Job Shop Scheduling Problem Neural Network Solver with Dispatching Rules

Job Shop Scheduling Problem (JSSP) is an optimization problem in computer science and operations research. Many problems in real-world manufacturing processes can be translated into JSSP. In recent years, Machine Learning has shown great promises in solving optimization problems and can be used to solve JSSP instances. In this paper, an Artificial Neural Network (ANN) was designed and trained to solve JSSP instances using the priority of the operations as the learning output. Dispatching rules were implemented to break ties during the decoding of the priorities. Our experiment results showed that a hybrid algorithm that combines the best of ANN with dispatching rules and standalone dispatching rule-based heuristic outperforms previously reported results.