A bottleneck prediction and rolling horizon scheme combined dynamic scheduling algorithm for semiconductor wafer fabrication
Jijie Deng, Zhengcai Cao, Min Liu
Abstract
Jijie Deng, Zhengcai Cao, Min Liu
Abstract
Semiconductor wafer fabrication is a very complicated manufacturing system which is characterized by re-entrance production flow, complicated constraints, and dynamic and uncertain environment. Bottleneck is the key factor to the wafer fabrication which has essential influence on the throughput rate, cycle time, time-delivery rate, etc. The optimal scheduling policies for bottleneck machines are conducive to cost decreasing, throughput target achieving and production-resource scheduling improving. In this paper, considering the characteristic information concerning bottleneck and the whole production line, a bottleneck prediction method based on improved adaptive network-based fuzzy inference system (ANFIS) is presented by predicting the wait time of the lot in workstation buffer and the workstation load. Then, based on the obtained bottleneck and directly considering the layer load levels of bottleneck and due date, we proposed a dynamic different layer bottleneck-based scheduling algorithm which includes releasing policy and dispatching algorithm for preventing starvation of bottleneck, balancing work in process (WIP) and higher throughput. Applied to a simulation wafer fabrication, our proposed bottleneck prediction method and bottleneck-based scheduling algorithm can reduced the cycle time and improve machine utility compared with other common heuristic scheduling method.
OpenAlex reports 6 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Semiconductor wafer fabrication is a very complicated manufacturing system which is characterized by re-entrance production flow, complicated constraints, and dynamic and uncertain environment. Bottleneck is the key factor to the wafer fabrication which has essential influence on the throughput rate, cycle time, time-delivery rate, etc. The optimal scheduling policies for bottleneck machines are conducive to cost decreasing, throughput target achieving and production-resource scheduling improving. In this paper, considering the characteristic information concerning bottleneck and the whole production line, a bottleneck prediction method based on improved adaptive network-based fuzzy inference system (ANFIS) is presented by predicting the wait time of the lot in workstation buffer and the workstation load. Then, based on the obtained bottleneck and directly considering the layer load levels of bottleneck and due date, we proposed a dynamic different layer bottleneck-based scheduling algorithm which includes releasing policy and dispatching algorithm for preventing starvation of bottleneck, balancing work in process (WIP) and higher throughput. Applied to a simulation wafer fabrication, our proposed bottleneck prediction method and bottleneck-based scheduling algorithm can reduced the cycle time and improve machine utility compared with other common heuristic scheduling method.
Key concepts: Bottleneck, Computer science, Scheduling (production processes), Wafer fabrication, Job shop scheduling, Semiconductor device fabrication, Algorithm, Dynamic priority scheduling