A Hybrid Approach Based on Artificial Neural Network and Genetic Algorithm for Job-shop Scheduling Problem

Job-shop scheduling problem (JSSP) is very common in a discrete manufacturing environment. It deals with multi-operation models, which are different from the flow shop models. There are some difficulties that make this problem difficult. Firstly, it is highly constrained problem that changes from shop to shop. Secondly, its decision mainly depends on other decision which are not isolated from other functions. It is an NP-hard problem. In this paper, we proposed a new hybrid approach, combining ANN and GA, for the job-shop scheduling. The GA is used for optimization of sequence, neural network (NN) is used for optimization of operation start times with a fixed sequence, thanks to the NN's parallel computability and the GA's searching efficiency, the computational ability of the hybrid approach is strong enough to deal with complex scheduling problems. The results indicate that the proposed algorithm can obtain satisfactory for the job-shop scheduling problem