Machine Learning-Based Assessment of Construction Project Delays

Delays in construction projects lead to significant negative consequences, including cost increases, resource wastage, and failure to complete the project on time, posing serious risks to both investors and project stakeholders. Accurately identifying the causes of delays and predicting these adverse outcomes in advance are critical for minimizing risks in project management processes and enabling more efficient use of resources. In this context, the use of machine learning techniques for analyzing and predicting delay factors in construction projects has gained increasing importance. This study is based on a publicly available dataset collected through a survey conducted among stakeholders involved in the investment and construction process, encompassing 32 different delay factors. Each factor in the dataset was evaluated by participants using a 5-point Likert scale. To determine the importance levels of delay causes and predict delay risks in projects, various machine learning algorithms, including Support Vector Machines (SVM), Artificial Neural Networks (ANN), and K-Nearest Neighbors (KNN), were applied. Comparative analyses revealed that the factors with the most significant impact on delays could be successfully predicted. The obtained results contribute to the development of strategic decision-making mechanisms aimed at improving project planning and management processes in the construction sector. Furthermore, this study provides an important contribution to the literature regarding the applicability of machine learning techniques in construction management. For future work, it is recommended to conduct comprehensive comparisons of different machine learning algorithms to enhance model performance, update the dataset with broader and more diverse samples, and develop dynamic time series-based prediction models. These approaches will enable more effective and reliable management of construction projects.