Research on Production Decision-Making Based on Dynamic Programming

For the current enterprise in the production process faces the problem of high scrap rate, high inspection cost and low profit, how to optimize the decision-making in the multi-stage production process has become an urgent issue. This study focuses on innovating the decision-making optimization model of the multi-stage production process, and firstly, the production process of the electronics factory is clearly defined into four stages: part inspection, finished product inspection, defective product disassembly and disassembly of parts after disassembly. In this paper, we innovatively propose to combine the interdependence and feedback mechanism between different stages in the production process, use dynamic programming and economic benefit evaluation models to optimize the cost of each stage, and further construct the Markov decision process by combining the reward function and state transition probability, and use the Bellman equation to calculate the optimal decision scheme and verify its accuracy. This study provides an effective new fusion model for decision-making optimization in the production process, and makes a breakthrough in multi-stage joint optimization, and finds that strict quality control, especially in the semi-finished product stage, can significantly reduce costs, which provides guidance for enterprises to improve economic benefits. To reduce production costs and improve corporate profits.