Abstract
Due to increasing uncertainty and complexity, construction projects are frequently exposed to disruptions that compromise their performance, continuity, and long-term value creation. Although project resilience has attracted growing scholarly and practical attention, there remains a critical gap in systematically identifying the key resilience factors and understanding how resilience can be forecasted within construction project environments. This research addresses this gap by first identifying key project resilience factors from the construction context and then developing a predictive modelling framework to assess resilience capacity.
Given that project resilience is dynamic, non-linear, and difficult to forecast using conventional analytical methods, this study employs an Adaptive Neuro-Fuzzy Inference System (ANFIS) to model and predict resilience levels. ANFIS enables the integration of expert knowledge and empirical data while capturing complex interdependencies among resilience factors. By learning from observed project conditions and performance patterns, the model generates a resilience score that supports proactive risk mitigation and strategic decision-making. The findings reveal that leadership, cooperation and trust, and internal and external resources are the most influential drivers of project team resilience, with leadership emerging as the dominant predictor. Additionally, risk management practices and digital technology integration were found to reinforce adaptive capacity by enhancing anticipatory awareness and response efficiency.