Research On Re-delivery Strategy Of Process Material Considering In-process Inspection Results

With the implementation of China Manufacturing 2025,the intelligent workshop is becoming an important mode for transformation and upgrading of integrated manufacturing enterprises,and 3C manufacturing enterprises are typical of them.Because of their mature automation capabilities,they often have advantages in the transformation and upgrading of intelligent workshops.As a typical 3C manufacturing enterprise,the mobile phone frame manufacturing enterprise is mostly a unitized flow layout,which has many features such as many processes,fast cycle time and high delay penalty.In the production process,due to the randomness of the production system,there are non-conforming products in the processing.In order to meet the demand output,the enterprise often inputs more in the delivery process.Because the product unqualified rate is not fixed or difficult to predict accurately,companies often do not know how much material to input,and more delivery can meet the production requirements,but it causes waste.Meanwhile,less investment may not meet the order requirements leading to delayed delivery.This paper studies the inspection process of workshop production,and proposes a process re-delivery strategy based on the minimum sampling capacity.By correcting the sampling qualification rate at different stages and fitting the actual production qualification rate to guide the re-delivery of the materials in the workshop process,the material input meets the order demand,which reduces the production cost and improves the delivery of the order.According to the proposed research ideas,this paper mainly carried out the following work.Firstly,according to the current condition of production process of the phone frame workshop,the process inspection and process delivery business process are analyzed,and the mathematical model of process material re-delivery is established,and the original problem is decomposed into two sub-problems: prediction problem(small sample qualified rate prediction large sample qualified rate problem)and optimization problem(re-delivery strategy optimization problem).For the prediction problem,the Bayes estimation method is adopted.For the optimization problem,a re-delivery strategy based on the minimum capacity of sampling is proposed.Secondly,the Plant Simulation software was used to build a single-process multi-unit intelligent workshop simulation platform based on production characteristics,abnormal factors and process inspection.The parameters,operation mechanism and performance indicators of the simulation platform were given.Among them,the non-conforming product generation mechanism in the simulation platform is obtained through the following analysis: the factor analysis of the unqualified workpieces,the tool wear is the main reason for the unqualified workpieces,and the influence of the tool on the product unqualified rate is different under different conditions.Four strategies based on sampling frequency and sampling method are proposed.Then,through the design of a large number of simulation experiments,the Bayes estimation method is validated and analyzed for the four sampling strategy samples.The trend relationship between the qualification rate of the process sampling and the overall qualification rate is analyzed.For the stability of the rate and the degree of error with the overall qualified rate,the sampling strategy and the Bayes estimation are analyzed.A sampling strategy was selected,and the minimum sampling capacity is determined considering the two main workshop environments.Finally,according to the sampling strategy selected above,a process re-delivery strategy based on the minimum sampling capacity is proposed.With the help of the simulation platform,the simulation experiment of the re-delivery strategy is carried out under the consideration of the two factors of the workshop environment and the production order requirements.Through statistical analysis of each performance indicator,verify the effectiveness of the strategy,and further determine the coefficient values involved in the strategy.