Optimization of Automobile Painting Operation Scheduling Based on DQN

Our research concentrates on determining an optimal operation sequence for the painting workshop in an automobile manufacturing factory. A deep - Q Network (DQN) algorithm is adopted to solve this problem, aiming to satisfy the sequential demands of the final - assembly workshop and minimize spray - gun color changes. DQN's advantages include its ability to handle complex decision - making in high - dimensional spaces and adapt to environmental changes. The training and test results demonstrate its effectiveness in converging to near - optimal solutions and maintaining good performance across different instances. Future research directions may involve enhancing DQN's generalization ability, integrating it with other techniques, and considering real - time factors in production.