The manufacturing process requires high-precision machining and assembly, where the degree of automation of assembly is particularly important for production efficiency. Therefore, taking the fuel cell assembly line as an example, this paper proposes an au-tomated parallel assembly line balance and worker allocation method considering the task type in view of the problem of different degrees of automation of parallel assem-bly lines and processes in automated assembly lines. Firstly, a mathematical model is established to minimize the number of workstations and the number of workers, and a distribution method that allows workstations to share workers under the condition of automated assembly line is proposed. Secondly, a hybrid variable neighborhood search genetic algorithm is proposed to solve large-scale problems. Finally, by comparing the solution of the standard study with a variety of algorithms, the actual problem of the fuel cell stack assembly line is at the same time.In this paper, the mathematical model of the assembly line is optimized, the efficiency of the assembly line is improved, and the actual production cost is reduced.