Research On Product Quality Control Based On Multi-agent Modeling And Quantifier Constraint Satisfaction

In the trend of constantly pursuing high-quality products,manufacturing-oriented enterprises are increasingly paying attention to the quality control of products in the industry competition.With the advancement of information technology,mechanical product manufacturing companies are actively exploring quality control methods combined with information technology.In this paper,the quality uncertainty of products in the manufacturing stage is analyzed,and the process of product quality is analyzed.This paper summarizes the factors affecting the formation of quality that includes man,machine,material,method,environment and measurement from the perspective of Total Quality Management.Interviewed with quality experts and compiled the Likert Scale Questionnaire,which was issued to a Precision Machinery Co.,Ltd.After the exploratory factor and confirmatory factor analysis of the questionnaire,an index system of factors affecting product quality loss was established.In order to thoroughly study the quality uncertainty in the production process,it is necessary to explore the evolution mechanism of quality loss.Based on the multi-agent modeling and simulation(ABMS)method and the product quality loss influencing factors index system,the product quality loss emergence model is constructed.Then,this paper uses the Netlogo simulation platform to study the various processes of the whole process of quality formation,and dynamically simulates the influence of six factors of man,machine,material,method,environment and measurement(5M1E)on the emergence trend of product quality loss.Finally,the article adjusts the various factors of the model,and carries out comparative simulation under the benchmark mode and the analysis mode.The analysis analyzes the influence degree of the uncertain factors of the six aspects of man,machine,material,method,environment and measurement on the product quality and the function cross rule between the factors.Mechanical products disturbed by uncertain factors in the manufacturing process and the output quality parameters fluctuated results in low robustness of quality characteristics.In order to deeply control the uncertain factors,this paper considers the uncertainty constraint in the mechanical product design stage,establishes the Quantifier Constraint Satisfaction Problem(QCSP)model to optimize the quality parameters,and studies the quality control problem from the design process to the production process.Firstly,the product parameters are set to existential and universal variables.Secondly,the obtained function cross rules are introduced into the uncertainty constraints.The model is established,and the robustness index that meets the designer's preference is set.The NSGA-II algorithm is used to search for the Pareto optimization solution in the feasible domain,and the entropy weight is calculated to select the optimal solution,and the product quality characteristics of the manufacturing stage are robustly controlled.Finally,combined with the above two methods,the quality parameters of a certain type of mechanical press are robustly designed,which proves the correctness and high efficiency of the two methods combined in engineering applications.