Browsing by Author "Ghasemi, Amir"

Now showing 1 - 2 of 2

Citation - WoS: 3
Citation - Scopus: 5
Demonstration of the Feasibility of Real Time Application of Machine Learning To Production Scheduling
(Institute of Electrical and Electronics Engineers Inc., 2022-12-11) Ghasemi A.; Kabak K.E.; Heavey C.; Ghasemi, Amir; Heavey, Cathal; Kabak, Kamil Erkan
Industry 4.0 has placed an emphasis on real-time decision making in the execution of systems, such as semiconductor manufacturing. This article will evaluate a scheduling methodology called Evolutionary Learning Based Simulation Optimization (ELBSO) using data generated by a Manufacturing Execution System (MES) for scheduling a Stochastic Job Shop Scheduling Problem (SJSSP). ELBSO is embedded within Ordinal Optimization (OO), where in the first phase it uses a meta model, which previously was trained by a Discrete Event Simulation model of a SJSSP. The meta model used within ELBSO uses Genetic Programming (GP)-based Machine Learning (ML). Therefore, instead of using the DES model to train and test the meta model, this article uses historical data from a front-end fab to train and test. The results were statistically evaluated for the quality of the fit generated by the meta-model. © 2022 IEEE.
Citation - WoS: 25
Citation - Scopus: 27
Optimizing Capacity Allocation in Semiconductor Manufacturing Photolithography Area - Case Study: Robert Bosch
(Elsevier Sci Ltd, 2020-01) Ghasemi, Amir; Azzouz, Radhia; Laipple, Georg; Kabak, Kamil Erkan; Heavey, Cathal
In this paper, we advance the state of the art for capacity allocation and scheduling models in a semiconductor manufacturing front-end fab (SMFF). In SMFF, a photolithography process is typically considered as a bottleneck resource. Since SMFF operational planning is highly complex (re-entrant flows, high number of jobs, etc.), there is only limited research on assignment and scheduling models and their effectiveness in a photolitography toolset. We address this gap by: (1) proposing a new mixed integer linear programming (MILP) model for capacity allocation problem in a photolithography area (CAPPA) with maximum machine loads minimized, subject to machine process capability, machine dedication and maximum reticles sharing constraints, (2) solving the model using CPLEX and proofing its complexity, and (3) presenting an improved genetic algorithm (GA) named improved reference group GA (IRGGA) biased to solve CAPPA efficiently by improving the generation of the initial population. We further provide different experiments using real data sets extracted from a Bosch fab in Germany to analyze both proposed algorithm efficiency and solution sensitivity against changes in different conditional parameters.