A Proposed Model for Predicting Cost and Schedule Overruns Caused by Risks in Bridge Construction Projects
Keywords:
Risk management, bridge projects, cost & schedule delay, artificial neural networks (ANN), sensitivity analysis.Abstract
Bridge projects in Syria face complex challenges due to unforeseen risks that impact costs, schedules, and execution quality, making risk management essential for ensuring planned project delivery.
This study proposes an advanced predictive model using Artificial Neural Networks (ANN) to analyze and evaluate the impact of various risks on bridge projects in Syria.
A comprehensive methodology was adopted, including a field study of 45 completed bridge projects and a specialized questionnaire distributed to experts to identify and classify key risk factors. Based on this data, two models were developed using Multilayer Perceptron (MLP) neural networks: one for predicting cost overruns and the other for schedule delays.
The models demonstrated high predictive accuracy, with R² values of 0.95 for cost and 0.94 for schedule delays. Sensitivity analysis identified the most influential factors: economic risks such as inflation and price fluctuations were the main drivers of cost overruns, while poor planning and design were leading causes of schedule delays. These findings highlight the urgent need for proactive, data-driven strategies based on the proposed model to effectively manage risks in Syria's bridge construction sector.