Track: Graduate Student Paper Competition

Abstract

The recent advancement in digital technologies has encouraged manufacturers to adopt intelligent data analysis and decision-making support tools to leverage their competitiveness. They have become challenged by the need for more flexible and complex manufacturing systems. This complexity and flexibility cannot be handled efficiently with traditional production management paradigms, such as lean management. Semiconductor industry is an example of a complex production, in which high level of flexibility is required to meet a continuously changing customer demand. This limits the efficiency of static modeling approaches and it reflects negatively on the overall production yield.

In this research, we propose a discrete event simulation logic for assessment of several challenging production management-related problems at a semiconductor manufacturer. The logic includes scheduling and processing a large variety of products, while respecting relative priorities and customers’ needs. It also accounts for production resources, equipment and operators, with variable availability. Our simulation logic is data-driven, in which data that drives the different actions taken within the execution of the simulation, is compiled. Consequently, the model is highly flexible to continuous changes in the model’s parameters. The proposed logic provides a quick and risk-free tool for capacity planning and reliable experimentation with different ‘What-If’ scenarios.