Applying Artificial Intelligence to Solving Multiple Optimization in Construction Management: Hybrid Slime Mold Algorithm With Opposition-Based Learning

This study analyzes the hybrid algorithm between Slime Mold Algorithm (SMA) and Opposition-based learning (OBL), in which Adaptive Opposition Slime Mold Algorithm (AOSMA) is proposed to simultaneously solve the triple-objective optimization problem in construction projects. The combination of OASMA aims to improve and upgrade the model in order that the exploration, exploitation and acceleration of the convergence are perfected and local optimizations are avoided during rapid convergence, accordingly, the best pareto solutions shall be provided to the project. The Time-Cost-Quality trade-off problem is one of the prerequisites which lay a solid foundation for the success of a construction project, one of the challenges in simultaneous optimization of the factors to obtain consistence with each specific activity of the project. In order to enhance the superiority and efficiency, the proposed model is compared with previous algorithms (MOSGO, MODE, MOPSO and NSGA-II) for the comprehensive development of the proposed model. According to the overall results, the AOSMA model shows diversification and provides a strong and convincing optimal solution for readers to recognize the potentialities of the proposed model. This paper's contribution is a new hybrid approach for solving optimization problems and determining pareto convergence. These findings have ramifications for the creation of future algorithms. However, this model has limitations, so it is critical to continuously improve the model if the artificial intelligence system for humans is fully developed.