Multi-objective optimization of construction project management based on an improved genetic algorithm

In construction project management, it is crucial to consider multiple objectives, such as duration and cost, to develop an optimal plan. This paper established a multi-objective optimization model, taking into account the construction period, cost, safety, and quality of projects. A genetic algorithm (GA) was selected as the solution method, and the non-dominated sorting genetic algorithm-II (NSGA-II) was optimized by cat mapping, adaptive crossover, and mutation operators to obtain an improved algorithm for the model solution. Experiments were conducted to evaluate the performance of the designed algorithm. It was found that the improved NSGA-II exhibited superior convergence and diversity when applied to the test functions ZDT1-ZDT3. The mean construction period obtained from the model solution was 124 days, with a cost of 1,204,782 euros. The quality and safety levels achieved were 0.93 and 0.95, respectively, which were significantly better than those obtained by the NSGA-II. These findings demonstrate the reliability of the improved NSGA-II developed in this paper, suggesting its practical applicability