Batch scheduling optimization method based on deep reinforcement learning and genetic algorithm

Abstract

The invention belongs to the field of production and manufacturing scheduling, and discloses a batch scheduling optimization method based on deep reinforcement learning and a genetic algorithm, and the method comprises the steps: building a mathematic model of a difference workpiece batch scheduling problem; establishing a strategy model of the problem by adopting a pointer network; training a pointer network model by using an actor-critic algorithm; defining and initializing parameters of a genetic algorithm; optimizing the initial population of the genetic algorithm by using the trained pointer network; further optimizing the scheduling scheme by adopting a genetic algorithm; and the optimal scheme obtained by the genetic algorithm being used as a production scheme for processing workpieces by the batch processor. Compared with a traditional heuristic algorithm, the pointer network can obtain a better solution; in addition, in the crossover operation of the genetic algorithm, a novelcrossover mode is provided, and the performance of the scheme can be further improved by improving the optimization capability of the genetic algorithm on the basis of the scheduling scheme obtainedby the pointer network.

Description

technical field [0001] The invention belongs to the field of production and manufacturing scheduling, specifically a batch scheduling optimization method based on deep reinforcement learning and genetic algorithm. Background technique [0002] Batch scheduling problems arise from burn-in operations used in final test in semiconductor manufacturing. In this operation, integrated circuits are placed in batches in a high-temperature oven, and after a long period of time, possible early failures of integrated circuits are detected. The burn-in operation is often a bottleneck in semiconductor manufacturing because it typically takes longer to process than other operations in final test. Therefore, it is very important to have efficient scheduling of ovens (or machines) to greatly increase their utilization. At present, the batch scheduling problem not only exists in the semiconductor manufacturing industry, but also widely exists in most manufacturing industries, such as foundr...

AI Technical Summary

Problems solved by technology

At present, there is little research and attention on the deep neural network to solve the manufacturing scheduling problem, and there is no application of the deep neural network in the batch scheduling problem of differential workpieces.

Images