Dynamical neural network approach to the job-shop scheduling

In recent years, neural networks have been used successfully in many fields because of the progress in computer technologies. This paper proposes a dynamical neural networks model that modifies Foo and Takefuji’s algorithm for solving job-shop scheduling problems. It also shows how to improve implementation efficiency including the arrangement of neurons, form of energy function and enhancement of the bias input matrix in order to promote the quality of solution during the schedule generating procedure. It provides, according to the results of numerical examples, a powerful approach for obtaining the optimal scheduling solution.