Comparative Analysis of Cost Estimating Methods in Construction Projects

Thesis Abstract

This study investigates the comparative performance of traditional, parametric, and digital cost estimating methods across diverse construction projects to address persistent discrepancies between estimated and actual project costs. The problem addressed is the inconsistent accuracy and reliability of cost estimates that undermine project viability, stress budgets, and erode stakeholder confidence in procurement processes. The aim is to determine how estimation method choice influences cost precision, risk exposure, and decision-making efficiency, with specific objectives to (i) quantify estimation accuracy across methods using a standardized metric, (ii) examine how project type, scale, and procurement route moderate method performance, (iii) identify cost drivers most sensitive to estimation technique, (iv) evaluate practitioner perceptions of ease of use, data requirements, and integration with digital BIM workflows, and (v) develop a normative framework for selecting estimation methods aligned with project characteristics. The research adopts a mixed-methods design, combining quantitative analysis of a cross-sectional sample of 120 completed projects drawn from public, private, and public–private partnerships over a five-year window, and qualitative insights from 24 semi-structured interviews with senior quantity surveyors and estimators. The quantitative component employs absolute and percentage error metrics, mean absolute percentage error (MAPE), and root mean squared error (RMSE) to compare traditional, parametric (e.g., unit rate and cost-indexed models), and digitally enhanced estimates generated from BIM-enabled data linked to historical price libraries. Regression analysis and ANOVA are used to test the relationship between estimation errors and project attributes (size, complexity, procurement route, and contract type). The qualitative component applies thematic analysis to interview transcripts, guided by the Technology–Organization–Environment (TOE) framework and the Lean Construction theory to contextualize adoption barriers and workflow integration. Data collection instruments include a standardized estimation dataset capturing initial estimates, revised estimates, final actual costs, quantity takeoffs, and time-to-estimate across cases; BIM-enabled data extraction templates; and interview guides designed to elicit practitioner experiences, challenges, and perceived reliability of each method. Validity and reliability are ensured through triangulation across data sources, pilot testing of instruments, inter-coder reliability checks for qualitative coding (Cohen’s kappa > 0.70), and convergence checks in regression diagnostics (variance inflation factors below 5, residual analyses confirming homoscedasticity). Expected findings indicate that digitally augmented cost estimates, when integrated with robust historical databases and BIM models, produce substantially lower MAPE and RMSE than traditional and standalone parametric estimates, with reductions in cost overruns by an estimated 8–15% in large-scale projects and 4–10% in medium-scale projects. The performance advantage is anticipated to be moderated by project type (institutional vs. residential), procurement route (design-bid-build vs. design-build), and data quality of historical records. Practitioner interviews are likely to reveal that digital methods improve transparency and early risk signaling but require greater upfront data governance and cross-disciplinary collaboration. The study contributes to knowledge by providing empirical cross-sectional evidence on the relative accuracy of cost estimation methods in current practice, clarifying how BIM-enabled data ecosystems interact with estimation techniques, and offering a decision-support framework for method selection grounded in project characteristics and organizational readiness. The main conclusion posits that digitally integrated cost estimation methods, properly governed and aligned with project context, outperform traditional and purely parametric approaches in both accuracy and actionable insight. Recommendations include (i) investing in standardized data governance and BIM interoperability, (ii) developing a tiered estimation protocol that prescribes method selection by project scale and procurement route, (iii) establishing organizational training to enhance proficiency with digital estimation tools, and (iv) expanding the dataset of historical project costs to continuously refine estimation models. The study advances knowledge on the practical efficacy of estimation methodologies and informs procurement policies, contract strategy, and risk management practices within the construction industry.