An Artificial Intelligence Model-Based Locally Linear Neuro-Fuzzy for Construction Project Selection
S.M. Mousavi, B. Vahdani, H. Hashemi and S. Ebrahimnejad
Selecting the best project in construction industry is a complex decision problem in which numerous conflicting factors should be considered for the assessment. Among well-known models in the soft computing field, artificial intelligence (AI) can be suggested to yield better predictions than traditional methods. For this purpose, this paper introduces an effective AI model based on new neural networks and fuzzy logic to improve the decisions for projects owners. A computationally AI model, namely locally linear neuro-fuzzy (LLNF), is proposed to precisely predict the overall performance of projects in construction industry. Proposed model can be properly employed for the long term prediction of performance data in construction industry. Finally, the model is applied to facilitate the assessment process in a real case study. To demonstrate the applicability of the proposed AI model, the computational results in terms of the performance and accuracy are compared to two widely-used regression methods.
Keywords: Artificial Intelligence, locally linear neuro-fuzzy model, learning algorithms, construction project selection.