Equipment throughput is one of the most critical parameters for production planning and scheduling, which is often derived by optimization techniques to achieve business goals. However, in semiconductor manufacturing, up-to-date and reliable equipment throughput is not easy to estimate and maintain because of the high complexity and extreme amount of data in the production systems. This article concerns the development and implementation of a throughput management system tailored for a semiconductor wafer fabrication plant (Fab). A brief overview of the semiconductor manufacturing and an introduction of the case Fab are presented first. Then, we focus on the system architecture and some concepts of crucial modules. This study also describes the project timescales and difficulties and discusses both tangible and intangible benefits from this project.
12 April 2017 in IEEE Transactions on Semiconductor Manufacturing
Photolithographic (Photo) plays a key role in semiconductor manufacturing because of its importance to advanced process shrinking. Even with a small improvement in its operational efficiency, the cost competitiveness in production can be enhanced as a result of the huge amount of share capital cost. However, it is difficult to stabilize the throughput rhythm of Fabs, while keeping a high equipment utilization for Photo. In the light of Industry 4.0 and big data, a huge potential of maintaining a desired system performance by (near) real-time dynamic system control is highly anticipated. But it also poses challenges to intelligently handling mass data acquisition and allocating computing resources. This research aims to maximize the equipment utilization in Photo by an efficient multi-model simulation optimization approach with big data techniques in the era of Industry 4.0. dynamic Photo configurator and abnormality detector are the two critical units in our proposed system framework; the former can make a quick decision to optimize the system configuration while receiving the adjustment request from the latter. The results from an empirical study show the practical viability of proposed approach that the capacity loss in Photo has been effectively improved.
12 July 2018 in IEEE Transactions on Components, Packaging and Manufacturing Technology
The advances in packaging technology in the past decade have overcome a few engineering limitations in integrated circuit (IC) manufacturing. This has greatly complicated the manufacturing process and created a huge challenge in the operations management of the semiconductor back-end production. In particular, the modern demand of lighter and smaller products expedites the multichip packaging technology, which requires reentrant processes and hence makes resource scheduling more difficult. Apart from the fact that IC packaging shares many key features with the semiconductor front-end production, the cycle time of back-end production is significantly shorter than that of the front-end production. Therefore, there is an urgent need of a rapid solution procedure to generate a reliable production schedule for IC packaging. To respond to customer requests efficiently, this paper models the production scheduling of IC packaging as an optimization model and formulates a hybrid genetic algorithm (GA) to solve the problem efficiently. The embedded structure of our model enables the decomposition of the original problems into many small-sized subproblems, which can be solved by available optimization solvers. These subproblems communicate via a master problem, which is solved by a GA to determine the due dates assigned to subproblems. The master and the subproblems are iteratively solved in turn to obtain a satisfactory solution. Computational experiments and an empirical study are performed to validate the efficiency and the feasibility of the proposed approach.
5 October 2013 in International Journal of Precision Engineering and Manufacturing