Optimal balancing of multi-objective U-shaped assembly lines using the TSGA method

Abstract

Many assembly lines in industries are now being designed as the U-shaped assembly line balancing (UALB) due to pressures of just-in-time manufacturing. Once compared to the straight assembly line balancing (SALB), it has better balancing, improved communication, fewer workstations, and more flexibility for adjustment. This paper proposes a hybrid intelligent approach to solve such the UALB problems. The TSGA method consisting of the tabu search (TS) and the genetic algorithm (GA) is used to identify solutions for the UALB problems. The multiple objectives including the workload variance, the idle time, and the line efficiency, are proposed and set as the objective function of search process. With the proposed approach, the TS can well address the number of tasks assigned for each workstation of the U-shaped line, while the GA can also assign the sequence of tasks for each workstation according to precedence constraints. The proposed approach is tested against three UALB problems from a survey of literature. Obtained results are compared with results obtained from the single-objective approach. As results, the proposed multiple-objective approach based on the TSGA method gives better solutions for all UALB problems.