New Hybrid Models Integrating the Firefly Optimization Algorithm with the Artificial Neural Networks and Adaptive Neuro-Fuzzy Inference Systems to Improve Estimation at Completion

Abstract

In the construction industry, project performance often faces challenges due to the unpredictable nature of operational environments, causing construction projects to frequently exceed their estimated initial budget. Construction companies must adopt a vigilant approach that continuously monitors construction project costs and promptly identifies and corrects discrepancies to maintain profitability. Estimation at Completion (EAC), which projects the total cost of a project at completion, is a crucial tool for managers to use to monitor project performance and mitigate associated risks. However, existing methods for predicting the EAC are criticized for their low accuracy. This highlights the need for new methods to improve the accuracy of EAC projections. The primary objective of this research is to create a vigorous artificial intelligence model that reliably predicts the final project cost based on the actual data obtained throughout the project's lifecycle. To achieve this, this study proposes new models that integrate Artificial Neural Networks (ANN) and Adaptive Neuro-Fuzzy Inference Systems (ANFIS) with the Firefly Optimization Algorithm (FFA). The major objectives of the proposed model are to reveal the most significant factors influencing the EAC and to enhance its prediction accuracy by utilizing fewer input parameters. The proposed models were developed using a historical data set comprising 306 construction projects completed between 2000 and 2007 in Taiwan. The findings illustrate that the FFA-ANN model outperforms the FFA-ANFIS and traditional standalone models in terms of prediction accuracy. Overall, this study demonstrates that using the FFA algorithm for parameter selection significantly enhances the performance of both the ANN and ANFIS models, offering a promising approach to improving cost estimation in construction projects.