A Decision-Support Framework for Predicting Construction Delay Disputes Using Multi-Factor Analysis and Relative Importance Index

Abstract

The primary issue addressed in this study is the occurrence of disputes within construction projects. These disputes typically arise from a range of underlying factors, including site conditions, project delays, acceleration requirements, coordination challenges, value engineering decisions, conflicting objectives, the inherent complexity of work, quality-related concerns, and variations in tender specifications. The research examines time-related delay disputes by drawing on extensive literature, which reflects variations across different time periods, project types, and geographical contexts. To identify and categorise these causes, personal interviews were conducted, leading to the formulation of a comprehensive list of 110 dispute factors, which were incorporated into a structured questionnaire. The questionnaire consisted of two sections: the first gathered demographic and professional information of experts, while the second presented a series of categorised questions. It was then distributed among construction professionals, including owners, consultants, engineers, and contractors. The participants possessed varying levels of experience, ranging from five years to over thirty years in Egyptian construction projects. For analytical purposes, the Relative Importance Index (RII) was applied to determine the top fifteen causes of time delay disputes in the Egyptian construction sector. Based on the questionnaire findings, a predictive framework was developed and designated as the Time Delay Dispute Prediction Model (TDDPM). The model utilises inputs such as project-specific information, the probability associated with each of the top fifteen identified causes, and different weighting approaches, which may be equal, manually assigned, or automatically determined. The model generates outputs in the form of predicted delay dispute percentages, classification grades, and a concise set of recommendations aimed at mitigating delay-related disputes. A case study was also conducted to validate the findings, where its conditions were compared against the identified root causes from the questionnaire. The comparison demonstrated a complete alignment between the case study and the identified factors. Furthermore, when the case study was analysed using the developed model, the results showed a strong correspondence between the model’s output and the observed case conditions. The model achieved an estimated prediction accuracy of 83%, and additional recommendations were proposed to reduce the occurrence of time delay disputes.