BIM-based solution to enhance the performance of public-private partnership construction projects using copula bayesian network

Public-private partnerships (PPPs) are efficient methods for constructing infrastructures that create many opportunities, specifically in financial crises. However, the cost/time overruns have raised many problems in these projects that show their feeble performance. That said, optimizing the key performance indicators (KPIs) have been neglected as a crucial part of performance enhancement. Therefore, analyzing and proposing a Building Information Modeling (BIM) based model to improve KPIs condition have been chosen as the main objective of this research. In this research, the Copula Bayesian Network (CBN) has been selected as a robust statistical technique to determine causal structure ability and estimate the variables' impact on each other. Unlike similar research, this study has used Python programming, Shapley Additive exPlanations (SHAP), and Extreme Gradient Boosting (XGBoost) to comprehensively analyze CBN and provide quantitative analysis to assess the BIM impact on PPPs. This research has estimated the influence of BIM on PPPs' performance and extracted the most prominent KPIs in BIM enabling conditions, including the feasibility study, finance/cost performance, appropriate financing option, risk identification, allocation, sharing, and transfer, finance infrastructure, and compliance to the legal and regulatory framework. This study proved that BIM could improve PPPs performance by 28.9% on average. The outcome of this research helps the private sector to gain a comprehensive perspective on implementing BIM and its effects on PPP projects and gives an insight into BIM's importance and efficiency in solving critical PPP issues for future researchers.

Automated summary

Public-private partnerships (PPPs) are efficient methods for constructing infrastructures, but cost and time overruns have affected their performance. This research proposes a Building Information Modeling (BIM) based model to improve key performance indicators (KPIs) and analyzes the impact of BIM on PPPs using statistical techniques and programming.