Re-thinking material selection for large scale post ...€¦ · Giulia Celentano, 1988, MSc in...

1

Re-thinkingmaterialselectionforlargescalepost-disasterreconstruction

GiuliaCelentano,VincentHischier,GuillaumeHabert,EdwinZeaEscamilla,VerenaGöswein

ChairofSustainableconstruction,Swissfederalinstituteoftechnology,ETHZürich,Switzerland

ABSTRACT

Thelastfiftyyearshavewitnessedanunprecedentedgrowthonthehumanpopulation.Atthesametimethenumberand intensityofnaturaldisastershasbeenescalating,duebothtoclimatechangeandthegrowthofurban settlements in unsafe lands. This situation is unceasingly increasing the existing housing gap of theaffectedregions,alreadystrugglingtocopewiththeshelterdemand.Thepresentstudyaimstobetterunderstandthekeyparametersdrivingthedeliveryofhousingonpost-disasterreconstruction projects and to propose a set of indicators to help support the decision making process ofmaterialselection. Inthiswork15casesofpost-disasterreconstructionprojects in10 locationswerestudied.Two types of housing units were identified: core and transitional. They used four distinctive constructionmaterials:(i)bamboo,(ii)concrete,(iii)woodand,(iv)steel.TheseprojectswereexecutedbytheInternationalFederationofRedCross(IFRC)betweentheyears2004and2011.Fromtheirassessmentwehighlightedanewwaytoapproachmaterialselection,analysingtheoutputthroughdifferentkeyparameters:materialdemand,cost, delivery time andmanpower requirement. These key parameterswere considered at thehousing unitlevelandthenscaleduptothereconstructionproject.Finally,weproposeabenchmarkbasedontheidentifiedkeyparametersandintegratedwiththeshelterenvironmental impact,whichisabletoassesstheproblemonitscomplexity. Itallowstore-thinkthematerialselectionprocessandtoopenthediscussiononhowtocope

withthechallengesofupscalingreconstructionprojects.

KEYWORDS

Shelter,post-disasterreconstruction,upscaling,affordablehousing,appropriatetechnology.

AUTHOR’SBIOGRAPHY:

GiuliaCelentano,1988,MScinArchitecturefromthePolytechnicUniversityofMilan,Italy.Herfieldofexpertiseis appropriate technologies implementation in developing countries, with specific practical training in earthconstruction and slum upgrade. Before joining the Chair of Sustainable Construction of prof. Habert at ETHZurich,DepartmentofCivilEngineering,Giuliahasbeendeployedonfieldonvariousprojects inKenya,EgyptandCentralAmerica.

1Introduction

The last fifty years have witnessed an unprecedented growth of the human population, shifting from ruraltowards urban settlements (UN-Habitat 2016).Meanwhile, the damages provoked by natural disasters havebeenescalating,duebothtoclimatechangeand tothegrowthof informalsettlementsonunsafe lands (UN-Habitat2011).AccordingtotheUnitedNationsOfficeforDisasterRiskReduction,justbetween2000and2012theworldhassuffereddisastersaffecting2.9billionpeopleanddamagingthelocaleconomyby1.3trillionUS$(UNISDR2016)whileincreasingtheexistinglocalhousinggap.

The complexity of large scale post-disaster reconstruction demands to balance speed and affordability ofhousingdelivery(InternationalFederationoftheRedCross2012,fromnowonIFRC).Moreover,reconstructionprojectsaresusceptible tobottlenecks,as lackofsuitable resources (Russell2005;ZuoandWilkinson,2008).

2

Theuniquenessofeverypost-disastersituationisbindingsinceresourceavailabilityisintrinsicallyrelatedtothespecific post-disaster context (Chang et al. 2011). Still, diverse technological options might be available,accordingtomaterial,fundsandlogisticcapacityattheemergencymoment(IFRC2012).Aproperframeworkfor an appropriate reconstruction is thennecessary for assistinghumanitarianorganizations froma technicalpointofview(DaSilva2010).Thisneedhasalreadybeenidentified(JohnsonandDavidson2010)withnoclearsolutionproposed.

Thepresentstudyaimstobetterunderstandthekeyparametersdrivingthehousingdeliveryofpost-disasterreconstructionsandtoproposeasetofindicatorstosupportthedecisionmakingprocessofmaterialselection,inthewaytooptimizeresourcesandtimewhileprovidinganaffordableandsustainablelarge-scaleresponse.

2DataandMethods

2.1Data

15post-disaster reconstructionprojects in10 locationsworldwidearestudied.Twotypesofshelterunitsareidentified: coreandtransitional. Intheseprojects, fourmainconstructionmaterialsareused: (i)bamboo, (ii)concrete,(iii)woodand(iv)steel.TheseprojectswereexecutedworldwidebytheInternationalFederationofRedCrossandRedCrescentSocietiesbetweentheyears2004and2011,asshowninFigure1.

Figure1.Projectlocations

Locations:_Afghanistan,Bangladesh,BurkinaFaso,Haiti,Indonesia,Pakistan,Peru,Philippines,SriLanka,Vietnam.

The data collected for the research (location, year of execution, type of shelter, covered living space,constructiontime,numberofpeoplerequiredfortheconstruction,projectandmaterialcost)wereextractedfromthereportspublishedbytheFederation:“Transitionalshelters–Eightdesigns”and"Post-disastershelter:Tendesigns”(IFRC2013;IFRC2011).Theshelterperformanceproposedbypreviousstudies,combininghazardrisksateachlocationandexpectedshelterperformanceforanaturaldisaster(Zea,2015),wasconsideredasanadditional qualitative indicator. Furthermore, the environmental impact of the shelters was adopted by thesamesource.Thedatarelatedtomaterialamountandprocurementarepresentedintable1.

3

Table1.Sheltermaterialspecification

The different key parameters: (i) covered surface, (ii) expected lifespan, (iii) project cost, (iv) amount ofmaterials,and(v)constructiontimepermanpower,arepresentedinfigure2.Eachparameterisrepresentedbycircles,which dimensions are given by the parameter value for the shelter case. The shelter performance is

colourcodedhavingredasunsatisfyingandgreenasoptimalperformances.

Figure2.Sheltersdatavisualization

Fortheassessmentatsettlementscale,datawerecollectedfromtheofficialOperationsUpdatesrecordedbyIFRC on each executed project (IFRC 2016) in order to better understand the upscaling dynamics of post-disasterreconstructionprojects.

2.2Methodology

Inthisstudy,wefocusedontwoscales:theshelterunitandthesettlement.

At the shelter level, a new way to approach material selection is proposed. Cost, time of construction andamountofmaterialsarecalculatedforasinglefunctionalunitwhichhasbeendefinedasthecoveredareaoveran expected lifetime. The indicators are presented in table 2. Additionally, the location of thematerials hasbeen recordedand thesheltershavebeenclassifieddependingon theshareof importedmaterials required,ranging from100% imported, 50% imported, 2% imported and all localmaterials. Finally, the environmentalimpactof the shelter’s constructionhasbeencalculated.Themethod IMPACT2002+hasbeenused.DetailedcalculationcanbefoundinZeaandHabert(2015).

4

Indicator Unit

Projectcost/coveredsurface/lifespan

CHF/m2/months

Constructiontimeperteampermanpowerrequired/coveredsurface/lifespan

man-days/m2/months

Amountofmaterial/coveredsurface/lifespan kg/m2/months

Table2:shelterscaleindicatorsThe second part of the work focuses on the larger scale of the whole settlement. The objective here is toaddressthespeedandpaceofareconstructionprograminordertoseeifcommonpatterncouldbeidentified.Theindicatorsusedforthisphaserepresenttherelationbetweenthenumberofsheltersbuildovertimeandtheirconstructiontime.TheyarepresentedonTable3.

Indicator Unit

Ratioofsheltersbuildamongthetotal

%

Ratioofreconstructiontimesincewhenthereconstructionstartedamongthetotal

%

Table3:settlementscaleindicatorsandunits

Diverse factors significantly impede the reconstruction process in disaster-affected countries (Chang et al.2010),astheabilitytopoolresources,theleadtimeofprocurementandtransportationintothedisasterzone(Potangaroa2010),disruptionofaccesstoavailableresources(Green,Bates,andSmyth2007),andlimitedwaysof material procurement (Brunsdon et al., 1996; Oxfam Australia, 2007). At this scale the location of thematerials has therefore also been identified in order to understand if this could be a driving parameter forchangesinreconstructionpace.

3.Results

The shelters were first ranked according to the investments required for their construction (project cost,constructionspeed,lifespanandamountofmaterial),asrepresentedinFigure3a,3band3c.Theshelterswerenamedaccordingtotheirlocationandtypeofmaterial(locations:1Afghanistan,2Bangladesh,3BurkinaFaso,4Haiti,5Indonesia,6Pakistan,7Peru,8Philippines,9SriLanka,10Vietnam.Materials:B-Bamboo,C-Concrete,S-Steel,W-Wood).Acorrelationbetweeninvestmentandoutputisvisibleinalltheindicatorsstudied.Moreover,itispossibletoseethatanincreaseintheeconomicinvestmentusuallyleadstoabetterperformance(Figure3a)aswellasinthecaseofalongerconstructiontime(Figure3b).OnthecontraryanincreaseintheconstructionmaterialamountdoesnotleadtoaperformanceimplementationasshowninFigure3c.Itisimportanttonoticethatsomeexceptions(sheltersC9SriLankaconcrete,W5Indonesiawood,S5Indonesiasteel)achievegoodperformancescoresdespitealowinitialinvestmentintwooftheindicators.Nevertheless,noneofthecasestudiesshowedapositivecorrelationbetweenthethreeselectedindicators.

5

Figure3a.Projectcost

Figure3b.Constructiontime

Figure3c.Materialamount

Abenchmarksystemwasdevelopedtobetterunderstandthecorrelationbetweenthethreeindicators.Thisnotonlyallowedtoevaluatethesheltersunderthenewlensoftheprojectoptimization,butalsoprovidedanewreferenceforfutureprojects.

4.Discussion

TheresultsarepresentedonabenchmarkwithConstructiontimeperworkforce/coveredarea/lifespanontheXaxisandontheYaxisprojectcost/coveredarea/lifespan.TheindicatorAmountofmaterials/coveredsurface/lifespanwasrepresentedbythesizeofthecircles.Additionally,theperformancescorewasaddedincolour.ThefullassessmentispresentedinFigure4.

6

Figure4sheltersbenchmark:performanceandinvestmentShelterslocations:1Afghanistan,2Bangladesh,3BurkinaFaso,4Haiti,5Indonesia,6Pakistan,7Peru,8Philippines,9SriLanka,10Vietnam.

SheltersmaterialsBBamboo,CConcrete,SSteel,WWood.

The figure shows that an increase in the project cost leads to a better performance while affecting theconstruction speed negatively. Rapid constructions seem instead to be achieved with smaller economicinvestment,butattheexpenseoftheperformanceachieved.TheonlynotableexceptionisshelterC9(SriLankaconcrete), performing highly despite the low construction time requirement and a very limited economicinvestment. Its material amount as well is extremely low if compared to the others. High potential forimprovement isrecognizedforshelterB5(Indonesiabamboo)similartoC9(SriLankaconcrete) foreconomicinvestment,materialamountandevenmoreadvantageousintermsofconstructionspeed.Itsperformanceisinsteadbelowthestandardandwouldsorequiredesignimplementation.

From the same benchmark, more results can be observed by considering the material procurement of theshelters,displayedincolours,andtheenvironmental impactrelatedtothem,representedbythecirclesizeinFigure5.

7

Figure5.sheltersbenchmark:materialprocurementandenvironmentalimpactShelterslocations:1Afghanistan,2Bangladesh,3BurkinaFaso,4Haiti,5Indonesia,6Pakistan,7Peru,8Philippines,9SriLanka,10Vietnam.

SheltersmaterialsBBamboo,CConcrete,SSteel,WWood.

Shelterserectedmainlywithtransportedmaterialproducedlowresultsbothfromaneconomicinvestmentandconstructionspeedpointofview.Onthecontrary,sheltersmanufacturedwithlocallyprocuredmaterialsshowvariable results. In particular, the cases W7A Peru wood, B5 Indonesia bamboo and C9 Sri Lanka concretedisplayedaninterestingproportionbetweenprojectcostandconstructiontime.Thesheltersconstructedwithareduced inputof importedmaterials (below5%)haveverypositiveresultsandcanbeacknowledgedasgoodpractices. Despite the promising results of the partially transported shelters, design revision is for themrecommendedsotoimplementtheirperformance

Moreover, considerable challenges arise in post-disaster reconstruction situations as the context ischaracterized by a lack of access, logistical issues and inadequate human resources (Davidson et al. 2007;Ophiyandrietal.2010;IFRC2010),becomingsoanextremelycomplexproblemtodealwith(Bilauetal.2015).Forthisreason,theresearchfocusedthenonthecamplevel,lookingforpotentialupscalingeffectboostedbytechnologicalchoices. Thereconstructionprocessesrelatedtothe15casestudies is represented in figure6,wherethesheltersbuildamongthetotal,ontheYaxis,arecombinedwiththeelapsedprojecttime,ontheXaxis.

8

Figure6.campdevelopment:%elapsedprojectvs.%shelterscompleted

ThereconstructionsfollowedanS-curvetrend.Thisdevelopment,evenifdiverseamongthecases,seemedtobeaffectedbythematerialsourcesprocurement:sheltersbuiltwithmainly localmaterials infactdisplayedamoreaccelerateddevelopment.The benchmark was used to rank the case studies in a new comprehensive system and allowed for theidentification of strategies that better combined the listed above indicators. Despite an overall coherencebetweeninvestmentgrowthandperformanceachieved,suggestionsofdesignrevisionarealsorecommendedforthepartiallytransportedonesthatseemedpromisingforagoodbalancebetweentheindicatorsbutdidnotperformadequately. From thematerialprocurementpointof view, theuseof localmaterialsemergedasaninterestingelement, able toboost the speedof the constructionat the camp scale. This informationdidnotemerge at the unit scale: this shows that amultiscale approach formaterial selection is fundamental whendealingwithlargescalereconstructionprojects.

5Conclusion

Inthisstudy,15post-disasterprojectsdeliveredworldwidebytheInternationalFederationoftheRedCrossandRedCrescentSocietywereassessedwiththegoalof investigatinghowdifferentmaterialchoicesaffectedtheprojectdeliveryprocessaswellas the largescalereconstructionprocess.Thedevelopmentofnew indicatorsallowedinfirstplacetorankandevaluatetheexistingprojectsunderanewperspective.

Thiswork shows that an increase of the project cost usually leads to a better performing shelter, but oftennegatively affects its speed of construction. Moreover, the majority of the rapidly erected shelters wereconsidered as affordable but not satisfying from the proposed technical performance. It is also possible toconclude that thesevariations in construction time, costsorperformancearenotmaterial related.However,

9

theimportationofanykindofconstructionmaterialsfortheshelterusuallycontributestotheconstructionofhighperformingshelters,butheavilyimpactsbothontheprojectcostandtheconstructiontime.Moreover,theimportofaminimumamountofmaterialpershelterseemstohaveahighpotential.Thesetypeofshelterscanbe considered as most cost-effective from the studied sample. Nevertheless, their technical performanceinstead requires further improvements. Finally, the results showed that the use of locally procuredmaterialdoesnotdrivecostsnortimeofconstructionattheshelterlevel,butseemstobeabletohavealeverageeffectatthesettlementlevelwhereveryeffectiveincreaseinreconstructionprogramcouldbeachieved.

From these results it is possible to conclude that a multiscale approached can better support the decisionmakingprocesson largescale reconstruction.Opportunities forproject implementationandupscalingshouldbefurtherinvestigatedasafundamentalissuesincetheplanningphase.Finally,byrethinkingmaterialselectionfor large scale post disaster reconstruction, it is possible to maximize the use of the available resources,overcoming the challenges faced on post-disaster reconstruction project while providing a rapid sustainableresponse to the affected population. This study also serves as a reference for the evaluation of newprojectproposal,assuringtherelevanceoftheproposedindicatorsafutureassessments.

Acknowledgement

The authors would like to acknowledge the International Federation of the Red Cross and Red CrescentSocietiesfortheirsupportonthedevelopmentofthisresearch.Particularly,toCorinneTreherne, IFRCSeniorOfficerattheShelterandSettlementsdepartment,forhervaluableinsights.

References

Bilau, Abdulquadri Ade, EmlynWitt, and Irene Lill. 2015. “A Framework forManaging Post-Disaster HousingReconstruction.”ProcediaEconomicsandFinance21:313–20.

Brunsdon, D.R., Charleson, A.W., King, A.B., Middleton, D.A.,Sharpe, R.D. and Shephard, R.B. 1996. "Post-earthquake co-ordination of technical resources: the need for a unified approach." Bulletin of the NZNationalSocietyforEarthquakeEngineering,29(4):280–283.

Chang, Yan, SuzanneWilkinson, Regan Potangaroa, and Erica Seville. 2010. “Resourcing Challenges for Post-DisasterHousingReconstruction:AComparativeAnalysis.”BuildingResearch&Information38(3):247–64.

Davidson,ColinH., Cassidy Johnson,Gonzalo Lizarralde,NeseDikmen, andAlicia Sliwinski. 2007. “Truths andMyths about Community Participation in Post-Disaster Housing Projects.” Habitat International 31 (1):100–115.

Green,Rebekah,LisaK.Bates,andAndrewSmyth.2007.“ImpedimentstoRecoveryinNewOrleans’UpperandLowerNinthWard:OneYearafterHurricaneKatrina.”Disasters31(4):311–35.

InternationalFederationofRedCrossandRedCrescentSocieties.2010.“WorldDisasterReport2010:FocusonUrban Risk.” Geneva, Switzerland. http://www.ifrc.org/Global/Publications/disasters/WDR/WDR2010-full.pdf.

InternationalFederationofRedCrossandRedCrescentSocieties.2011. “Post-DisasterShelter:TenDesigns.”Geneva, Switzerland. http://www.sheltercasestudies.org/files/tshelter-8designs/10designs2013/2013-10-28-Post-disaster-shelter-ten-designs-IFRC-lores.pdf

International Federation of Red Cross and Red Crescent Societies. 2012. “Post Disaster Settlement PlanningGuidelines.” Geneva, Switzerland. https://www.sheltercluster.org/sites/default/files/docs/IFRC-Post%20disaster%20settlement%20planning%20guidelines.pdf.

International Federation of Red Cross and Red Crescent Societies. 2013. “Transitional Shelters: EightDesigns.”Geneva, Switzlerand. http://www.ifrc.org/PageFiles/95186/900300-Transitional%20Shelters-Eight%20designs-EN-LR.pdf.

10

InternationalFederationofRedCrossandRedCrescentSocieties.2016.“AppealsandReports-IFRC.”Geneva,Switzerland.http://www.ifrc.org/en/publications-and-reports/appeals/.

DaSilva,J.2010.“LessonsfromAceh:KeyConsiderationsinPost-DisasterReconstruction.”Arup/TheDisastersEmergencyCommittee.http://publications.arup.com/publications/l/lessons_from_aceh.

Johnson, Lizzarralde, andDavidson. 2010. “A SystemsViewof TemporaryHousing in Project in Post-DisasterReconstruction.”ConstructionManagementandEconomics24(4):367-378.

Ophiyandri, T., R. D. G. Amaratunga, and C. P. Pathirage. 2010. “Community Based Post Disaster HousingReconstruction: Indonesian Perspective.” Paper presented at the CIB 2010, 10th - 13th, University ofSalford.

OxfamAustralia,AustralianRedCrossandWorldVision.2007."EmergencyResponseSupplyChainAssessment",Australian Council for International Development, Melbourne.http://www.pacificdisaster.net/pdnadmin/data/original/ACFID_emergency_response.pdf

Potangaroa, Seville. 2010. “Resourcing Challenges for Post-Disaster Housing Reconstruction: A ComparativeAnalysis.” Building Research and Information: the international journal of research, development anddemonstration38(3):247-264.

Russell, Timothy Edward. 2005. “TheHumanitarian Relief Supply Chain: Analysis of the 2004 South East AsiaEarthquake and Tsunami.” Master Thesis, Massachusetts Institute of Technology.http://dspace.mit.edu/handle/1721.1/33352.

UN-Habitat.2011."CitiesandClimateChange:GlobalReportonHumanSettlements,2011."Washington,DC:Earthscan. http://unhabitat.org/wpdm-package/cities-and-climate-change-global-report-on-human-settlements-2011/?wpdmdl=110715

UN-Habitat.2016.“WorldCitiesReport2016.”http://unhabitat.org/books/world-cities-report/.

UNISDR. 2016. “Disaster Statistics - UNISDR.” AccessedMay 31. https://www.unisdr.org/we/inform/disaster-statistics.

Zea, E and G Habert. 2015. "Global or local construction materials for post-disaster reconstruction?Sustainabilityassessmentoftwentypost-disastershelterdesigns."BuildingandEnvironment92:692-702.

Zuo, K. and S.Wilkinson. 2008. “Supply Chain andMaterial Procurement for Post-disaster Construction: TheBoxing Day Tsunami Reconstruction Experience in Aceh, Indonesia.” Paper presented at the CIB W89International Conference on Building Education and Research, Heritance Kandalama, Sri Lanka, 11–15February.