Abstract: In a U-type production line, often the operation time is subject to random fluctuation and it takes time for an operator to go from one workstation to another. With these factors considered, the U-line balancing problem is addressed in this paper. Given cycle time and workstation completion rate, a mathematical model of minimizing the number of workstations is presented. Then, a solution method that combines genetic algorithm and simulation is proposed. Based on the layout of the line and the precedence relation of the tasks, an operator's walking route can be determined by simulation. Thus, the time taken for operator's walking is obtained, so is the time for processing a task. In this way, the balancing problem can be solved. A numerical example is given to show the effectiveness of the approach.

Key words: U-line balancing, random processing time, simulation, genetic algorithm