Dynamic bottleneck analytical method of semiconductor production line

Abstract

The invention discloses a dynamic bottleneck analytical method of a semiconductor production line. The method includes the steps that firstly, a key parameter capable of representing the bottleneck device characteristic is selected to be used as a bottleneck analysis index; secondly, the comprehensive bottleneck degree of equipment is described quantificationally through a combined definition method from the aspects of the relative production load, the utilization rate, the queue length of a buffer area and the like of the equipment, the bottleneck equipment is recognized by combining a bottleneck judging mechanism, and an equipment number is obtained; furthermore, a bottleneck prediction model of the semiconductor production line is established through an increased trimming type neural network based on the key parameter and the bottleneck equipment number, and the network structure is corrected in a dynamic mode based on a closed-loop control concept; at last, the key parameter affecting the bottleneck is analyzed through a single-factor testing method, and the reasonable value range of the parameter is obtained during bottleneck stabilization. By the adoption of the method, the bottleneck equipment can be recognized rapidly and effectively, the bottleneck at the next moment can be predicted, bottleneck drift is analyzed quantificationally under the condition of the dynamic effect of the key parameter, and a basis is laid for optimal dispatching of the semiconductor production line.

Description

technical field [0001] The invention relates to a dynamic bottleneck analysis method of a production line, which belongs to the field of semiconductor processing and automatic control, in particular to a dynamic bottleneck analysis method of a semiconductor production line. Background technique [0002] The semiconductor manufacturing industry is a capital-intensive high-tech industry, which has great strategic value for economic development. Under the background of the transfer of the world's manufacturing center to the Asia-Pacific region and the sustained and rapid development of my country's economy, my country's semiconductor industry has developed rapidly. In order to continue to maintain the good momentum of the development of my country's semiconductor manufacturing industry and enhance market competitiveness, it is necessary not only to expand the scale but also to improve production efficiency. The semiconductor production line is considered to be the most complex...

AI Technical Summary

Problems solved by technology

At present, the research on the bottleneck mainly has the following problems: (1) There are many parameter variables affecting the bottleneck, and the selection of key variables is directly related to the accuracy of the analysis of the bottleneck equipment. (2) In terms of bottleneck prediction methods, static prediction based on mathematical analysis and simulation technology is mostly used, lacking consideration of uncertain factors such as workpiece types, batches, and equipment failures; (3) Scheduling methods based on fixed bottlenecks , lagging behind the change of the bottleneck equipment, making the scheduling strategy centered on the fixed bottleneck lack of pertinence, reducing the timeliness of production control

Images