MSc project - R.H.B. Roelofs

Automated production scheduling in semiconductor back-end operations


NXP Semiconductors is a global, world-leading semiconductor manufacturer with manufacturing sites all over the world. In the semiconductor back end production, wafers will be processed into chips. Part of the production schedudeling in this back end production takes place manually, however an automated solution method to schedule the production on a weekly basis is desired. Since the Assembly Plant Guangdong (APG) in China processes many orders per week, data of this company is used in this project. The project will be carried out at the Industrial Technology and Engineering Center (ITEC) of NXP Nijmegen.
A first step has been made in realising a computer program to schedule production, this program is called iSchedule. The heuristic used in this program performs still not as good as desired. This project focuses on providing and implementing a better heuristic to this scheduling problem.


This project emphasizes on improving the newly designed iSchedule tool to provide an automated and optimized production schedule for APG. Key performance indicators (KPI's) to assess the obtained production schedule are:
1. Minimal (conguration) switching time
2. Minimal make span against balanced work on the lines
To this end, a literature study should be conducted to investigate which heuristic(s) is/are best suitable for this planning problem. The most promising heuristic(s) should be implemented in the iSchedule environment. The performance of the new method can be compared with both the current heuristic and the manual planning, based on the described KPI's. The objective is to implement a heuristic in the iSchedule application, which subsequently can be used by the production planners at APG.