Multi-Stage Assembly Optimization Study Based on Decision Tree Algorithms

This paper presents a comprehensive study on multi-stage assembly optimization using decision tree algorithms. A novel approach to writing and recognizing “decision codes” is introduced and applied to decision trees to acquire a comprehensive and readable set of decision-making options. To address potential cyclic repetition that could affect stability, a limit fitting method is employed for control. The decision tree algorithm is utilized to process various parameters across scenarios, culminating in a comparison of outcomes based on ability metrics, thereby identifying the most advantageous assembly strategy. Furthermore, this research extends to the examination of independent multi-process models, analyzing the transition from components to semifinished products and then to finished products. By evaluating the overall weight of each component, the study identifies elements with minimal impact on the total system, thereby optimizing the model without compromising accuracy. This optimization significantly enhances model efficiency while maintaining accuracy through the preservation and transfer of state data from one stage to the next, ensuring stable and precise multistage assembly processes.