Empowering water consumers through smart metering:

evidence from a field study in a residential suburb of Montpellier (south of France)

Marielle Montginoul et Arnaud Vestier

SH2016, 22-25 August 2016, Monte Verità – Switzerland

An increasing water scarcity (less supply – more demand)

INTRODUCTION

Smart metering: a technology with many advantages

Advantages Forusers*/forwatermanagers**

Highmeterreadingfrequency

Awaterleaks’detection(*+**)Analert toinformusersincaseofsuspicionofwaterleaks(*)Todesignwaterpricingtakingintoaccountwaterscarcityandothermanagerialconstraints(**)

Betterwaterconsumptionmonitoring

Aprecise knowledgeofindividualwaterconsumption(*+**)Benchmarking(*)Configuringalerts(SMS orEmail)toinformuserswhenwaterconsumptionexceedsapre-definedthreshold(*)

AutomatedremotemeterreadingAnabsenceofdisturbances(*)Productivitygains(**)A billonrealdata,anopportunitytoincreasebillingfrequency(*)

↔Aneedtosavewater

… but a low subscription rate (even if free of charges)

INTRODUCTION

2%ofthe23,000watersubscribers(February2015)

Why a such gap? Looking forward factors determining the adoption:1. Information2. Intends to take action (the theory of planned behavior, Ajzen, 1991)

Smart metering: a technology with many advantages …

Being informedFactors determining smart metering adoption

Who says What to Whom in Which channel with What effect (Lasswell - 1902-1978)

Mediaofcommunication CODAH Metz Mulhouse SEDIF SMGC

Directly: towatersubscriberorwateruserPersonalized

Mail x x x x x

Flyer x x x x x

TargetedNew subscription x

Meter inspectors x x

Neighborhood council x x x

Mayor communication x 1district

GeneralizedWeb page x x x x

Newspaper article x x

Municipalnewsletter x x x x x

Indirectly:through localauthoritiesormayorsMail x

Oral communication x

Being « well meaning »(Plannedbehavior– Ajzen,1991)

Factors determining smart metering adoption

The attitude toward water and new technologyEnrolment

Subjective norm or perceived social pressureSmartmeterasasocialobjectconsideredasrewarding

Perceived behavioral controlAbilitytocontrolthenewtechnologyAbilitytotranslateinformationintoaction- workingknowledge(Darby,2010)(e.g.understandingiswaterlevelisnormalornot)

Factors determining smart metering adoption

Information

Being informedIndividualinformation:facetofaceand/orbyletter(signedbytheMayorinchargeofwater)DirectlydonebyMontpellierMétropole staff(andnotbythewateroperator:Veolia)Aleafletexplainingthesmartmeteringservice

A residential areaThe case study

261householdswithsmartmetersofMontpellierMétropole (2015)Anaturalfieldexperiment

Being « well meaning »Testedthroughaface-to-facesurvey

A residential areaThe case study

261householdswithsmartmetersofMontpellierMétropole (2015)

• ToanalyzerelationshipsbetweenhouseholdsandwaterserviceEx:waterpriceperception,computerskills,paymentofwaterbillthroughthewebsite

• ToidentifysocialrepresentationsofwaterandsmartmetersTogive3wordsorphrasesreferringtoitems« water »/« intelligentmetering »,toorderthemofimportance,andtoprioritizethemonascalefrom-3to+3

• TocollecthouseholdcharacteristicsEx:size,housingtype,socio-professionalcategory…

The subjectsThe case study

Targetedhouseholds

Surveyedhouseholds

Media of communication

House (even-numbered) 123 35 Face-to-face (for surveyed pop., letter to others)

House (odd-numbered) 68 25Letter M3MApartment (social building) 16 6

Apartment (other) 54 11Total 261 77

Average water consumption (liters/day/household) (jan-june 2015)

Targeted households Surveyed households

Apartment House Total Apartment House Total

- Mean

- Median

211

187

403

304

354

278

247

168

402

324

374

318

Observations (number) 54 157 211 11 50 61

Surveyedhouseholds’characteristics2,7peopleperhouseholdAveragedateofbirthofthesurveyedrespondent:1961Retiredpeople:31%Profession:managers(47%),employees(31%)Renters:27%Houses:60%withswimmingpool,12%withrainwatercollector,33%withborehole

Social representationThe case study

Throughprioritizedevocations’method

Levelofinterest

High (rank<2) Low(rank≥2)

Frequencyofoccurrence

High(≥10%)Thecore:quantitative andqualitativecentralityarea

Firstperimeter

Low(<10%) Contrasting elements Secondperimeter

Motsassociésàl’eauNombredecitation

Fréquencedecitation

Rangmoyendecitation

Echellemoyennedu

motZonedunoyaudelareprésentation lavie 27 12% 1,11 2,9lavage 27 12% 1,81 2,7Zonedepremièrepériphérie plaisirdétente 22 10% 2,45 2,1Elémentscontrastésdelareprésentation boissonetalimentation 20 9% 1,75 2,5nécessité 16 7% 1,69 2,7pureté 11 5% 2,00 2,4rareetprécieux 5 2% 2,00 1,4soif 4 2% 1,50 0,8eaupotable 3 1% 2,00 2,3qualité 3 1% 2,00 2,7liquide 2 1% 1,50 0,0manqued'eau 7 3% 1,86 -2,6inégalitédepartage 2 1% 2,00 -2,5Zonedesecondepériphérie lanature 13 6% 2,46 1,8fraîcheur 12 5% 2,50 1,8économiserl'eau 10 4% 2,20 2,0environnement 7 3% 2,57 1,9abondancedébitaccèsàl'eauconsommation 6 3% 2,17 1,0arrosage 5 2% 2,40 0,8relaxationpurification 4 2% 2,25 2,5banal 1 0% 3,00 0,0coûtdel'eau 10 4% 2,30 -1,8mauvaisequalitédel'eau(calcaire,goût,odeur) 4 2% 2,50 -0,5inondationpollution 2 1% 2,50 -3,0gaspillage 2 1% 2,50 -1,5demoinsenmoinsnaturel 1 0% 3,00 -2,0

Togive3wordsorphrasesreferringtoitems« water »/« intelligentmetering »Toorderthemofimportance

Toprioritizethemonascalefrom-3to+3

Water social representationThe case study

Water-relatedwordsFrequency

ofoccurence

Citationaverageranking

Wordaveragescale

Corearea Life 12% 1,11 2,9Washing 12% 1,81 2,7Firstperimeter Pleasureand/orrelaxation 10% 2,45 2,1Contrastingelements Drinkandfood 9% 1,75 2,5Aneed 7% 1,69 2,7Purity 5% 2,00 2,4Scarceandprecious 2% 2,00 1,4Thirst 2% 1,50 0,8Drinkingwater 1% 2,00 2,3Quality 1% 2,00 2,7Liquid 1% 1,50 0,0Watershortage 3% 1,86 -2,6Inequityofsharing 1% 2,00 -2,5

Intelligent metering social representationThe case study

Intelligentmetering-relatedwords

Frequencyof

occurence

Citationaverageranking

Wordaveragescale

Corearea Useful 18% 1,49 1,94Improvedconsumptionmonitoring 11% 1,77 2,05Simple 10% 1,95 1,58Contrastingelements Realtimealert 7% 2,00 2,15Leakageoroverconsumptionalert 6% 1,73 2,45Modernity 6% 2,00 2,00Watersaving 4% 1,57 2,00Abillbasedonrealconsumption 2% 2,00 3,00Effective 1% 1,50 2,00Indifference 1% 1,50 2,00Environmentallyfriendly 1% 2,00 1,50Havetobepaidbywatermanager 1% 2,00 3,00Assistance 1% 2,00 3,00Billmanagement 1% 1,00 2,00Unemployment 4% 1,75 -3,00Unuseful 3% 1,40 -1,60Manipulation 1% 2,00 -3,00AneedtobeconnectedonInternet 1% 1,00 -3,00Privatisation 1% 2,00 -3,00

Still a low subscription rate (2 months after being informed)

… but not because of a negative representation: aneedtoexplorethelaststepthatonewhichgoesfromintentiontoaction

Ø Aneedtobeperceivedasanimportantissue:WaterbilllevelAFrenchwaterlawwhichprotectswaterusersagainstexceptionalleakages(debtcancellationwhenplumberinvoiceprovingleakagerepair)

Ø Enhancedbilling(e.g.awaterpricinglinkedtointelligentmetering):Likeelectricitywheretime-of-daypricing(Ehrhardt-Martinezetal.,2010)

Ø Aneedofafeedback(Schleich etal.,2013)orofanautomatedprogramwhichinformswateruserwhenheoverconsumes(Lafaye etal.,2013)

Conclusion

Media of communication Number of households Before being informed 2 Letter communication only 6 Face-to-face communication 4 Total 12

LowincomeHighwaterusers

Information

Intention

Behavior

Conclusion

Further researches: Ø Exploringadatasetof23,000watermeters(numberofobservationsfrom2012toJune2016:

44,000,000):detectionoftrendsandbreaks,explorationofpotentialcorrelationwiththesubscriptionornottointelligentmeteringserviceortypeofwateruser…

Montpellier

Many thanks for your attention

marielle.montginoul@irstea.fr