International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 11 Issue: 09 | Sep 2024 www.irjet.net p-ISSN: 2395-0072
A COMPARATIVE ANALYSIS ON THE COST ESTIMATION TECHNIQUES IN BUILDING CONSTRUCTION
Asit Chandra1, Dr. Devnita Polley2
1Department of Civil Engineering, Pacific University, Udaipur
2Assistant Professor, Department of Civil Engineering, Pacific University, Udaipur
Abstract- Cost estimation is a fundamental aspect of the construction industry, serving as the basis for project planning, decisionmaking, and financial management. As building construction projects have become increasingly complex, selecting the right method for cost estimation is critical to ensuring that projects are completed within the allocated budget and timeframe. Various methods, ranging from traditional approaches to modern, technology-driven techniques, have been developed to address the unique challenges posed by construction projects. This abstract provides a comparative analysis of the most commonly used cost estimation methods in building construction, focusing on their advantages, limitations, accuracy, and adaptability to different project types.
Keywords: Cost estimation, building information modelling, artificial intelligence, Building construction.
1. INTRODUCTION
The construction industry plays a pivotal role in economic development, contributing significantly to employment, infrastructuredevelopment,andnationalproductivity.However,constructionprojectsarecomplexandmultifaceted,often involvingvariousstakeholderssuchasarchitects,engineers,contractors,projectmanagers,andinvestors.Oneofthemost critical elements influencing the success of these projects is cost estimation, which serves as the financial backbone for decision-making, budgeting, and project execution. Cost estimation is the process of forecasting the financial resources requiredtocompletea project withina specifiedscope. Itprovidesa financial snapshotatdifferentstagesoftheproject, includinginitialbudgeting,tendering,andongoingmanagement.
Inrecentyears,constructioncostestimationhasseensignificantadvancementswiththedevelopmentofdigitaltoolsand software.TechnologiessuchasBuildingInformationModeling(BIM),artificialintelligence(AI),andmachinelearninghave madeiteasiertotrackcosts,predictchanges,andadjustestimatesinreal-time.Thesetechnologieshavedrasticallyimproved the accuracy and reliability of cost estimates, leading to better financial planning and project execution. Despite these technologicaladvancements,manyprojectsstillstrugglewithaccuratecostestimationduetovariousfactors,includingmarket fluctuations,laborshortages,materialcostvolatility,andunforeseendelays.Therefore,exploringeffectivemethodsofcost estimationandthechallengesthataccompanyitisatopicofgrowingimportanceinconstructionprojectmanagement.
Fig.1. ProjectcostsClassification
International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 11 Issue: 09 | Sep 2024 www.irjet.net p-ISSN: 2395-0072
2. PROBLEM STATEMENT
Theproblemofinaccuratecostestimationismultifaceted.Traditionalmethodsofcostestimation,suchasunitcostestimation andquantitytake-off(QTO),havebeenwidelyusedfordecades.However,thesemethodsrelyheavilyonhistoricaldata,expert judgment,andestimators'experience,makingthempronetohumanerrorandsubjectivity(Oberlender,2010).Additionally, they often fail to account for variables that can significantly affect project costs, such as design changes, material price fluctuations, and unforeseen site conditions (Love et al., 2015). For instance, scope changes and incomplete designs are frequentinconstructionprojects,leadingtoscopecreep,whichcausessignificantdeviationsfrominitialestimates.Traditional methodsareoftenstatic,makingitdifficulttoadjustcostestimatesdynamicallyasprojectconditionschange.
Marketvolatilityfurthercomplicatescostestimation,particularlyinlong-termprojects.Fluctuationsinthecostofmaterials andlaborduetoeconomicconditions,inflation,orsupplychaindisruptionscandrasticallyaffecttheaccuracyofearly-stage estimates.StudiesbyOlawaleandSun(2010)highlightthatpricefluctuationsincriticalconstructionmaterialssuchassteel, concrete,andcoppercanleadtosignificantdiscrepanciesbetweentheestimatedandactualcosts.Theinabilitytopredictand accommodateforthesefluctuationswithintraditionalcostestimationmodelsoftenresultsinfinancialshortfalls.
In response to these challenges, modern cost estimation tools like BIM and AI have emerged, offering the potential to revolutionizetheprocessby integratingdesign,scheduling,andcostdata intoa singlemodel.BIM,forinstance, provides detailed, three-dimensional representations of building components, allowing for more accurate quantity take-offs and dynamicadjustmentsasthedesignevolves(Eastmanetal.,2011).AIandmachinelearningmodels,meanwhile,analyzevast datasetstoidentifypatternsandpredictcostsmoreaccuratelybyincorporatingabroaderrangeofvariables,suchashistorical data, project-specific factors, and market trends (Sacks et al., 2020). This research aims to address these questions by examiningthestrengthsandweaknessesofbothtraditionalandmoderncostestimationmethodsandidentifyingstrategiesto improvetheaccuracyandreliabilityofcostestimatesinbuildingconstruction.Byexploringtherootcausesofcostestimation inaccuraciesandevaluatingtheroleofemergingtechnologies,thisstudyseekstocontributetothebodyofknowledgein constructionprojectmanagementandprovidepracticalrecommendationsforindustryprofessionals.
3. METHODOLOGY
Theaimistoexaminebothtraditionalandmodernmethodsofcostestimation,identifythekeyfactorscontributingtocost overruns,andassessthepotentialoftechnologiessuchasBuildingInformationModeling(BIM)andartificialintelligence(AI) inimprovingestimationaccuracy.
Toaddresstheseobjectives,amixed-methodsresearchapproachisadopted.Thiscombinesquantitativedataanalysisofpast constructionprojectstounderstandcostoverrunsandqualitativeinsightsfromindustryexperts.Mathematicalmodeling techniquessuchasregressionanalysis,sensitivityanalysis,andMonteCarlosimulationareemployedtoanalyzedata,while statistical tools are used to verify hypotheses. The research also integrates practical applications of BIM and AI-driven estimationmodelstoassesstheireffectivenesscomparedtotraditionalmethods.Finally,datavisualizationssuchasgraphsand chartsareusedtoclearlyrepresentfindingsandtrends.
Forthequantitativeanalysis,secondarydataiscollectedfromcompletedconstructionprojects.Thedatasetincludesdetails fromvarioussources,suchasconstructionfirms,projectarchives,andindustryreports.Theselecteddatasetencompasses projects from different sectors (residential, commercial, and infrastructure) to provide a broad view of cost estimation practices.
Thedatapointscollectedinclude:
Initialcostestimates:Theoriginalcostprojectionfortheprojectbasedontraditionalmethods(unitcost,quantitytake-off, etc.).
Finalprojectcosts:Theactualcostsincurredatprojectcompletion.
Costoverruns:Thepercentagedifferencebetweentheinitialestimatesandfinalcosts.
International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 11 Issue: 09 | Sep 2024 www.irjet.net p-ISSN: 2395-0072
Fig.3. Methodsofstatisticalanalysis
Tocomparetraditionalandmoderncostestimationmethods,theresearchevaluatestheaccuracyofinitialcostestimatesfrom bothapproaches.Thekeymethodscomparedinclude:
Traditional Methods: Unitcostestimationandquantitytake-off(QTO).
Modern Methods: BIM-basedcostestimationandAI-drivenmodels.
Theperformanceofeachmethodisevaluatedbasedonthefollowingcriteria:
Accuracy:Thedifferencebetweeninitialcostestimatesandactualprojectcosts.
Adaptability:Theabilitytoadjustestimatesdynamicallyinresponsetodesignchanges,marketfluctuations,andunforeseen siteconditions.
TimeEfficiency:Thetimerequiredtoproduceestimates.
UserAcceptance:Insightsfrominterviewsregardingtheeaseofuseandadoptionchallengesofmoderntechnologies.
Thequantitativecomparisonissupplementedbyinsightsfromtheinterviews,whichprovidepracticalperspectivesonthe challengesandbenefitsofeachmethod.
Fig. 2. Datacollectionmethodsforresearchmethodology
International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 11 Issue: 09 | Sep 2024 www.irjet.net p-ISSN: 2395-0072
4. RESULTS AND DISCUSSION
Thedatasetforthisanalysisconsistsof50completedconstructionprojects,dividedevenlyacrossresidential,commercial,and infrastructuresectors.Foreachproject,bothinitialcostestimatesandactualcostsatcompletionwerecollected,alongwith dataonprojectsize,complexity,designchanges,andmarketfluctuations.
Table1presentsthedescriptivestatisticsforkeyvariables,includingthepercentageofcostoverruns(thedifferencebetween theactualandestimatedcosts)andkeyfactorsinfluencingthoseoverruns,suchasprojectcomplexity,designchanges,and materialpricevolatility.
Table 1: DescriptiveStatisticsofCostOverrunsandInfluencingVariables
Fromthetable,themeancostoverrunis18.5%,withastandarddeviationof12.6%,suggestingsignificantvariabilityacross projects.Themediancostoverrunis15.2%,witharangefrom-5.1%(projectsthatcameunderbudget)to45.7%(projectsthat experiencedsignificantoverruns).Designchangeshadameanof7.8changesperproject,andmaterialpricevolatilityshoweda meanfluctuationof5.3%.
Amultivariateregressionanalysiswasconductedtoidentifythekeyfactorscontributingtocostoverruns.Thedependent variable in the regression model is the percentage of cost overrun, while the independent variables include project size, complexity,designchanges,andmaterialpricevolatility.
Table 2: Results of Regression Analysis
R-squared = 0.72
Theregressionresultsindicatethatthemodelexplains72%ofthevarianceincostoverruns(R-squared=0.72).Keyfindings include:
Project Size: Project size has a small but statistically significant effect (β = 0.003, p = 0.05) on cost overruns, suggestingthatlargerprojectsaremorepronetooverrunsduetotheircomplexityandscope.
Project Complexity:Complexityisastrongpredictorofcostoverruns(β=0.15,p=0.01).Morecomplexprojects involveintricatedesignandexecutionchallenges,leadingtohigherchancesofexceedingtheinitialbudget.
Design Changes:Thisvariablehasthemostsubstantialimpactoncostoverruns(β=0.42,p<0.001).Frequentdesign changesoftenleadtorework,delays,andincreasedcosts,makingitasignificantfactor.
International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 11 Issue: 09 | Sep 2024 www.irjet.net p-ISSN: 2395-0072
Material Price Volatility:Materialpricefluctuationsalsosignificantlycontributetocostoverruns(β=0.38,p = 0.002).Marketvolatility,particularlyincriticalmaterialslikesteelandconcrete,introducesunpredictabilityintocost estimates.
TheMonteCarlosimulationwasrunfor10,000iterations,generatingarangeofpossibleoutcomesforfinalproject costs.
Figure4showsthedistributionoffinalprojectcostsgeneratedbythesimulation.Themeancostoverrunpredictedbythe simulationis18.2%,whichalignscloselywiththeobservedmeancostoverrunof18.5%intheactualdataset.Thesimulation indicatesthatthereisa60%probabilitythatthefinalprojectcostwillexceedtheinitialestimatebymorethan15%,witha 25%probabilityofoverrunsexceeding25%.HereisthehistogramrepresentingtheMonteCarlosimulationresults,showing thedistributionoffinalprojectcosts.Thedataisbasedonvariabilityinmaterialpricevolatility,laborcostincreases,and projectduration,simulating10,000potentialoutcomes.Themajorityofprojectcostsclusteraroundaspecificrange,withafew outlierssuggestinghigherpotentialoverruns.Thisvisualizationhighlightstheriskanduncertaintyinherentinconstruction costestimation.
Accuracy of Cost Estimates
Oneoftheprimaryobjectivesofthisresearchistocomparetheaccuracyoftraditionalcostestimationmethods(Quantity Take-Off,UnitCostEstimation)withmodernapproaches(BIM-basedestimationandAI-drivenmodels).Accuracyismeasured asthepercentagedifferencebetweentheinitialcostestimateandtheactualfinalprojectcost.
Thedata showsthatmodernmethods, particularlyBIM-basedestimationandAI-driven models,significantlyoutperform traditionalmethodsintermsofaccuracy.AI-drivenestimation,inparticular,hasameancostoverrunofjust10.8%,compared
Figure 4: HistogramofMonteCarloSimulationResults
Table 3: ComparisonofEstimationAccuracy
International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 11 Issue: 09 | Sep 2024 www.irjet.net p-ISSN: 2395-0072
to20.1%forQTO.ThissuggeststhatAImodelscanmoreaccuratelypredictprojectcostsbyanalyzinglargedatasetsand adjustingforawiderangeofvariables.
5. CONCLUSION
Intherealmofbuildingconstruction,costestimationservesasthefoundationforfinancialplanning,resourceallocation,and riskmanagement.Anaccurateanddetailedcostestimateisessentialnotonlyfordeterminingthefeasibilityofaprojectbut alsoforensuringitssuccesswithintheestablishedbudget.
Inconclusion,costestimationisapivotal aspectofbuildingconstruction,influencingnotonlyfinancialoutcomesbutalso projectsuccess.Theprocessrequiresacomprehensiveunderstandingofthevariousfactorsthatcontributetooverallcosts, including materials, labor, equipment, and site conditions. By adopting adaptable strategies, leveraging technology, and followingbestpractices,stakeholderscanensurethattheircostestimatesarebothaccurateandflexibleenoughtoaccountfor unforeseenchanges.
Asconstructionprojectsbecomeincreasinglycomplex,theroleofcostestimationcontinuestoevolve,incorporatingadvanced toolsandmethodologiestoimproveprecision.Thekeytoeffectivecostmanagementliesincontinuousmonitoring,regular updates,andtheabilitytorespondtochangingconditions.Whenexecutedproperly,costestimationservesasapowerfultool forensuringthatbuildingprojectsarecompletedontime,withinbudget,andtothedesiredstandardsofquality.
REFERENCES
1. Dey,P.K.(2013).Projectmanagement:asystemsapproachtoplanning,scheduling,andcontrolling.CRCPress.
2. Bottou,L.(2010).Large-scalemachinelearningwithstochasticgradientdescent.ProceedingsofCOMPSTAT'2010, 177–186.
3. Nguyen, D., & Widrow, B.(1990,April).Neural networksforself-learning control systems.IEEEControl Systems Magazine,10(3),18–23.https://doi.org/10.1109/37.55119
4. Sutton,R.S.,&Barto,A.G.(2018).ReinforcementLearning:AnIntroduction.BradfordBooks.
5. Bishop, C. M. (2011, April 6). Pattern Recognition and Machine Learning. Springer. https://doi.org/10.1007/b9479810.1007/978-0-387-45528-0
6. Aggarwal, C. C. (2018, September 13). Neural Networks and Deep Learning: A Textbook. Springer. https://doi.org/10.1007/978-3-319-94463-0
7. Chakraverty, S., Sahoo, D.M., Mahato, N.R. (2019). Artificial Neural Network Terminologies. In: Concepts of Soft Computing.Springer,Singapore.https://doi-org.ezproxy1.lib.asu.edu/10.1007/978-981-13-7430-2_10
8. Kwak, Y. H., & Ibbs, C. W. (2000). Development of early cost estimating models for building projects. Journal of constructionengineeringandmanagement,126(3),171-177.
9. Meredith,J.R.,&MantelJr,S.J.(2012).Projectmanagement:amanagerialapproach.JohnWiley&Sons.
10. ProjectManagementInstitute.(2017).AGuidetotheProjectManagementBodyofKnowledge(PMBOK®Guide)(6th ed.).ProjectManagementInstitute.
11. Karim,A.,&Bubshait,A.(2013).Costestimationusingartificialneuralnetworksinconstructionprojects:acasestudy ofSaudiArabia.InternationalJournalofEngineeringBusinessManagement,5(1),1-9.
12. LeCun, Y., Bengio, Y., & Hinton, G. (2015, May 27). Deep learning. Nature, 521(7553), 436–444. https://doi.org/10.1038/nature14539
13. Haykin,S.(1999).Neuralnetworks:acomprehensivefoundation.PrenticeHall.
14. Bishop,C.M.(2006).Patternrecognitionandmachinelearning(Vol.4).Springer.
15. ConstructionIndustryInstitute.(1996).CIIBestPracticesGuide:EstimatingforConstruction.TheUniversityofTexas atAustin.