A hybrid metaheuristic for scheduling problem in sheet metal processing

Abstract

In recent years, manufacturing industries are required to provide various kinds of products in a short time while reducing burden on the environment. To achieve such requirement, optimization techniques incorporated with ICT are attracting a rising attention. In sheet metal processing, reducing wastes of material and managing production schedule are considered to be very important for the efficient manufacturing. To work with these problems, we attempt to simultaneously minimize wastes of material and total tardiness through nesting and scheduling. The nesting and scheduling in sheet metal processing are affected with each other and there occurs certain conflicts in their decisions. For entire efficiency of manufacturing activities, therefore, they should be decided through managing trade-off between them. Under such understanding, in our previous study, we proposed an integrated method of nesting and scheduling. This method determines a schedule under the condition that the operational sequence of punching process as the first process of sheet metal processing is fixed. In order to realize more efficient manufacturing, in this paper, we have included the punching process in our consideration. Then, this study has proposed a hybrid heuristic approach to the scheduling problem of sheet metal processing. Genetic algorithm, dispatching rule and local search technique are applied to create a schedule together with a bottom left algorithm for nesting. Finally, in order to validate the effectiveness of the proposed method, we implemented prototype of nesting scheduling system and carried out a few computational experiments.

A Heuristic Solution Method of Mixed Problem of Materials' Cutting Plan Decision and Production Scheduling for Foundation Reinforcement Units Manufacturing

“…The equation (7) shows that operation time PT p k j of part j of order p is decided based on set-up time MT kl and operation time t kl of the lot that part j belongs to. Decision variable y p kl j denotes if part j of order p is batched into lot l at process k. Z kml ′ shows the processing sequence between lot l ′ and lot l on machine m in process k, and can be used to calculate set-up times as shown in equation (6).…”

“…al. [6] have proposed a hybrid heuristic approach to the scheduling problem of sheet metal processing, which genetic algorithm, dispatching rule and local search technique are applied to create a schedule together with a bottom left algorithm for nesting. Toyoda [7], [8] made a revision to first-fit algorithm and proposed a new heuristic solution method for multiple stock length CSP.…”