Abstract:Line balancing problem and buffer allocation problem are often studied separately, but there is a complex interaction between them. Considering stochastic assembly line, the operating time variability exacerbates the interaction between the two problems. Optimizing sequentially is difficult to get the global optimal solution. Therefore, it is necessary to solve the two problems simultaneously. The probability distribution of operating time is measured as station complexity based on information entropy. An integrated optimization model is established. The optimization objectives are as follows: maximizing of the production rate and minimizing of the smoothness index of standard workstation time, the smoothness index of workstation complexity and total buffer capacities. Parametric modeling simulation is used to calculate production rate. An improved genetic algorithm is put forward to obtain integrated optimization solution. An instance of a gearbox assembly line is calculated and verified, which proves the effectiveness of the method.