In the field of road engineering construction, project schedules are usually very tight, cost pressures are high, and engineering quality and safety standards are also very strict. These key objectives are prone to contradictions, but traditional optimization methods, such as linear programming or single objective heuristic algorithms, often struggle to coordinate the complex balance between multiple objectives, and have weak adaptability to common process constraints and resource limitations in actual construction. To address the aforementioned multi-objective optimization requirements, this study proposes an improved multi-objective genetic algorithm called MOGA-II. This algorithm mainly includes three parts: firstly, an optimization model is established that includes four aspects: duration, cost, quality, and safety, quantifying and integrating practical constraints such as process logic relationships and resource constraints; Secondly, the algorithm adopts a fitness evaluation method based on Pareto sorting and crowding distance, which can better balance the comparison of solution quality and maintain population diversity; Thirdly, the addition of elite retention strategy and parameter adaptive adjustment significantly improved the convergence speed and global optimization ability of the algorithm. A simulation experiment using a city’s main road expansion project as an example shows that compared with traditional genetic algorithms, MOGA-II has increased the number of Pareto solution sets by 40% and shortened convergence time by 25%; Compared with commonly used NSGA-II and particle swarm optimization algorithms, MOGA-II has reduced the comprehensive indicators of construction period and cost by 12.7%, and the quality pass rate has also increased to 98.5%.
Yang ziqiao (Thu,) studied this question.