A Production Cycle Prediction Method for Multi-product Production Line Based on Variability Measurement

Abstract

The invention belongs to the technical field of production line management. The production cycle prediction method of a multi-product production line based on variability measurement provides the variability measurement and corresponding product-oriented production cycle forecasting method. The method includes: extracting and processing the original data, historical operation data, and future cycle planning data of each station in the production line; constructing the logical station of the product according to the actual link of the product, and connecting the logical stations in series to obtain the logical production line of the product; Composed of data and logic production lines, calculate the variability of each logic station on the product logic production line and the variability of transmission between logic stations; calculate the queuing time of each logic station in the future cycle based on the VUT equation, and accumulate and sum each logic station The queuing time and effective processing time of the station can be used to obtain the overall production cycle of the product logic production line.

Description

technical field [0001] The invention belongs to the technical field of production line management, and in particular relates to production cycle prediction under the condition of multi-product mixed production. Background technique [0002] The production cycle (Cycle Time, CT) is one of the key performance indicators of the product (such as semiconductor chip packaging and testing) production line. It refers to the average time spent by a batch of products as a work-in-process (WIP) in the production line. The production cycle is a key parameter for making a production plan. If the CT of the production line can be accurately predicted, the production plan will be more reasonable and the production efficiency will be improved accordingly. The semiconductor chip packaging and testing production line includes many processes, and each process corresponds to a corresponding workstation (Workstation). The product is processed as a work-in-progress through each workstation in sequ...

AI Technical Summary

Problems solved by technology

This multi-product mixed production and "serial-parallel-serial" equipment composition characteristics make the effective processing time composition of the workstation very complicated, which in turn leads to the complexity of the measurement of the variability of the workstation. At present, there is no corresponding scientific and reasonable The measurement method, which has become the biggest obstacle to predict the production cycle of this type of production line based on the VUT equation

[0004] Although, the application number is 200810201503.X, and the patent application titled "production cycle target measurement and system" discloses a method of predicting (or measuring) the production cycle by directly fitting the relationship between utilization rate and queuing time , although this method can also achieve the goal of production cycle prediction, it only considers the utilization parameter for modeling, so once the variability changes (product and link changes) during the production cycle, the method will fail; in addition, the The scheme does not distinguish the cycle of each product in the case of multi-product mixed flow, so it can only calculate the overall production cycle of multiple products

That is, because it completely regards the workstation as a black box, it cannot distinguish the specific impact of each device connected in parallel in the workstation (equipment corresponds to a link) and different products passing through the same workstation on the production cycle, and cannot provide more detailed information for further performance optimization. Granular Directional Guidance

Images