A Research Model for Production Process Inspection Based on Machine Learning and Monte Carlo Simulation

This article addresses the decision-making issues related to inspection in the production processes of enterprises and proposes a decision model based on machine learning and Monte Carlo simulation. The research aims to optimize the inspection process, reduce costs, and maximize profits by providing a scientific decision-making framework for enterprises. Firstly, a sampling inspection scheme is designed using hypothesis testing methods to determine the defect rate of spare parts supplied by vendors with minimal testing. Secondly, a decision tree model and Monte Carlo simulation are employed to make optimal decisions for each stage of the production process and calculate expected returns. Finally, the multi-step process problem is transformed into a combination of multiple spare parts and finished products, applying the decision tree model to solve it and using sampling inspection methods to estimate the defect rate, optimizing the model to minimize total costs and maximize the profit from qualified products. Experimental results indicate that this model demonstrates significant advantages in terms of accuracy, cost control, and profit maximization.