Hierarchical Transfer Learning for Cycle Time Forecasting for Semiconductor Wafer Lot under Different Work in Process Levels

Abstract

The accurate cycle time (CT) prediction of the wafer fabrication remains a tough task, as the system level of work in process (WIP) is fluctuant. Aiming to construct one unified CT forecasting model under dynamic WIP levels, this paper proposes a transfer learning method for finetuning the predicted neural network hierarchically. First, a two-dimensional (2D) convolutional neural network was constructed to predict the CT under a primary WIP level with the input of spatial-temporal characteristics by reorganizing the input parameters. Then, to predict the CT under another WIP level, a hierarchical optimization transfer learning strategy was designed to finetune the prediction model so as to improve the accuracy of the CT forecasting. The experimental results demonstrated that the hierarchically transfer learning approach outperforms the compared methods in the CT forecasting with the fluctuation of WIP levels.

Modeling and Simulation of a Digital Twin of a Production System for Industry 4.0 with Work-in-Process Synchronization

“…On the other hand, the Work-in-Process inventory is one of the basic components involved in the cost of production in an enterprise and it is expressed in inventory units. Articles about WIP are related to optimizing the size of the inventory [31], especially for batch production [32], or cycle time forecasting according to the fluctuation of WIP levels [33] and the allocation of buffers on the effectiveness of an assembly manufacturing system [34]. The research [31,32] shows the phenomenon of the accumulation of WIP in the process route in the manufacturing industry, which is related to a great and uneven distribution that is dependent on different factors, including batch supply transport, the maximum volume and maximum load capacity of the buffers, various types of products, complex scheduling, and a long production cycle with a bottleneck on some machines [33][34][35].…”

Preface to the Special Issue “Mathematical Modeling in Industrial Engineering and Electrical Engineering”—Special Issue Book

“…The paper authored by J. Wang, P. Gao, Z. Li and Q. Bai [18] has as its main purpose the construction of a model for forecasting the cycle time (CT) unified under dynamic work-in-progress levels. Particularly, the work proposes a transfer learning approach to fine-tune the predicted neural network in a hierarchical way.…”