Abstract: When multiple projects work at the same time, rational allocation of various resources and pursuit of greater economic profits through human resource management have become the focus of research and development enterprises. From the perspective of project cost and satisfaction of multi-skilled workers, the second-generation non-dominated sorting genetic algorithm and ant colony algorithm are used to study the allocation of multi-skilled workers. Considering the matching between the skill mix of multi-skilled workers and the needs of project tasks and the impact of skill proficiency on task operation time, an optimization model is constructed that aims to achieve the total duration of multi-projects and the workload balance among multi-skilled workers. According to the constraints of the model, a series of heuristic rules are proposed to improve the efficiency of the algorithm. The algorithm is developed by combining the characteristics of the second-generation non-dominated sorting genetic algorithm and ant colony algorithm to solve the model. Numerical examples verify the effectiveness of the model and algorithm.