infrastructure (HBO) (published 2March 2015) may help,but theyneed to be accompanied by a changein culture, attitude and forms ofcommunicating by water utilities.

The TRUST project, infrastructure assetmanagement and communicationMuch has still to be done with thisregard.Some steps were accomplishedwithin the European Research projectTRUST (www.trust-i.net). In parallelto technologies and analytical tools,TRUST results include an objective-oriented approach to infrastructureasset management, theTRUST/AWARE-P IAM approach,designed tosupport a continuous improvementmanagement process (Alegre et al.2015). It is an outcome-oriented IAMplanning for long-term sustainability,embedding key ISO 55000 require-ments.Above all,TRUST/AWARE-Pis a transparent,defendable planningmethodology to support the identifica-tion of problems and in the compari-son and selection of alternativesolutions.This is fundamental for goodcommunication.TheTRUST/AWARE-P software

platform offers a growing andcustomizable number of tools(baseform.org), summarized in 1.Twoof these tools were developed within

theTRUST project, and are the mostinstrumental for internal and externalcommunication:PLAN and IVI.The focus of this paper is IVI

(InfrastructureValue Index).Thefollowing sections introduce theconcept, explain the rational of thetool and exemplify uses as a communi-cation and negotiation support tool.

The concept of IVIManaging water systems,heredesignated as IAM, faces the challengeof dealing with assets of very differentnature,useful life and cost.Typically,utility managers inherit an infrastruc-ture with assets in diverse conditionsand stages of their life cycle.They areexpected to manage their value inorder to ensure a service of adequatequality, and make sure that what theypass on to their successors is capableof continuing to do so.Due to theindefinite life of an urban waterinfrastructure as a whole, a classicallife cycle approach is not directlyapplicable – only to the individualassets. Instead, long-term planning isneeded, embedding the life cycle ofthe individual assets without losingsight of sustained and sustainableservice provision.The InfrastructureValue Index

(IVI) is the ratio between the current(fair) value of an infrastructure andthe replacement cost on modernequivalent asset basis (Alegre,2008),as stated in (1):

InfrastructureValue Index (%) =

Infrastructure Current (fair)ValueInfrastructure Replacement Cost

The Infrastructure CurrentValuewould be, in a competitive marketactivity, its market value. In a mono-polistic activity, as in urban waterservices, alternative valuation

Table 1Tools offered in theTRUST/AWARE-Psoftware platform


How long is long-term?Is life cycle approach applicable to a public infrastructure as a whole?What is better for the service: concentrated investments or continuous renovation overtime?Are we renovating our assets at the right rate?Are prevailing project financing and PPP contracts adequate for indefinite liveinfrastructures?

Or, more importantly than all:

How much and when are my renovation needs?

Compare & decide. A decision-support environment where planning alternatives or competing projects are measured up, compared andprioritized through objectives-guidedmetrics

Performance Indicators. Quantitative assessment of the efficiency or effectiveness of a system is provided, through the calculation ofperformance indicators. A tool for selection and calculation of kPI based on organized libraries, including industry standards (IWA) and user-developed libraries.

Financial project. Assess the net present value (NPV) and the investment return rate (IRR) of any financial project from a long-term/ assetlifecycle perspective.

EPANet Network Modeling. An efficient, Java-implemented EPANet simulation engine and natively integratedMSX library for full-range hydraulicand water quality network simulation, with advanced 2D/3D visualization and Google Earth integration.

Failure analysis. Using system component failure records, such as work orders, predict present and future probability of failure of pipesor sewers.

Component Importance. Simulate the failure of each individual pipe in a water supply network to measure its hydraulic impact onnodal consumption.

Unmet demand. Quantify water supply service interruption risk through the expected reduced service, calculated as the volume of unmetdemand over a given period.

Performance Indices. Technical performancemetrics based on the values of certain features or state variables of water supply is provided.Simulation-based tool provides detailed technical performance assessment, related to capacity, level-of-service, network effectiveness and

efficiency, water quality and energy system behavior.Infrastructure Value Index. Analyze the ageing degree of an infrastructure as a ratio between the current and replacement values of its

components, and project short- and long-term investment needs.

PLAN

PI

FIN

NETW

FAIL

CIMP

UNMet

PX

IVI

(1)


approaches must be adopted.Onepossibility might be the use of theaccounting value.However, thisoption is not recommended formultiple reasons, as discussed inAlegreet al. (2014).The Infrastructure Replacement

Cost is the expected cost of a modernequivalent if the infrastructure werebuilt at the year IVI refers to.IVI shall refer to a specific date, as it

changes over time. It can be assessed inseveral different ways,derived fromtwo main families, as discussed inAlegre et al. (2014):• Asset-oriented:Calculation is basedon the useful life of each asset,ondepreciation curves and on replace-ment costs for each category ofassets;

• Service-oriented:Calculation isbased on the performance offunctional units of the infrastruc-ture.

Whenever an asset-oriented strategyis applied, IVI may be determinedconsidering the individual contribu-tion of each asset (2):

where: t : reference time; IVI(t) :InfrastructureValue Index at time t;N:Total number of assets;rci,t : replacement cost of asset i at timet ; fvi,t: fair value of asset i at time t.

IVI is a standardized measure of socialresponsibility.The IVI of mature,well-maintained infrastructures is in theorder of 50% (45% - 55%).Lowervalues mean that the current genera-tion is accumulating a debt to thecoming generation and putting intorisk the service provision.Highervalues,pointing towards infrastructureyouth, are only apparently good. Inreality they mean that in the mediumor long-term high levels of invest-ments for renovation will be needed ina concentrated way. Ideally, therenovation needs and the correspond-ing investment made should be evenlydistributed over time. IVI is a long-term planning index and aims atassisting stakeholders understand,communicate and negotiaterehabilitation policies and associatedinvestments.

The TRUST / AWARE-P IVI tool:objectives, root assumptionsIVI Tool objectivesBased on data that can be eitherdetailed (asset by asset) or simplifiedby cohort, the IVI tool assesses the IVI,the capital needs and the remaining

percentage of“living assets” (i.e. assetsthat did not reach the end of theiruseful lives) over time. It encouragesthe users to carry out sensitivityanalysis to rehabilitation rates andpolicies, to planning horizons and touseful lives estimates. It can accommo-date bothAM of linear assets and ofvertical assets, although the utilityimplementations so far have beenmainly for linear assets. Its applicationto equipment and assets for whichuseful lives can be defined is straight-forward.Assets such as civil works,subject to renovation interventionsthat prolong their lives indefinitely,

is a matter currently being workedout, still marginally implemented inreal situations.The development of the IVI tool

had the following objectives:1.Assess IVI evolution over time;2.Assess capital needs for renovationover time;

3.Assess the percentage of assets (interms of replacement values) thatdid not reach the end of its usefullife, as a proxy of the quality ofservice provided;

4.Be easy to use;5.Be applicable to cases where detailedinformation exists on the assets, butalso to cases where limited information exists;

6.Accommodate both linear andvertical assets;

7.Accommodate assets of differentnature (e.g.,water,wastewater, stormwater), different replacement valuesand different useful lives;

8.Be based on and easy to understandrational, even for non-experts;

9.Allow for sensitivity analysis to themain parameters used;

10.Produce easy to understandoutputs.

The first version of IVI,developedprior toTRUST,could only respond toobjectives 4 and 7. Implementation inreal cases showed the need to fullyrethink and redesign the tool.Theutilities’ response to the new versionhas been very encouraging.

Simplification options resulting from theobjectivesThere is a major difference betweencomplexity and accuracy.There is amajor difference between being simpleand being simplistic.The IVI wasdeveloped in order to be as simple touse and interpret as possible,withoutbecoming simplistic. Some optionsmade are often object of discussion byscientists and researchers.This sectionpresents the most relevant optionsmade and,when applicable, somefrequent questions raised.Replacement cost: the implementation

of the current value of an infrastructurein the IVI tool assumes that thereplacement and the current values ofthe infrastructure may be assessed as the


Figure 1Input data for the

IVI tool

Figure 2Screenshots

from the TRUSTdeployment

Σi=1 fvi,t

Σi=1 rci,t

N

N1IVI(t) = (2)



sum of the values of its components.To allow for flexibility, the softwarehas just a column with the replace-ment cost,without any calculationsincluded.Users are free to assess thereplacement cost of an asset or a groupof assets as they consider appropriate.Import / export from Excel allow toeasily customizing calculations asappropriate.This option results fromexperience.Different utilities assessingIVI manually used different approach-es.Although general guidance isprovided (e.g.use costs of modernequivalent;use average replacementand renovation market values insteadof historic book values), context maydictate different procedure and theapplication would became either toocomplex or too restrictive if thereplacement cost calculations wereincluded.Useful lives: users may specify a single

value (e.g.60 years) or a range (e.g.

50-60 years); if a range is specified,corresponding cohorts are uniformlydistributed,processed as a uniformdistribution by the software.Given therandom nature of failures, it has beenoften suggested that the IVI toolaccepts a statistical distribution instead.This suggestion is not taken on boardfor one main reason:useful life is a veryloose concept.Assets, and particularconstructed assets as pipes and civilworks,do not“die”.There useful lifestops when they no longer fit forpurpose due to its condition (e.g. leaks,structural resistance), capacity, ease ofoperation and maintenance, etc.Statistical studies on past history mayeasily create bias.The use of sophisti-cated statistical functions often providesa false appearance of higher accuracy,while adding complexity and reducestransparency to the tool.The optionis therefore to keep a very simplerationale that easily allows to carry

out sensitiveness analysis on the impactof adopting different useful lives.TheIVI tool is not designed to supportdetailed analysis of what,when andhow to rehabilitate each asset. It is along-term planning tool, to be at anaggregated value.Current value: the implementation of

the current value of an infrastructurein the IVI tool assumes simplificationsresulting from careful analysis of prosand cons.The current value of an asset(or cohort of assets) is calculated as thereplacement value multiplied by thepercentage of residual life.Actualcondition, if known and different fromthe average for the cohort,may giveplace to adjustments in the useful livesspecified.A linear depreciation isadopted.Utility users report back thatthese simplifications are reasonable.Potential features accommodating, forinstance,non-linear depreciation rates(e.g.bathtub curve functions) wouldonce more give a false impressionof accuracy and would reducesimplicity and ease of communicationbetween technical and non-technicalstakeholders.Parameters to run the simulations: users

can easily change some parametersduring the analysis and stakeholdersdiscussions at any moment.Thisincludes the period of analysis, thepossibility to simulate what happens ifreplacement at the end of useful lifesystematically occurs and the possibili-ty of setting a target in terms of IVI,annual replacement rate and annualinvestment for renewal rate.Thesefeatures resulted from suggestions fromusers of the first generation of applica-tion and are proving to be rathereffective in terms of communication.

Data inputData input (Figure 1) is a table with asmany lines as needed and columnswith an (asset or cohort) ID; a descrip-tion of the component type (thatmay be used to filter the results percomponent type; installation date;useful life (one single value or a range);and the replacement costs.Althoughall data can be loaded directly andmanually in the IVI tool, import from/ export to MS Excel allows users toprepare or modify the input filesquicker and more easily.

Software implementationThe IVI tool is implemented in theTRUST deployment,web-based,available for allTRUST partners.However, it is also available in theAWARE-P Suite Community Edition(baseform.org) that can be used in itsweb-based form inAWARE-PTrial, orcan be downloaded and used locally.Corporate versions are also availableunder a licensing scheme.

Figure 3IVI tool output for alarge networkdescribed in detail

Strategy 1:replacement at theend of assets usefullife

Strategy 2:rehabilitation rate= 0.5 % per year

Table 2Example data for asimplified use of theIVI tool



Examples with detailed andsimplified dataIf complete detailed inventories exist,an all of assets input data for IVI canautomatically be generated and used.Figure 3 shows an example with71,200 pipes of a gravity sewer system.It might represent a higher diversity ofassets, including for instance water andwastewater, linear and vertical assets,equipment and civil works.Replacement costs were estimatedby the utility based on diametersand replacement materials (modernequivalent material). The pinkcolumns represent the annual invest-ment in rehabilitation.The yellow lineis IVI, and the blue line the percentageof the system that did not reach theend of the useful life yet.In an ideal world,Figure 3a

(replacement at the end of assets usefullife) would show a flat yellow nine at0.5 and a constant capital maintenance(pink columns). Instead,we have adeclining IVI (that will presumablycause deterioration in the quality ofservice), some decades withoutinvestment needs, and afterwards somedecades of very intensive investment incapital maintenance.To smooth thecurve of investment needs, constantrehabilitation rates might be considered(e.g.:Figure 3b). In this case thealternative devised does not totallyrespond to the problems occurred andafter 2080 it will be needed to increaserehabilitation rated.If detailed inventories do not exist

or the information is not easily avail-able to use, cohorts can be easily usedas input date of the IVI tool.Theexample inTable 2 corresponds to areal case of anAfrican town,where twomain areas in town exist: the so called“cement town”,corresponding to amore classical urbanized area, and aperi-urban area,of informal housingand slums.The outputs in Figure 4 show a

higher granularity (thicker columns),but in any case rather useful to call theattention for the needs of investmentsneeded in some years, and for rapiddecrease in service (many assets beyondits useful life) if these investments arenot made.The life discussion generatedaround these two graphics and othersimulations carried out by the stake-holders demonstrated the value of thistool for creating awareness, facilitatingcommunication and assisting innegotiation and consensus-building.

ConclusionCommunicating the needs for invest-ing in renovation of urban watersystems at a level that ensures long-term service sustainability is a battlestill to be won. Simple-to-use andunderstandable methods and tools are

needed. This paper presents IVI and asoftware tool, developed under theTRUST project and already broadlyused by many utilities, that helpsresponding to this need.�

AcknowledgementsThe research leading to these results hasreceived funding from the European UnionSeventh Framework Programme (FP7/2007-2013) under grant agreement n. 265122.

ReferencesAlegre,H.;Coelho,S.T.;Vitorino,D.;Covas,D.(2015). Infrastructure asset management - theTRUST approach and professional tools.Citiesof the Future Conference.TRUST and IWA,Mulheim,Germany,April 2015.Alegre,H.;Vitorino,D.;Coelho,S.T. (2014).

InfrastructureValue Index: a powerful modellingtool for combined long-term planning of linearand vertical assets, 16th Conference onWaterDistribution SystemAnalysis,WDSA 2014,Bari, Italy,Procedia Engineering,Volume 89,2014,Pages 1428-1436Alegre,H.;Vitorino,D.;Coelho,S.T. (2014).

InfrastructureValue Index: a powerful modellingtool for combined long-term planning of linearand vertical assets, 16th Conference onWaterDistribution SystemAnalysis,WDSA 2014,Bari, Italy,Procedia Engineering,Volume 89,2014,Pages 1428-1436Alegre,H. (2008).“Water infrastructure asset

management”,Research Program,Series «Thesisand Research Programmes»,LNEC,Lisbon,ISBN 9789724921341 (in Portuguese).AMQI (2014).AMQ International.

Strategic Asset Management,Ed.PennyBurns, Issue 400,September 22,2014,

[email protected] (2011).Failure to acct.The Economic

impact of current investment trends in water andwastwater treamente infrastructure,ASCE,[email protected],S.T.,Vitorino,D.,Alegre,H. (2013).

“AWARE-P:a system-based software for urbanwater IAM planning”.2013 IWA LESAM,Sydney,AustraliaBaseform (2013).

www.baseform.org/np4/apps/awareApp.htmlasset management.

This paper was presented at the Cities of theFuture Conference, TRUST and IWA, Mulheim,Germany, April 2015.

Figure 4IVI tool output for alarge networkdescribed in asimplified way

a) Strategy 1: Noinvestment forrehabilitation

b) Strategy 2:replacement at theend of assets usefullife


ENERGYEFFICIENTWATERSUPPLYSYSTEMSINALBANIA

Figure 1Overview ofAlbanian townscovered by GIZenergy scans

Despite the fact that Albania isblessed with abundant water

resources, the sector is facingsignificant challenges in terms ofproviding safe, affordable andsustainable drinking water andsewerage services.The currentchallenges are perhaps bestexplained by looking at somekey facts that highlight sectorperformance (Water RegulatoryAuthority, 2013 SectorPerformance Report):• On average the municipal utilitiesonly supply water for about 12hrs/day and in 2013 only nine out of

58 utilities were able to provideregular services of more than 20hrsdaily (the target is for all utilities toprovide more than 20hrs dailyservice by 2017)

• Average Non-Revenue-Water(NRW) for all utilities stands at68.1% (the target is for less than 40%NRW by 2017)

• In 2013,only 18/58 utilities wereable to meet their direct operationalcosts (target: all utilities need to meettheir operational cost by 2017)

Energy efficiencyIn their efforts to cover all operational

costs many utilities struggle with theoften old,poorly maintained andamortised pumping systems, (that is,the system comprising pumps,motors,suction and transmission mains andreservoirs).As a result many utilities

Energy efficient water supply systems in Albania:moving from energy scans and cost-benefit analysisto the planning, implementation and management ofenergy efficiency water infrastructure in Albania

Many Albanian water and sewerage utilities have energy costs that make up 30-40 % of

their direct operating costs, in several cases even well above 50%. Energy scans and

cost-benefit analysis conducted in 14 out of 58 towns have revealed that existing

pumping systems often have efficiencies that are well below 50% of a comparable new

system. This is due not only to the age of the pumping systems, but also because the

systems have been designed and built for purposes different to those they are being

used for today. Considering the high proportion of energy costs in the operations along

with the low efficiency of existing systems, it is clear that there is a considerable need to

improve the use of energy in water and waste water service provision. Efficient use of

energy is an important factor to ensure the sustainable use of resources and is also a

key ingredient when it comes to the provision of affordable and cost recovery water and

sewerage services. Fridtjof Behnsen, Pirro Cenko, Kastriot Shehu and Piro Ndreu

outline economically-viable investments and practices to achieve this aim.

Fridtjof Behnsen, MSc, MBA,GIZWater SectorReformProgram,AlbaniaEmail: [email protected]

Pirro Cenko, MSc Eng.GIZWater SectorReformProgram,AlbaniaEmail: [email protected]

Kastriot Shehu, Dr. Eng.Consultant, AlbaniaEmail: [email protected]

Piro Ndreu, Dip. MSc Eng.Consultant, AlbaniaEmail: [email protected]


Skenderbeg Square, Tirana,Albania (Credit: FridtjofBehnsen)


low network pressure, very manyprivate households and businesses inAlbania have in-house water pumpingsystems and storage facilities to copewith the erratic supply situation.Thismeans that a very significant propor-tion of theAlbanian energy demand isdue to water and sewerage serviceprovision, services that are charac-terised by many inefficiencies that havesignificant potential for improvement.

Energy optimisationIn addition to the low energy efficien-cy of existing pumping systems, theutilities also face a significant challengein terms of energy optimisation in theirnetworks.Many of the systems operat-ing today are rather old,with largelyamortised assets. In addition, thesesystems were originally designed fordifferent purposes when they wereconstructed.This leads to a situationwhere some key assets like reservoirsare not located in the ideal place,or arenot matched to the volumes of waterthat need to be provided.As a result, reservoirs are quite often

sited at high levels that don’t allow forthe efficient use of energy for pump-ing.These reservoirs would better be atlower elevations that are better suitedto providing adequate supply pressurein the networks.Transmission mains aretoo small, resulting in high flowvelocities that lead to higher frictionlosses and the need for bigger pumps.These are major contributing factors

to the very high NRW figures inAlbania,which currently stands at a

sector average, according to sectorbenchmarking data,of about 68%.Alarge part of these losses is due totechnical losses incurred through leaksand bursts in the transmission anddistribution networks. In the case ofpumped supplies, this means a lot ofenergy is being used to produce waterthat is lost before it is consumed andtherefore does not generate anyrevenue for the utilities.So in addition to new pumping

systems with efficient pumps andmotors (that is, energy efficient), areduction in the very high levels ofNRW through energy optimisation isrequired to ensure better use of energyand a significant reduction in theamount of water that needs to bepumped.Many of the supply systemsvisited do not have operationalnetwork zones or any pressuremanagement,maintenance plans oroperational plans in place.This furtherincreases the losses.

Future use of energy scans andcost-benefit analysisTo systematically improve the energyefficiency of pumping systems inAlbanian water and wastewaterutilities, the following 10-stepapproach was defined from initial datacapture, energy scans and cost-benefit-analysis to finally implement measuresand monitor the performance ofinvestments.This approach is similar tothat found in guidelines such as theEPA Guidelines for EnergyManagement for example:

Figure 2Energy Scans inrelation to theAsset Life Cycle

Figure 3Performanceimprovement logic

have energy costs that account for30-40% of the direct operating costs,and for some utilities this goes beyond50% as in the case of Durres at 56%.The situation with the dilapidatedcondition of many pumping systemswas confirmed in 14 of the 58 townsthrough energy (system) scans in 2014.The results are documented in cost-benefit-analysis, and by mid-2015scans of a further six utilities will alsobe completed (see Figure 1).With many utilities currently unable

to meet their operational costs,significant electricity arrears areregularly accrued.As a result thegovernment ofAlbania has repeatedlyprovided energy subsidies over theyears to clear arrears with the electrici-ty distribution company on behalf ofthe utilities, i.e. about€20 million($22.6 million) in 2014 and a further€45 million ($50.8 million) in 2015.These subsidies,however,were merelypayments of arrears and did not carryany conditions for utilities to makenecessary investments and/or necessarychanges at the operational level tocounteract a future build-up of newelectricity arrears.The energy scans and cost-benefit

analysis completed to date come to theconclusion that most of the existingpumping systems that have been visitedonly attain efficiencies generally below50% of a comparable new system,resulting in high energy inputs for thewater produced.Knowing that the reported

operational data from utilities isgenerally inaccurate (for instance onlyabout 25% of the water production ismeasured by bulk meters, and therest is estimated), the overall use ofelectrical energy in the water sectorprobably ranges between 12-15% ofthe billed energy consumption(excluding water heating).There arehundreds of smaller communalschemes not included in this figure.Also, and as a direct result of the

intermittent water supply and resulting

In 2013 the 58municipal utilities reported to have produced 273million m3 from pumpedand gravity fed sources, serving approximately 2.5 million people in the respective serviceareas. This volume equates to a per capita production of 300 litres/day. This is in contrast tothe average per capita consumption of 150 litres/day.

In 2013 utilities reported to have used 173Million kWh to pump 160million m3 of water.This results in a unit energy input of about 1.1 kWh/ m3 and represents 5% of the overallbilled energy consumption in Albania.




sector that requires energy efficiency aswell as energy optimisation measures.To systematically tackle the

challenges it is suggested that utilitiesfocus on the asset life cycle approachdepicted in Figure 2.The initial datacapture (system and operational data)that led to the energy scans werecarried out in the operational phase ofthe cycle for the existing water supplysystems.The results obtained fromthese operational assessments aredocumented in the form of cost-benefit analysis,which needs to beconsidered in the planning phase ofthe cycle to ensure future infrastruc-ture is energy efficient.

Planning phaseFor the planning phase the NationalStrategy ofWater Supply and Sewerage(2011-2017) provides clear guidanceon the kind of service that shall berendered by water utilities inAlbaniaby 2017.The main aspects that to behighlighted in this paper are:• A 24 hour supply withadequate pressure

• A supply with good qualitypotable water

• Affordable water and sewerageservices for the end consumers

• Sustainable services (economicallyand environmentally), that is,services that attain operational costrecovery and at a later stage totalcost recovery

To check whether utilities inAlbaniaare moving towards attaining thesesector strategic targets, a system ofperformance indicators has been inplace since 2008.This system has beencollecting relevant operational datafrom utilities over the years and it isused to calculate sector specific KeyPerformance Indicators (KPIs).However, so far this data is mainly

being collected to calculate therespective KPIs that are reported in thesector.The information from the KPIsis not turned into actual knowledgeoften enough through assessments, andthat knowledge used to define clearactions as depicted in Figure 3.More effective performance assess-

ments are needed at the operational,regulatory and policy levels. If suchassessments are made effectively againstthe sector strategy targets, for example,they can be used to define manage-ment actions to address shortfalls andtrigger necessary investments.Thiscould include decision making onsubsidy allocations to utilities as well,based on performance.From the operational/system data

capturing that was undertaken for theenergy scans it has become clear that inmany cases the data at the utility levelis either not (readily) available and is

• Define main challenges (objectives)• System information (data capture -system)

• Operational information (datacapture - operations)

• Situation assessment (energy scans)• Technical options (cost benefitanalysis)

• Feasibility study• Secure funding• Detailed design/tender documents• Project implementation/commissioning

• Monitor and communicate outcomes

The work done under theWater SectorReform programme, a technicalassistance programme funded by theGerman government and implementedby GIZ (Gesellschaft für InternationaleZusammenarbeit), only dealt with steps1 to 5 of the above approach, that is,energy scans/cost-benefit-analysis.Thedesign and implementation of actual

infrastructure measures to mitigatesome of the problems was not part ofthe mandate of the programme.This,however, leaves the question of whatcan be done with the valuable findingsand analysis.How can theAlbanianwater sector use the findings andrecommendations in order to dealwith the situation in the sector?

Asset life cycleWith the water sector inAlbaniaresponsible for an estimated 12-15% ofthe billed electricity consumption, andat the same time operating manyinefficient pumping systems and watersupply networks with multiple opera-tional challenges, isolated improvementmeasures by some water utilities willnot have sufficient impact for thenecessary long-term change to providesustainable and improved servicedelivery inAlbania.There is consider-able energy saving potential in the

Figure 4Ability to influencethe costs ofinfrastructureprojects alongthe projectdevelopmentprocess

Figure 5Schematic of theWOBIS System

Life Cycle Costs for pumping systems have shown that 5% of the overall costs are due to theinvestment cost for the implementation and 10% for the maintenance costs over the lifetime. However, 85% of the costs that a utility incurs from a pumping system are related toenergy costs. Therefore, using more expensive energy efficient motors pays off over the life-cycle and ensures lower operating costs.


ENERGY EFFICIENT WATER SUPPLY SYSTEMS IN ALBANIA

also often not accurate or reliable.Butunless the data variables are correct andaccurate, everything else will notrepresent the actual situation on theground,which will lead to erroneousassessments.To achieve this aim,WOBIS (Water

Operational Business InformationSystem) was developed to assist theutilities in capturing key operationaldata variables on a monthly basis andusing them to calculate selected KPIs.Thus the utilities are put in a positionto monitor progress on a monthly basissuch that they can see if operationalimprovements are being made thatlead toward attaining the envisagedKPI targets.Based on the data collected for the

energy efficiency assessments, thecost-benefit analyses that have beenprepared generally outline economi-cally-viable measures in the 14 utilities,looking at investments with a simplepayback period generally in the rangeof two to five years.

Implementation phaseFor the implementation phase it is vitalthat the recommendations from theplanning phase are based on soundassessments using valid data. It is at theplanning stage of any project where thebiggest opportunity exists to influencethe final costs of new infrastructure asdepicted in Figure 4.If the planning is wrong from the

start, this could well lead to inefficientinfrastructure with high operatingcosts.Therefore it is important thatfor pumping systems, for example,professional pump selection softwareis used to determine the most suitablepump/motor for the pumping system(for the transmission main).This alsomeans that the impeller diameter ofthe pump has to be matched to actualoperating conditions, and if thedemand in the supply area varies a lot,variable speed pump sets should beconsidered.Finally the use of energy-efficient IE3 motors has to be consid-ered, and appraisals of differenttechnical options need to considerthe full life-cycle cost to ensurethat operational costs are properlyconsidered.Since NRW is such a big issue in

Albania, a special focus needs to beput on improving the water supplynetworks.First, the capacities ofexisting networks have to be checkedagainst realistic demand scenarios.During the energy scans it was foundthat many transmission mains forexample were operating with flowvelocities that were far too high.Undersized pipes lead to excessivehead losses and require large pumpsand motors to transport water to thereservoirs.This uses a lot of energy, and

could be remedied by providingadditional transmission capacity.Second, the reservoirs are not always

located where they are needed.Manyreservoirs were constructed to serve adifferent purpose to that which theyare used for today.This,however,meansthat a lot of water is lifted to elevationswhere it is not needed.This consumes alot of energy that could be saved byconstructing reservoirs in suitablelocations.Third,many networks don’t have

any functional pressure zoning.Not only does this affect the overalloperation of the water supply net-works,but with reservoirs in highlocations the water pressure in thenetwork is excessive in many low-lying areas.All of the above leads to increased

wear and tear on the infrastructure, andincreases the NRW.This is why there isa strong need inAlbania to design andimplement energy optimisationmeasures.This requires the use ofexperienced planners and designers,and the need to specify and tenderthe right equipment and materials,complying with minimum technicalstandards,which ensure the implemen-tation and use of efficient infrastructureover the asset life cycle.Finally,professional construction supervisionand commissioning is needed to ensurethat what was designed and specifiedis constructed and installed to therequired standard.

Operational phaseTo ensure that good quality services areprovided by a utility, it needs to employqualified staff that will ensure the day-to-day operations are done accordingto the operating manuals. In many

places that were visited,operatingmanuals were not available and keyoperational data like water productionand pump operating hours were notrecorded.Also the utilities generally didnot have an up-to-date asset registerand networks maps and schematicsavailable that would provide therequired information to carry out anassessment like an energy scan at apumping station.As a result an Excel tool has been

developed to capture key pumpingsystem data and help determine theenergy efficiency of the pumpingsystems. In addition,WOBIS (WaterOperational Business InformationSystem) has been developed to help theutilities to routinely capture keyoperational data variables on a monthlybasis. Figure 5 outlines theWOBISsystem.The operational data to becaptured is first and foremost requiredby each utility to regularly analyseand assess its performance againstoperational targets,which would beset as part of a utility’s business planor corporate plan.Additionally, thesystem can also be used to satisfyreporting needs to theWaterRegulatoryAuthority (ERRU) andthe Benchmarking Unit (BMU)within theWater and SewerageDirectorate (DPUK) in the MinistryofTransport and Infrastructure.From the recorded operational data

variables, theWOBIS system automati-cally calculates the respective indicatorsand can compare these with annualtargets for the KPIs,provided thesehave been set.These regular performance assess-

ments need to be linked to the defini-tion of specific management andinvestment actions geared towards

Figure 6BenchmarkingCycle (adoptedfrom DWA-M 1100E)

A single leak from a 6mm-hole in a water distribution pipe or transmission main, operated at5bar pressure, will lose about 40-45m3 of water per day. This amount of water could fill anOlympic size swimming pool with a capacity of 2500m3 in 2months. With an assumed waterproduction cost of 0.3 €/m3 this single leak will cost a utility almost €5000 annually



closing the gap between currentperformance and envisaged or targetedperformance, as outlined in the bench-marking cycle in Figure 6.Such assessments have also been

integrated into the revised tariffapplication process of theWaterRegulatoryAuthority (ERRU).Since2014, tariff applications using the newtariff-setting tool need be accompaniedby a tariff justification.This can essen-tially be seen as a summary version of abusiness plan that requires a utility tohighlight the strategies and plannedinterventions it wants to carry out, inan effort to meet the forecast improve-ments in the regulatory KPIs. It wouldbe ideal if in future other decisions likeallocations of subsidies would also bebased on such assessments and plans.

Retirement phaseIn the case of planned disposals ofassets, it is necessary that steps 1 to 5 ofthe asset life cycle (see Figure 7) aredone before the old assets can beretired.Only at this stage can these oldassets be taken out of service and alsoremoved from the asset registers,whilethe new assets need to be entered. Itwas noted that many investments in thepast have not been recorded properly inasset registers and their status andcondition have not been monitoredand updated.This would be a huge helpfor decision making if and when assetsneed to be replaced.From the work in the field for the

energy scans it was also noted thatbroken assets (for example pumps ormotors) are quite often replaced withwhatever is available at the time, toensure that some kind of service can beprovided.These makeshift solutions,however,often remain in place and leadto situations where, for example,oversized pumps and motors are notmatched to the system curve andoperate very inefficiently.Effortsshould be made to properly replace

these assets with well-planned anddesigned solutions as soon as possible toattain cost effective operations.

ConclusionIn the current situation for manyutilities inAlbania,well planned,designed and implemented energyefficiency and energy optimisationmeasures could have a significantimpact on their ability to cover theiroperational costs.New pumpingsystems with highly efficient pumpsand motors (that is, energy efficient),along with measures to better controlthe very high NRW (through energyoptimisation) would lead to a moreefficient use of energy and a significantreduction in the amount of water thatis pumped into the system.Suitableinvestments have been identified in thecost-benefit analysis and many havebeen found to be economically viablewith a simple payback period of two tofive years.To ensure that the right infrastruc-

ture investments are planned andimplemented, it is paramount thataccurate and reliable operational data isavailable and also used for regularperformance assessments.This requiresqualified,well-trained staff in theutilities and for the sector to performthese duties better.Being able to useenergy more efficiently and pump lesswater would also have a big impact onthe future development of water/wastewater tariffs, and thus the long-term affordability of services for theend consumer.To achieve this, significant infrastruc-

ture investments are required, as manyof the water supply systems are oldwith amortised networks and pumpingsystems.According to the Master Planfor theAlbanianWater Sector, there isan overall infrastructure investmentsneed to the amount of€5.1 billion($5.8 billion) until 2040 (or€6.8billion ($7.7 billion) including all

related costs).This equates to an annualinvestment of€180 million ($203.9million) to ensure that by 2040 allAlbanians enjoy a 24 hour watersupply service with good potablewater, and adequate wastewaterservices.Finally, it would be considered very

beneficial for theAlbanian water sectorif the planning, implementation andmanagement of water and wastewaterinfrastructure were set up followingan asset life cycle approach.Todetermine how well current processesand practices in the sector arealigned with such an approach, adetailed analysis would be required todetermine exactly where there aregaps and weaknesses.Based on such a gap analysis, con-

crete measures could be formulated toclarify the roles and responsibilities ofall actors involved in the planning,design, implementation and opera-tions.An effective communicationstructure between all actors wouldneed to be developed based on well-designed processes and appropriatesteering and coordination structures,supported by an appropriate capacitydevelopment programme at sector,institutional and individual level.Within the ongoing sector reforms

and as part of the current administra-tive territorial reforms inAlbania, thiswould be an ideal time to develop andintegrate such an asset lifecycleapproach for the water sector as part ofthese reforms.Through its expertise incapacity and organisational develop-ment and its long-term involvement intheAlbanian water sector,GIZ(Gesellschaft für InternationaleZusammenarbeit) would be an idealdevelopment partner to assist in theconceptualisation,design and imple-mentation of such a multi-layered andmulti-faceted change process.�

ReferencesWater RegulatoryAuthority ofAlbania (2014),Report on the Performance ofWater Supply andSewerage Companies.Water RegulatoryAuthority ofAlbania

(2015),WRAAnnual Report 2014.Albanian Official Journal (2011).NationalStrategy ofWater Supply and Sewerage (2011-2017).DWAGermanAssociation forWater,

Wastewater andWaste (March 2008),AdvisoryLeaflet DWA-M 1100E,Benchmarking inWater Supply andWastewater Disposal.Albanian Ministry of PublicWorks and

Transport (MPWT) - General Directorate ofWater Supply and Sewerage (DPUK) (2013),Water Supply and Sewerage Master Plan forAlbania.US Environmental ProtectionAgency (EPA),

Energy Star Program (2013),The Guidelines forEnergy Management(www.energystar.gov/guidelines).

Figure 7Asset Life Cycle

This article isbased on a paperpresented at theIWA Regional UtilityManagementConference, Tirana,Albania, 13-15 May2015.


ANALYSIS OF THREE EFFICIENCY MODELS FOR ALBANIAN WATER COMPANIES

At the global level,water isconsidered a rare resource

and there are several factors thatcause this phenomenon.Theseare global warming, populationgrowth,which increases pressurefor more water, and demographicchange in the population due tothe shift from rural to urbanareas.The debate on water hasbecome an important issueworldwide, as shown by theinclusion of water issues inAgenda21 of the United Nations(UN) in 1992 and the MillenniumDevelopment Goals.Water, from being a free or social

commodity has become a consump-tion commodity and many globalconditions that have occurred andare occurring have led to its scarcity.As a result one of the challenges ofthe 21st century is the effective andefficient management of water.To thatend,water management theory haschanged by introducing the conceptof sustainable water management, andits relationship with other sectors suchas health, environment, agricultureand industry.Albania is a country rich in water

resources, and with the currentcapacity of the resources it possesses itis able to meet the needs of the wholepopulation.But the challenge remainsefficient use of water resources.Concretely, the annual averagecapacity of all sources used forproduction of drinking water is654M.m3,while the average volumeof water produced is 278M.m3

(Drejtoria e Përgjithshme e UjësjellësKanalizimeve,2010).This indicates that only 43% of the

country’s water resources are used.Assuming that the entire populationliving in the area of jurisdiction was

supplied with drinking water,due tolosses and misuse – using water forother purposes – production is 31%higher than the demand for drinkingwater.Meanwhile, the volume of waterbilled (sold),meets only 54% of thedemand at sector level.These statistics

Table 1: List of Water/Sewerage Companies with both services

No. Name of the Company Number of population Number of population withserved with water access to sewerage

1 Berat & Kuçovë 103743 875122 Burrel 23025 164753 Delvinë 6300 20924 Durrës 267612 1812405 Elber (sh.p.k) 116000 1122506 Ersekë 5900 59007 Fier 126250 760428 Fushë Arrëz 1986 17009 Fushe Kruje 9616 836010 Gjirokastër (Q) 33156 1402511 Kamëz 40768 2010412 Kavajë 61071 2375113 Korçë (Q) 87500 7740014 Krastë 2213 201315 Krujë 16837 1600016 Kukës 40000 2500017 Lezhë 31378 3170618 Libohovë 3082 56719 Librazhd 18812 1610020 Lushnjë (Q) 47500 4000021 Mallakastër 17975 1072022 Mirditë 6074 526123 Pogradec 58142 4388524 Pukë 2892 211225 Rrogozhinë 9007 350426 Rubik 2322 220727 Sarandë 43220 3234628 Shkodër (Q) 82649 7346529 Tiranë (Q) 828200 655000

Total number of population for 29 companies 2093226 1586735Total number of population for 58 companies 2573023 1586735Weight of sample /Total population number 81% 100%

Source: General Water and Sewerage Directorate

Analysis of three efficiency models for water supplyand sewerage companies: the case of AlbaniaEvis Gjebrea analyses three efficiency models for the water supply and sewerage sector

in Albania, which aim to evaluate economies of scale by focusing on the factors that

impact the efficiency of water and sewerage companies. The study analyses financial,

technical and operational data for the period 2006-2012 for 29 water and sewerage

companies that cover 81% of the population of Albania. For the purposes of this study,

this research uses an econometric model for efficiency of water supply and sewerage

companies previously tested in a study on the relative efficiency of water utility

companies in Brazil, which has similarities to companies in Albania.

Evis Gjebrea, PhDLecturer, Department of Economics andManagementEuropean University of Tirana, AlbaniaEmail: [email protected] [email protected]



ANALYSIS OF THREE EFFICIENCYMODELS FOR ALBANIANWATER COMPANIES

sectors, and the total cost of the privatenetwork is a substantial part of the totalcost of water, leading to a naturalmonopoly (Mergos,2005).Specifically,competitors cannot build parallelwater supply and sewerage systems inthe same city (Urio,2010).Quiggin (2011) argues that most

cases of public ownership exist forinfrastructure services such as energyand water and there are two reasons forthis: first, these services require hugeinvestments and what counts is the lowcost of public borrowing, and second,such services are a natural monopolyand if they were privatized they wouldrequire regulation by the government.In this sense, the choice is betweenone form of government interventionand another.One way to address this issue is to

merge several water companies andcreate big regions,which is knownas regionalization.This process hasoccurred both in developed and indeveloping countries.According toKingdom (2005) there are fewpublications on regionalization.TheEuropean Union (EU) and theWorldBank are promoting this scheme indeveloping countries.The mainreasons for regionalization are toachieve economies of scale by servinga large number of people, and theability to access private sectorfinancing or funds from internationalinstitutions that are encouraged tofinance based on the benefits fromeconomies of scale.The most important factor in

regionalization is achieving economiesof scale.This is especially importantfor small cities which,being small insize,have difficulties in achievingeconomies of scale.According to areport prepared forWestern cityprojects (McFarlane,2003), the reasonsfor sharing access to water resourcesare financial, geographic inter-dependence,public health concerns,regulatory changes, and publicexpectations.According to SMCMartin, Inc

(1983), regionalization favorseconomies of scale, as the total costswill decrease with the increase inpopulation served.Also the savingsarising from the aggregation offinancial, human resources andtechnology contribute to achievingthe economies of scale.Since 1998, the most important

studies that have looked at theeconomies of scale in the water sectorare those by Kim and Clark (1998) inthe US,Garcia andThomas (2001) inFrance, and so on.These econometricstudies failed to show that economiesof scale are more powerful for smallcompanies in providing services to theWS sector, and that economies of scale

show the high financial costs of thewater and sewerage companies (WSCs)inAlbania.InAlbania there are currently 58

water and/or wastewater companiesoperating, and the performanceindicators show an unhealthy financialsituation.Overall, companies face moreor less the same problems, regardless oftheir severity,which varies from oneregion to another.Some of theseproblems faced by companies in thesector level according to the data byWater RegulatoryAuthority (WaterRegulatoryAuthority,2012) are asfollows:• A relatively low level of watersupply: there is 80% coverage ingeneral, of which 90% is in urbanareas and 60% in rural areas

• Coverage of sewerage services is at alevel of 50% in total, of which 70% isin urban areas and 17% in rural areas.

• The degree of losses (the water thatis produced and introduced into thesystem,but that does not generateany income) is approximately 67%,while the standard inWesternEurope is 20%,and furthermore thestandard according to theWaterRegulatoryAuthority (WRA) is30%.This is due to low levels ofmeter installation, illegal connectionsto the network and the networkbeing overused.

• Continuity of water supply is onaverage 10.8 hours,while thestandard inWestern Europe is 24hours a day.

According to theWorld Bank studyundertaken in 2010, it is estimated that80% ofWSCs are small cities with less

than 50,000 inhabitants, a fact thatshows there is a lack of economiesof scale (Padeco,2009).The poorperformance of these companies exertspressure on the government,which isobliged to provide subsidies.Althoughthese have been reduced, they stillremain high at 8% of total operatingcosts for 2012.The subsidies areprovided to cover the financial lossesof theWSC,while today's debate isthat subsidies should be provided tohelp the poor.InAlbania, there are no in-depth

studies on increasing the efficiency oftheWS sector.The only institutionsthat have conducted several studies inthe field are theWorld Bank and theGerman Bank for Reconstruction andDevelopment (KfW).Specifically, theWorld Bank has conducted studieson decentralization, increasing theefficiency of the sector and investmentprojects,while KfW has conductedstudies for investment projects carriedout in several cities including Korça,Pogradec,Kavaja,Kukës Rrogozhina,Lushnja,Berat,Kuçovë,Saranda,Gjirokastra,Lezha and Fier.Recentlya master plan for the necessaryinvestments in theWSC was preparedby a German consulting company,which analyzed the financing needsof theWS sector until 2040.

Literature reviewIt is known from the literature thatthe introduction of competition inthe water sector is more difficult thanin other sectors such as energy andtelecommunications,because thetechnical characteristics of the watersector differ from those of other

Table 2: Summary Statistics for the Basic OLS Models

For the Volumes Models:Variable Observations Mean Standard Minimum Maximum

DeviationLN OPER COST 203 10.78009 1.354354 8.317204 14.45287LN WAGE 203 5.844669 0.3050549 5.072671 6.808056LN WATER VOL PROD 203 7.500128 1.658615 4.60517 11.59958LN SEWER VOL COLL 186 6.167105 1.523327 2.079442 9.917996

For the Population Models:Variable Observations Mean Standard Minimum Maximum

DeviationLN OPER COST 203 10.78009 1.354354 8.317204 14.45287LN WAGE 203 5.844669 0.3050549 5.072671 6.808056LN POP SERV WATER 203 10.06328 1.466459 7.482214 13.7233LN POP ACCESS TO SEWER 195 9.698738 1.579222 5.607639 13.57662

For the Connections Model:Variable Observations Mean Standard Minimum Maximum

DeviationLN OPER COST 203 10.78009 1.354354 8.317204 14.45287LN WAGE 203 5.844669 0.3050549 5.072671 6.808056LN WATER CONN 199 8.15966 1.947109 .6931472 12.02472LN SEWER CONN 196 8.292021 1.414973 5.62852 11.92949



are more stable when company size isdetermined by the volume of waterproduced.But these studies also showthat the most powerful economies ofscale exist when the number ofconnections or the population servedis used as an indicator for the size ofthe company (Tynan and Kingdom,2005).The overall result is that in allcases analyzed in various studies, theregionalization process has as its mainfactor increased efficiency as indicatedby the economies of scale theory.With the introduction of the

concept of integrated water manage-ment,merging companies becomesa necessity in order to maximizeeconomic and social welfare in anequivalent way without compromisingthe sustainability of ecosystems.Forexample in England andWales,economic growth and the emergenceof pollution problems forced thecentral government to reorganizemanagement of water resources bybringing together more than 200water companies and 1400 seweragecompanies to form only 10 regionalcompanies.These companies had theobjective of managing water and alsosewerage services in an integratedmanner (Kingdom,2005).According to McFarlane (2003),

financial benefit is provided bygeographic interdependence,meaningthat some areas may have problems ingaining access to water resources, andthus merging water companies alsooffers access for areas that are unableto access water.

MethodologyThe basic units of analysis in this studyareWSCs operating inAlbania thatoffer both types of services,namelyaccess to water supply and seweragesystems for the period 2006 to 2012.First, the analysis started for all waterand/or sewerage companies, 54 intotal, for that period.Currently inAlbania there are 58 water and/orsewerage companies operating,butout of these four did not exist duringthe period of analysis.From the data analysis, it was found

that not all companies offered bothwater and sewerage services. In thiscontext, the analysis was reduced from54 to 29 companies that constitute theselected sample.The sample of 29companies is representative because itcovers 80% of the population served.The data used are publicly available onthe web page of the GeneralDirectorate ofWater and Sewerage.Research questions raised in this

study were:• What are the factors that affect thecompany's operating costs?

• Which of the factors is the mostimportant in the operating costs?

• What are the criteria used forrankingWS companies by theirefficiency?

The quantitative hypotheses are asfollows:• The reduction of technical lossesreduces the volume of waterproduced and the annualoperating costs

• The effect of scale in relation tothe population served by thewater/sewerage company should beimportant in reducing relativeoperating costs

• Increasing the number of metersinstalled leads to a significantincrease in operating costs

• The current methods for rankingthe companies by efficiency shouldbe changed, taking into accountranking according to operatingconditions

This paper is based on econometrictechniques that use parameters orvariables. In general, a non-parametricdata envelopment analysis model canprovide answers to some importantquestions.However, this study relies onthe parameter model to evaluate themost important parameters and to testthe relative importance of variables(Cubbie &Tzanidakis, 1998;Berg &

Lin,2005).When it comes the questionof measuring efficiency with aneconometric model, analysts have twooptions: cost functions and productionfunctions.Neither is perfect, due tolimited data.The preference for one orthe other depends on the particularcircumstances as analyzed bySabbione (2006).One of the most important circum-

stances is the environment in whichthe company operates.When we talkabout maximizing profit, a productionfunction can be a natural choice fora company,which can choose itsproduction level.However, if thecompany has an ‘obligation to serve’, itwill need to produce as much as isrequired by customers. In this regard,the production function may not beadequate. In the case ofAlbania a costfunction may be applicable because theWSCs are of different size, and theservice areas are also different.Such companies have an obligation

to serve the population, and in thisregard it is more appropriate for acompany to produce as much as isrequired by its customers. Since 42%of the companies serve fewer than50,000 inhabitants, this indicates thatthere is a lack of economy of scale, andprofit maximization on the basis ofproduction would not be appropriate.

Table 3: Basic OLS Models for 2006-2012

Dependent Variable: (ln) Operating Cost

Variable MODELVolumes Population Connections

Constant 3.850199*** -0.1922845 2.971766***(4.82) value t (-0.30) value t (3.83) value t(0.000) value p (0.764) value p (0.000) value p

Wage 0.231528 0.4476861*** 0.0913152(1.61 value t (4.01) value t (0.65) value t(0.110) value p (0.000) value p (0.517) value p

Water Volume Produced 0.6544855***(12.05) value t(0.000) value p

Sewer Volume Collected 0.107497*(1.84) value t(0.067) value p

Population served 0.6366187***with water (8.94) value t

(0.000) value pPopulation with 0.2021717***

access to sewer (3.02) value t(0.003) value p

Water Connections 0.0770307***(2.67) value t(0.008) value p

Sewer Connections 0.8050484***(20.56) value t(0.000) value p

Observations 186 195 192R2 0.8364 0.8918 0.8498



which exclude the external variable Z.Although there is limited information,specifications show that a highpercentage of variability in operatingcosts of companies is explained onlywith selected independent variablesthat are variables of the production andprice inputsW.Each model has an output variable

associated with the water service, andan output variable associated withthe sewerage service. Independentvariables are grouped into threecategories (Sabbione,2006):TheVolumes model except wage

includes as variables:• The volume of water produced inthousands of m3 (000 m3)

• The volume of sewerage collectedin thousands of m3 (000 m3)

• The volume of water measured inthousands of m3 (000 m3)

The Population model except wageincludes the variables:• The population served by water thatis the number of people served withwater services

• The population number with accessto sewerage services

The Connection model except wageincludes the variables:• The number of water connections• The number of sewerage connections

The theoretical hypotheses are asfollows:• Hypothesis H0: the annual operatingcost is not influenced by the inde-pendent variables of the threemodels, and in this case there is nolinear relationship

• The alternative hypothesis H1 ≠H0:annual operating costs are signifi-cantly influenced by the indepen-dent variables of the three models,or at least one of the variableslinearly affects the dependentvariable,which in this case is theannual operating cost

Results of the three modelsFrom the results of the Basic OLSmodels, some preliminary andexpected conclusions can beextracted.First,we see that thecoefficient of the variableWAGE isalways positive,but not always impor-tant.WAGE is important whenchoosing production by having thepopulation served as an output. In allthree regressions, the models confirmthe hypothesis that the firm's operatingcosts will increase if the price of laborgoes up,but confirmation of thehypothesis becomes more specificwhen we choose to produce on thebasis of the population served,because all indicators are positiveand significant.

We assume that aWSC canmaximize its profits by minimizing thecost of producing some exogenouslygiven output level, subject to theavailable technology (that is, theproduction function).The solution tothis optimization (cost minimization)problem is a cost function (Sabbione,2006):

Equation (1):C= f(Y,W,Z)

WhereY is the output vector,W is thevector of input prices, and Z is thevector of environmental characteristics.It is assumed that this cost function C(.) can be deconstructed as the productof two functions g (.) and h (.).Thefunction g (.) will have the input pricesW and outputsY as arguments,whilethe function h (.) will incorporate theexogenous variable Z that will affectthe technology of the company.Thecost function is then:

Equation (2):C= g(Y,W) h (Z)

Characteristics of the cost functionHaving introduced the mathematicalside of the cost function, this sectionanalyses the regression model used andthe results of the test of the significanceof the model.The model used is amultiple linear regression,whichincludes several variables that affectthe efficiency of aWS company.Theoretically, the model has thefollowing formula (Ramanathan,2002):

Equation (3):Yt= β1 + β2 Xt 2 +……..βkXtk + µt

Where:t is the number of surveys whichvaries from 1 to t;Yt is the dependentvariable;β1 is the constant;β2…..βkare the coefficients to be determined;Xt 2…. Xtk are the independentvariables; and µt is the error term.In this paper, the dependent variable

is the annual operating cost of thecompany, excluding amortization. InAlbania, the majority of companieshave financial losses, and subsidies fromthe central government are allocated tocover the annual operating costs.The challenge for allWSCs, as

described in the National SectorialStrategy 2011-2017 for water andsewerage, is for the company to initiallycover its operating costs and then as afinal objective to cover its total costs,which would be an indication thatthe company is moving to financialautonomy.Specifically, the Strategystipulates that by the year 2017,coverage of direct operating costs withrevenues and payables must reach100%.This is a strong argument forcompanies to understand whichvariables most affect the variability ofthe dependent variable.The first set of results consists of

three different specifications of the costfunction depending on the outputvariables chosen: aVolumes model, aPopulation model and a Connectionsmodel.These versions are called basicOrdinary Least Square (OLS) models,

Table 4: Summary Statistics for the Basic OLS Models with the Environmental Variable Z

For the Volumes Models:Variable Observations Mean Standard Minimum Maximum

DeviationLN OPER COST 203 10.78009 1.354354 8.317204 14.45287LN WAGE 203 5.844669 0.3050549 5.072671 6.808056LN WATER VOL PROD 203 7.500128 1.658615 4.60517 11.59958LN SEWER VOL COLL 186 6.167105 1.523327 2.079442 9.917996LN VOL WAT MEASURED 186 5.612998 1.955937 -1.07881 10.57029

For the Population Models:Variable Observations Mean Standard Minimum Maximum

DeviationLN OPER COST 203 10.78009 1.354354 8.317204 14.45287LN WAGE 203 5.844669 0.3050549 5.072671 6.808056LN POP SERV WATER 20 10.06328 1.466459 7.482214 13.7233LN POP ACCESS TO SEWER 195 9.698738 1.579222 5.607639 13.57662LN VOL WAT MEASURED 186 5.612998 1.955937 -1.07881 10.57029

For the Connections Model:Variable Observations Mean Standard Minimum Maximum

DeviationLN OPER COST 203 10.78009 1.354354 8.317204 14.45287LN WAGE 203 5.844669 0.3050549 5.072671 6.808056LN WATER CONN 199 8.15966 1.947109 .6931472 12.02472LN SEWER CONN 196 8.292021 1.414973 5.62852 11.92949LN VOL WAT MEASURED 186 5.612998 1.955937 -1.07881 10.57029



Specifically, the mathematicalrelationship between variables for thethree models is given below:

Equation (4):Volumes modelOperating cost = 3.8 + 0.23 × wage +0.65 × volume of water produced +0.11 × volume of sewerage collected +0.54

The result is that the variables of theVolumes model have a positive linearrelationship with the operating costs.What is evident in this equation is thatan increase of one percent of thevolume of water produced willincrease operational costs by 0.65%and an increase of one percent of thevolume of sewerage collected willincrease the operational costs by0.11%.

Equation (5):Population modelOperating Cost = -0.19 + 0.45 ×wage + 0.64 × population served withwater + 0.20 × population with accessto sewerage services + 0.44

The result is that the variables of thepopulation model have a positivelinear relationship with the operatingcosts except for the constant.What isevident in this equation is that anincrease of one percent in the popula-tion served with water will increaseoperational costs by 0.64% and anincrease of one percent in the numberof people with access to sewerageservices will increase operational costsby 0.20%.

Equation (6):Connections modelOperating Cost = 2.9 + 0.09 ×Wage+ 0.07 ×Water Connections + 0.8 ×Sewerage Connections + 0.52

The result is that the variables of theconnections model have a positivelinear relationship with the operatingcosts.What is evident in this equationis that an increase of one percent in thewater connections through meters willincrease operational costs by 0.07%,and an increase of one percent in thesewerage connections will increase theoperational costs by 0.8%.In all three models, the coefficient of

determination R2 is high,which showsthat 83% of operating cost variability isexplained by the variables of thevolumes model, 89% of the variables ofthe population model and 85% by thevariables of the connections model.Asa general result, all three models havehigh R2 to explain the variability ofannual operating costs, and none ofthem is excluded,but of the threemodels, the model that explains bestthe variability of operating costs is thePopulation model,with a coefficient ofdetermination of 89%.

The next step is to test the environ-mental variable Z to isolate its effect onthe operating cost and evaluate it indetail.The additional variable Z wasincluded only if it added significantexplanation to the operating cost byhaving a stable,meaningful and statisti-cally significant coefficient.The result isthat the variables of all the threemodels have a positive linear relation-ship with the operating costs. In allthree models the coefficient of deter-mination R2 is high,which shows that85% of the operating cost variability isexplained by the variables of thevolumes model, 88% of the variables ofthe population model and 84% by thevariables of the connections model.Asa general result all three models have ahigh R2 to explain the variability of theannual operating cost, and none ofthem is excluded,but of the threemodels, the model that explains bestthe variability of operating costs is thePopulation model,with a coefficientof determination of 88%.As a finalconclusion, adding the environmentalvariable in our testing does not create asignificant change.As mentioned in this article, the level

of water losses in the water and sewer-age sector is high.According to the

WRA, sector performance report for2012, it is reported that the level oflosses in the sector is 67%.Similarly, intheWS sector strategy for 2011-2017the objective is that by year 2017 lossesbe reduced to 40%,a change of 27%.On the basis of the results reached

through the regression formula, thefinal conclusions are as follows:• A reduction of technical losses from67% to 40% will reduce the volumeof water produced by 45%,andoperating costs will decrease by 29%.(Hypothesis 1 is verified)

• Serving a greater population of 1%to 5%, the operating cost increases byless than the increase in population(from 0.57% to 2.8%). (Hypothesis 2is verified)

• Increasing the number of metersinstalled (from 10% to 20%) theoperating costs increase much less(0.7< C < 1.4 %). (Hypothesis 3 isrejected)

Rankings of the water and seweragecompaniesTo estimate the efficiency level oftheWSCs, three stochastic costfrontier models were run.Basic OLSregressions were valid to identify thesignificant variables that most explain

Table 5: Basic OLS Models for 2006-2012 only for the Environmental Variable Z

Dependent Variable: (ln) Operating Cost

Variable MODELVolumes Population Connections

Constant 4. 27644*** - 0.0614173 3.162032***(5.31) value t (-0.08) value t (3.87) value t(0.000) value p (0.935) value p (0.000) value p

WAGE 0.1656319 0.3893714*** 0.0796103(1.17) value t (3.18) value t (0.55) value t(0.245) value p (0.002) value p (0.584) value p

Water Volume Produced 0.53318***(9.22) value t(0.000) value p

Sewer Volume Collected 0.1128604*(1.87) value t(0.063) value p

Population served 0.5749276***with water (7.02) value t

(0.000) value pPopulation with 0.2840244***

access to sewer (3.62) value t(0.000) value p

Water Connections 0.0475238(1.41) value p(0.160) value p

Sewer Connections 0.7664642***(15.51) value t(0.000) value p

Water Volume Measured 0.1500166*** 0.002597 0.0768459 **(5.11) value t (0.09) value t (2.23) value t(0.000) value p (0.930) value p (0.027) value p

Observations 175 178 179R2 0.8507 0.8822 0.8413


the variability of annual operatingcosts, but the construction of anefficiency ranking needed to accountfor both inefficiency and randomnessin the error term.The error termexplains the random factors arisingfrom natural disasters or otherconditions that are outside thecompany’s control and do not dependon it, such as geographic position -being located in an area that serves alimited number of people or that isa long distance from access to theriver basin.What is evident from the ranking

results is that the five bestWSCs forefficiency during 2012 are the samecompanies across the three models.These utilities are:Tirana,Durrës,Berat& Kuçovë,Fier and Elber Ltd.Some ofthe factors that could have contributedto the efficiency are good manage-ment,qualified staff, accumulatedexpertise and so on.The fiveWSCswith the worst efficiency are the samecompanies across the three models.These are:Delvinë,Rrogozhinë,Libohovë,Krastë and Rubik.Theconstant term in the stochastic modelin this case takes into account otherfactors besides those above.Factorsthat affect inefficiency could be badmanagement,unqualified staff, lack ofexpertise, and so on.The stochastic test results are in line

with the theory of efficiency on the

economy of scale and the conclusionsof empirical studies carried out byrecognized authors,who conclude thatthe greater the increase in the popula-tion served, the lower the cost per unit.This finding is relevant in the case ofAlbania, that an increase in the numberof population served leads to anincrease in population density,whichimpacts the achievement of economiesof scale.

Comparison of the PerformanceRanking of WSCs with ranking by theWRAInAlbania,performance monitoring oftheWS sector started in 2006 based ona benchmark set by the benchmarkingMonitoring Unit and later by theWRA.It is important to compare theranking of the companies based on theresults of this scientific study and theranking produced by theWRA.Specifically, theWRA groups the

companies in three categories using ascriteria the number of connections tothe water system.According to theWRA, the grouping as per connectionscontrary to the size of the service areahelps to differentiate between large andsmall companies.For each group, thebenchmark is calculated for the mainsector indicators and the performanceof each company is then calculated bycomparing with the benchmark ofeach indicator.


Ranking of the best companiesaccording to both evaluations is asfollows:

Scientific Paper WRA1.Tirana 1.Tirana2.Durrës 2.Korçë3.Elber 3.Elber4.Berat & Kuçovë 4.Vlorë5.Shkodër 5.Shkodër

Ranking of the worst companiesaccording to both evaluations is asfollows:

Scientific Paper WRA1.Delvinë 1.Pukë Fshat2.Rrogozhinë 2.Orikum3.Libohovë 3.Gjirokastër Fshat4.Krastë 4.Libohovë5.Rubik 5.Has

The test results confirm Hypothesis 4,that the current methods for rankingcompanies according to efficiencyshould be changed to take intoaccount the ranking as per operatingconditions.This was shown by theconstant term that explains otherfactors not captured by the volumesmodel, population model and connec-tions model.As is evident,both rankings are

different.There are several reasons forthe differences in the rankings foundin this comparison.Both types ofrankings use different methodologies.Also, the ranking of the scientific paperconsiders both types of services, and asa result companies that offer onlywater service are not included,whichis an added value of the scientificpaper,while theWRA ranks allcompanies regardless of the servicesthey provide.In addition, the ranking criteria

used by theWRA to group companiesaccording to their size are on thebasis of water connections,while thecriteria used by the scientific paper areaccording to three models: theVolumes model, the Population modeland the Connections model.What isfound is that the ranking used by theWRA does not take into considera-tion the part of the population that issupplied with water,but is notequipped with meters. It is a fact thatpart of the population served withwater may not be equipped withmeters, but still companies receivetheir payables using a fixed method forprice calculation. In this context, this isrevenue for the company and affects itsfinancial performance.This can be apoint of strength in the ranking usedin the scientific paper.

ConclusionsAlbania is a country rich in waterresources, and with the current

Table 6: Final Ranking for year 2012 for W/S Utilities according to stochastic cost frontier models

Volumes Model Population Model Connections Model

Tiranë Tiranë TiranëDurrës Durrës DurrësBerat & Kuçovë Berat & Kuçovë Berat & KuçovëElber Elber ElberShkodër Shkodër ShkodërFier Fier FierGjirokastër Gjirokastër GjirokastërKorçë Korçë KorçëSarandë Sarandë SarandëLushnje Lushnjë LushnjëKavajë Kavajë KavajëLezhë Lezhë LezhëPogradec Pogradec PogradecBurrel Burrel BurrelKamëz Kamëz KamëzKrujë Krujë KrujëKukës Kukës KukësMirditë Mirditë MirditëLibrazhd Librazhd LibrazhdMallakastër Mallakastër MallakastërErsekë Ersekë ErsekëFushë Krujë Fushë Krujë Fushë KrujëPukë Pukë PukëFushë Arrëz Fushë Arrëz Fushë ArrëzDelvinë Delvinë DelvinëRrogozhinë Rrogozhinë RrogozhinëLibohovë Libohovë LibohovëKrastë Krastë KrastëRubik Rubik Rubik



capacity of the resources it possesses itis able to meet the needs of the wholepopulation.But water management isnot efficient because only 43% ofwater resources are used, the produc-tion is 31% higher than the demandfor drinking water, and the volume ofwater billed (sold) only meets 54% ofthe demand at the sector level.Thedegree of losses is approximately 67%,while the standard inWestern Europeis at 20%,and furthermore thestandard according to theWRA is30%.This is due to the installation ofmeters being at low levels, illegalconnections to the network and thefact that the network is overused.One way of addressing the

inefficiency of the sector is to mergeseveral water companies and createbig regions,known as regionalization.This process has occurred bothin developed and in developingcountries.According to Kingdom, themain reason for regionalization is toachieve economies of scale by servinga larger population, and the ability toaccess private sector financing or fundsfrom international institutions.According to SMCMartin, Incregionalization favors economies ofscale, as the total costs will decreasewith the increase in population served.Also, the savings arising from theaggregation of financial, humanresources and technology contributeto the achievement of the economiesof scale.In all three models, the coefficient

of determination R2 is high,whichshows that 83% of the operating costvariability is explained by the variablesof theVolumes model, 89% by thevariables of the Population model and85% by the variables of theConnections model.As a generalresult, all three models have a high R2

to explain the variability of the annualoperating cost, and none of them isexcluded,but of the three models, themodel that explains best the variabilityof operating costs is the Populationmodel,with a coefficient of determi-nation of 89%.The ranking criteria used by the

WRA to group companies accordingto their size are are on the basis ofwater connections,while the criteriaused by the scientific paper are accord-ing to three models: theVolumesmodel, the Population model and theConnections model.What is found isthat the ranking used by theWRAdoes not take into consideration thepart of the population that is suppliedwith water but it is not equipped withmeters. It is a fact that a part of thepopulation served with water maynot be equipped with meters, butcompanies still receive their payablesusing a fixed method for price

calculation. In this context, this isrevenue for the company and affectsits financial performance.This can bea point of strength in the ranking usedin the scientific paper.The testing used for efficiency

through the econometric model is ascientific contribution recommendedfor consideration by theWRA.Efficiency is just one aspect associatedwith economies of scale. In addition tothis factor there are other factors suchas institutional, political, administrativeand legal that the companies mustconsider. It is recommended that inorder for the companies to achieveeconomies of scale, regionalization andthe creation of bigger regions could bean option. In this context, in theframework of the new territorial andadministrative reform that is happeningright now inAlbania,policy makersshould define a regionalization modelthat takes into consideration not onlythe economic factor but also otherfactors as mentioned here.Clearly,much work remains.For

the purposes of rewarding goodperformance and penalizing weakperformance, scholars and practitionersneed to develop benchmarkingprocedures that can pass legalchallenges.Econometricians mustcontribute to the debate on increasingthe efficiency of theWS sector inAlbania. In order to address theefficiency issue, academicians mustcollaborate more closely with theWRA and other policy makers todecide on a standardized rankingassessment.In the context of the new territorial

and administrative reform, the govern-ment should promote the offering ofwater and sewerage services by thesame operator.This could help thecompany to reduce costs and increasesavings, thus improving efficiency.�

ReferencesRamanathan,R (2002) Introductory economet-rics with application,South-Western CollegePub,USA,pp.9,82,50-51,144.Banka Botërore (2004) Studimi mbi

Decentralizimin dhe Privatizimin e Sektorit tëUjit në Shqipëri.Banka Botërore (2011) Raporti nr. 64412-

AL mbi Decentralizimin dhe Realizimin eShërbimeve në Shqipëri:Qeverisja në Sektorin eUjit.Berg,S and Lin,C (2005) Consistency in

performance rankings: the Peruvian water sector,PURCWorking Paper.Cubbin, J andTzanidakis,G (1998)

Regression versus data envelopment analysis forefficiency measurement: an application to theEngland andWales regulated water industry,Utilities Policy 7, pp.75-85.Drejtoria e Përgjithshme e Ujësjellës

Kanalizimeve (2011),Raporti 5 vjetBenchmarking, pp.17,

http://www.dpuk.gov.al/5_vjet_benchmarking_vilma_bibolli_2011.pdf.Enti Rregullator i Ujit (2012),Raporti i

Performancës për vitin 2012,pp.46-48.Garcia,S andThomas,A (2001)The

structure of municipal water supply costs:application to a panel of French local communi-ties, Journal of ProductivityAnalysis 16(1), pp.5–29.Kingdom,W (2005) Models of aggregation

for water and sanitation provision,Water Supplyand Sanitation Notes,Note No.1, January, pp.12,15.McFarlane,S (2003),Regional water works:

sharing urban water services,Western CitiesProject Report #28,pp.7.Mergos,G (2005) Private participation in the

water sector: recent trends and issues,EuropeanWater 9/10,pp.59-75.Ministria e Punëve Publike dheTransportit

(2011),Strategjia Kombëtare e Furnizimit meUjë dhe Kanalizimeve,2011-2017.http://www.erru.al/doc/Strategjia_Kombetare_e_Furnizimit_me_UK.pdfPadeco CO,LTD (2009),Feasibility Study

of Regional Utilities in theWater andWastewaterSector ofAlbania, pp.13.Quiggin, J (1995) Does privatization pay?

Australian Economic Review,No.110 (2ndquarter) pp.23-42.Sabbioni,G (2006) Econometric measures of

the relative efficiency of water and sewerageutilities in Brazil, pp.3-6.SMCMartin, Inc (1983) Regionalization

options for small water systems,Washington,DC:U.S.Environmental ProtectionAgency, pp.

19.Tynan,N and Kingdom,B (2005) Optimalsize of utilities?,Return to scale in water:evidence from benchmarking.Urio,P (2010) Public Private Parnterships:

success and failure factors for in-transitionalcountries,University Press ofAmerica,Maryland,pp.26,30.Youn,KH and Clark,RM (1988)

Economies of scale and scope in water supply,regional science and urban economics 18,no.4,pp.479.http://www.un.org/geninfo/bp/enviro.html

This article is based on a paper presented atthe IWA Regional Utility ManagementConference, Tirana, Albania, 13-15 May 2015.


BudapestWaterWorks (BWW),which celebrated the 147th

anniversary of its establishmentthis year, is one of the significantwater utility suppliers of theCentral-Eastern-European regiongiven the size of the populationserved and the outstanding level oftechnology.During its over 140-year lifetime,

BWW has developed along with thecapital city.The first temporary watertreatment works was established in1868, followed by the construction oftwo water treatment works in Buda in1882.Then the Káposztásmegyer watertreatment works was put into service in1904,which was the most up-to-datesuch facility in Europe at that time.The company has made immense

progress in the past 100-plus years.Theearly plants that used to be operated bysteam engines have been replaced bymodern, automated facilities that meetthe most exacting hygiene require-ments. In almost a century and ahalf, newer and newer water towers,pumping stations, and pipe networkshave been built, and wells put intooperation.The company’s pipelinenetwork has reached more than 5,000km in length throughout the city.Thecompany’s water treatment,networkoperation and water quality testingactivities are enabled by cutting edgetechnologies.Today, the potable waterproduction and supply systems arecompletely automated.Looking at its core activities, the

company deals with potable water

treatment,potable water production,network operation and potable watersupply – today it supplies more thantwo million people with healthy pipedpotable water – as well as with waste-water treatment and related services,which are supported by the applicationof world-class technologies.

BWW’s international activitiesThe professional queries we receivefrom countries all around the worldindicate a demand for cooperationamong our partners.Negotiations havebeen started recently with severalEuropean andAsian cities in the fieldsof consultancy, technology develop-ment and main contracting, andcontracts have been signed.Amongthem, the reconstruction and capacityincrease of two water treatment plantssupplying potable water to the capitalcity of Sri Lanka is worth mentioning.The NationalWater Supply andDrainage Board of Colombo awardedthis contract to our company.Apartfrom that, the Hungarian governmenthas concluded a comprehensive watersector framework agreement withVietnam,within which, as per theplans,BWWwill build two watertreatment plants.Furthermore,memo-randa of understanding in the field ofcooperation have been concludedbetween our organisation and theutility providers of Baku inAzerbaijan,Ohrid in Macedonia,Tirana inAlbania, Istanbul inTurkey, and Urunqiin China.It is of special value that in the

period since the change of regime,BWW has already walked a path ofmarket, technological, operationaland economic development that isstill ahead of these cities.Regional

THE BUDAPEST WATERWORKS WATER SAFETY SYSTEM

Milestones and results for the Budapest Waterworkswater safety systemThe Budapest Waterworks (BWW) was among the first to establish a water quality

system, both in Hungary and in Europe. As a first step, a system was established for the

Csepel island water treatment works, Budapest, as this has complex drinking water

treating technology due to the water’s iron and manganese content, which necessitates

the active intervention of the operator to maintain water quality. Géza Csörnyei and

Genovéva Frank look at the eight-year-long development process for the BWW potable

water safety system, the structural changes, water quality events and water quality

event management experiences. In addition, the authors look at modification of the

structure of the drinking water safety manual due to mergers and legal changes, and

water quality results achieved through more polished, efficient operation.

Géza CsornyeiOperation Director of Water Division,Waterworks of Budapest, Hungary.Email: [email protected]

Genovéva FrankOperational engineer,Waterworks of Budapest, Hungary.


The control centre(credit BWW)



co-operation,by strengtheningalternative revenues not directlygained from potable water sales, byincreasing the value of the company,by gaining experience related to thecore activities, and by extendingknow-how, serve to achieve ourowners’ goals.

BWW water safety systemsWater safety plan legal backgroundWater utility companies were firstobliged to prepare water safety plansaccording to the requirements of the65/2009. (III. 31.) GovernmentDecree,on the basis ofWHO guide-lines.Requirements for the content ofthese plans were not fixed then in anyHungarian legislation.Later, theserequirements were determined in the201/2001. (X.25.) GovernmentDecree regarding quality requirementsfor drinking water and the watersampling process.

AuthoritiesThe water safety plan has first to besent to the National Center for PublicHealth for professional acceptance,then,having this expert opinion, to theNational Public Health and MedicalOfficer Service for authorisation.

Steps for preparing the water safety planBWWwas among the first in Hungaryand Europe to start building a watersafety management system.Hungarian

legislation did not yet deal with thepreparation of water safety plans,butBWWwas already planning theimplementation of a water safetymanagement system.During theimplementation,BWW used the ISO22000:2005 standard on food safetymanagement systems and requirementsfor any organization in the food chain,as being more comprehensive.• The water safety managementsystem was certified:

• firstly for southern water production

in 2007• then for northern water productionin 2007

• then for the entire water productionand network operation in 2009

Later,BWW implemented an integrat-ed management system by amalgamat-ing four management systems:• quality according to ISO 9001:2008• environmental according to ISO14001:2004

• water safety according to ISO

Figure 1: Waterlevel of riverDanube

Wells under waterduring flooding(credit BWW)



actions for drought periods.A collec-tion of decision matrices related towater quality complaints was createdto manage complaints about waterquality and background pollution.

Decision matricesSteps for managing complaints weretaken in 2005 according to:• BWW’s technological process stepsWater quality parameters deter-mined by the 201/2001. (X.25.)Government Decree regardingquality requirements for drinkingwater and the water samplingprocess

• Technological water quality resultsdetermined by the water safety team

Various corrective actions arenecessary to manage non-compliantresults from wells and well pumphouses, collection points, reservoirsand consumer points.

Implementation of and experienceswith the integrated managementsystemPseudomonas compliance in SurányPseudomonas was found duringscheduled sampling of the water in theROCLA pipe at Surány in 2010.Theconclusion was that the sampling taphas to be located as close as possibleto the sampling location,with noaffecting factors in between (suchas a pump,or online measuringinstrument).

Red sludge disasterThe red sludge disaster happened on 4October 2010 at Kolontár-Devecser-Somlóvásárhely.The corner of the10th section ofAjkaAlumina plant’stailing pond burst, and the entire floraand fauna ofTorna creek weredestroyed by red sludge.The pollutionpassed through the rivers Marcal,Rába, and Mosoni-Danube andreached the river Danube.BWW’slaboratory started a samplingprogramme for the river Danube andwells,monitoring pH,conductivity,heavy metals, earth metals andperforming toxicological tests on thewaters of the river Danube.TheWaterProduction Department completedthese tests by measuring the pH everyfour hours.On the basis of measurements

lasting for one and a half months, it canbe declared that the red sludge disasterhad no effect on the drinking watersupplied by BWW.

Supervision of reservoir vents (see tables)Venting of water storage reservoirs isnecessary due to fluctuating waterlevels.Reservoir vents create a waterquality risk as drinking water cancome into contact with the environ-

22000:2005• occupational health and safetyaccording to BHOHSAS18001:2007

The BWWwater safety plan achievedprofessional acceptance from theNational Center for Public Health on23 June 2011, then permission fromthe National Public health and MedicalOfficer Service on 7 November 2011.BWW has until 7 November 2015 toinitiate public health surveillance ofthe plan.

Main elements of the water safety planThe BWWwater safety plan contains adescription of the water supply system,hazard analysis and risk evaluation,determination and evaluation ofmeasurements for checking risks,establishment of a checking monitor-ing system,and establishment of adrinking water risk managementsystem.

The relationship between the watersafety plan and the quality manage-ment systemAdvantages and disadvantagesCouncil directive 98/83/EC directive3 November 1998 on the quality ofwater intended for human consump-tion gives recommendations formember states to implement anappropriate quality managementsystem for checking whether drinkingwater intended for human consump-tion complies with the requirementsof this Directive.The ISO 22000 standard determines

uniform requirements for HACCPplan and makes HACCP certifiable,

and describes an internationallyaccepted quality management system.The HACCP plan and water

safety plan contain specifics on riskevaluation,operational monitoring,hazard analysis, protection of waterresources,which are essential, butneither of them can be certified per se.However, the food safety managementsystem according to ISO 22000:2005can be.The food safety management system

according to ISO 22000:2005 canbasically serve as a good basis for adrinking water safety plan,but cannotcontain sufficiently detailed hazardanalysis and risk evaluation for thewhole water supply system.Application of the food safety

management system (ISO 22000:2005)provides an enormous advantageand continuous development forthe water safety plans and HACCP,with its all advantages at BWW:itincludes an annual examination, isinfrastructure and resource-centered, and offersdocument management andinternational certification.

Specialties of risk assessmentDue to the condition of BWW’s waterresources even the raw water can bementioned as a hazard in some cases.During the operation of bank filtrationwells, potential industrial or municipalcontamination, temporary floodingand drought periods have to be consid-ered. Internal regulations have beencreated for the fast management ofsudden operational statuses suchas event management at water produc-tion facilities, a flood manual, and

Laboratory testing(credit BWW)



ment if the vents are incorrectlyprotected.To prevent contamination, aclosing fiberglass filter layer has beeninstalled at the end of the vents since2010.The filters are replaced everyyear.They are contaminated by iron(Fe),manganese (Mn), antimony (Sb),aluminium (Al), zinc (Zn), lead (Pb),Phenanthrene, fluoranthene,pyrene,chrysene,benz(b)fluoranthene,pollen,plant parts and insect wing scales.CFDsimulation was performed to examinethe flow of dust.

FloodingRegarding earlier high water levels,flooding in 2013 brought challengesfrom an operational point of view.

Hypo dosersMany chlorinators have beenreconstructed and have operated ashypo dosers since 2014.Among othermethods tested,use of a separatedflowing stream of ‘driving water’ fordilution with the modification of the

PLC program was successful.Theminimum dosage limit was eliminatedand flow-rate proportional dosage hasbeen implemented.The precipitationof hypo crystals has been accelerateddue to the use of the driving waterstream.Checking and maintenance ofthe dosing lance is performed everyweek, and an acid wash every threeweeks.

Actions to improve water qualityAccredited laboratory and rapid testlaboratoryThe laboratory was established in1960.For the accreditation, the entiresampling system was reformed;use ofcars for sampling and manual measure-ments were implemented.The processof sampling was accredited in 2004.Later, to make the administration of thesampling process easier, a PDA mobileworkstation and the LIS laboratoryinformation program were implement-

ed.The laboratory instrumentation hasbeen continuously expanding anddeveloping over the past 10 years.Informative microbiological rapid

test measurements,which are used toexamine reservoir status followingwashing,have been operating efficient-ly for years.The results are available 24hours earlier than accredited laboratoryresults, and are in line with them.

Mobile water purification and packagingsystemBWW has been dealing with the topicof cleaning and packaging water in caseof catastrophes and extraordinarysituations.Our equipment can be usedto supply smaller settlements, partsof settlements, industrial facilities,institutional consumers and settlementpopulations with healthy drinkingwater.Though the equipment is partof our own emergency water supply, ithas been used in several emergency

Table 1: Contaminants of fiberglass filter

Parameter Measured value Limit valueFe (total) 35 µg/l 200 µg/lMn (total) 2,5 µg/l 50 µg/lSb 1,2 µg/l 5 µg/lAl 60 µg/l 200 µg/lZn 13 µg/l - µg/lPb 2,7 µg/l 10 µg/l

Table 2: PAH content of fiberglass filter

Phenanthrene 22,47 µg/lFluoranthene 29 µg/lPyrene 16,5 µg/lChrysene 15,72 µg/lBenz(b)fluoranthene 17,91 µg/l

UV treatmentinstalled inBudapest’s oldestpumping station.(credit BWW)



Lead in the drinking water networkThere are almost 4000 pieces ofconnection pipe that are known to bemade of lead in BWW’s network.Regular analysis is performed on leadconnections at different points in thesupply area to check compliance withthe 10 µg/l dissolved lead limit valuefor water from consumer taps definedby the 201/2001. (X.25.) GovernmentDecree.The entire connection pipe isreplaced when the lead concentrationis above the limit. Similarly, leadconnections are replaced instead ofbeing repaired during maintenancework.

Asbestos cement pipe networkSome 42% of the material of the pipenetwork operated by BWW is made ofasbestos cement.Asbestos cementmeans a risk mainly during repairworks when it is necessary to cut thepipe.Exposure of workers to asbestosfibres in the air must be kept below 0.1fibre/ml.So ensuring the use of therequired labour protection devices,and the choice of cutting andassembling technology and appropriatewaste management are essential.Nevertheless, reconstruction ofasbestos cement pipes is due becauseof their age and to ensure the safetyof water supply.

ConclusionImplementation and continuousdevelopment of the drinking watersafety management system in theeveryday life of BWW is a good toolfor improving water safety actions,enabling BWW to cope with the waterquality problems that occasionallyoccur.BWW’s risk assessment methodand decision matrices, andthe water quality improvement actionstaken over the past years have madeit possible to react efficiently towater quality events and,with thecontinuous improvement, to provideour consumers with highly-compliantdrinking water and give the chanceto solve potential challenges in thefuture.�

ReferencesAll the information above is BWW’s ownmethodology and system, and is based onBWW’s own data.

This article is based on a paper presented atthe IWA Regional Utility ManagementConferences, Tirana, Albania, 13-15 May2015.

situations abroad.The equipment canbe transported on EUR pallets or in10 or 20 feet standard containers.Theequipment is considered to be amobile plant, so the HACCP system isimplemented and operated for them.

Drinking water dispenser bottle fillerBWW owns a little drinking waterdispenser bottle filler plant,whichinitially served only the hot and coldwater supply for our own employees.The plant underwent technologicaldevelopment and reconstruction in2012.Monitoring of the product andrelated raw and auxiliary materials iselaborated in the operating HACCPsystem, in accordance with theCodexAlimentarius and other legalrequirements.

Online measuring instrumentsThe installation of online measuringinstruments has become a reasonableoption due to the need to monitorwater quality results and rapidlyrecognise operational errors.Measurements results (for Fe,Mn,nitrite, pH,chlorine and turbidity) areavailable 24 hours a day, are document-ed and can be retrieved in the SCADAprogram and are monitored by thedispatcher continuously for possibleinterventions.

Challenges and future developmentsNetwork cleaning technologiesBeside the implementation of sched-uled network flushing, the improve-ment of water quality in the networkis also ensured by various networkcleaning technologies such as airscouring, soft sponge cleaning,chain-and-drag and RAK cleaning.Two new pipe cleaning technologieshave been tested on an experimentalbasis: dry ice-WOMA cleaning andjelly cleaning.

Inspection and review of fire hydrantsFire hydrants have to be examined andflushed every six months according tothe 28/2011. (IX.6.) Department ofHomeAffairs Decree.An advantage isthat stagnant pipe sections and deadends are regularly flushed,whichminimises sediments and secondarycontamination in the pipe network.

Installing non-return valvesBWW has been installing non-returnvalves since 1 July 2005 to preventwater flowing contrary to the properflow direction.Non-return valvesshould be installed in domestic potablewater pipes just before the valve afterthe water meter, according to theMSZ 22115:2002 Consumer waterconnections standard.

Geza Csornyei Genoveva Frank

Horizontal filteringwells (credit BWW)

SEPTEMBER2015 asset management - IWA Publishing€¦ · SEPTEMBER2015 ISSUE3 VOLUME11 PAPERS 3...

Documents

Transcript of SEPTEMBER2015 asset management - IWA Publishing€¦ · SEPTEMBER2015 ISSUE3 VOLUME11 PAPERS 3...