Industrial big data–based scheduling modeling framework for complex manufacturing system

Abstract

Scheduling modeling for manufacturing system has always been a great challenge in both industrial and academic community. With the growing complexity of the manufacturing system, traditional scheduling modeling and optimization methods cannot satisfy all the demands of current manufacturing environment, so data-based methods are brought into practice. Since the popularity of the concept of cyber physical system and Industry 4.0, more information and interaction systems are applied into the manufacturing system, and more industrial big data produced during the production process can be acquired; the knowledge within these data needs to be uncovered to better schedule the system. In this article, the research progress of scheduling modeling methods for complex manufacturing system are systemically reviewed and evaluated with the semiconductor wafer fabrication environment. Then, we propose the industrial big data-based scheduling modeling framework in cyber physical system condition and discuss how to implement it for the semiconductor manufacturing system, also our demonstration unit of an intelligent semiconductor manufacturing system based on this framework is introduced. Finally, the future work of our application case is discussed.

Simulation and deep reinforcement learning for adaptive dispatching in semiconductor manufacturing systems

“…A product or production is considered to be complex if it is subjected to variable customers' demands with recurrent changes in product's specifications and/or production flow. This scheme of production confronts many manufacturing enterprises (Zhu et al 2017) and requires internal logistic flexibility to satisfy their customers' needs (Freitag and Hildebrandt 2016).…”

PlantTwin simulation system as a tool for verifying production plans and supporting decision-making to improve the efficiency of machine-building industries

“…However, such systems do not allow us to take into account a number of factors, including probabilistic ones, which are typical for any real production, such as equipment failures, repairs, transport restrictions, and others. At the same time, such factors can have a decisive impact on the timing and ability to complete a given production plan [4][5][6].…”