Analysis of a manufacturing system to improve performance using characteristics curve and modeling & simulation

• Main Author: Vivek, Suresh Babu
• Other Authors: Sivakumar A. I.
• Format: Final Year Project
• Language: English
• Published: 2019
• Subjects: DRNTU::Engineering::Manufacturing::Production management
• Online Access: http://hdl.handle.net/10356/77723
• Institution: Nanyang Technological University

This project aimed to present an analysis for a semiconductor assembly and test manufacturing system. The objective of the final year report was to suggest improvements to enhance the performance of the factory for the NMPCC package. This was done by studying the characteristic curves for the manufacturing system and by experimenting with a simulation created on SIMUL8. Actual factory data was investigated; characteristic curves were plotted to illustrate the best-case, worst-case and practical worst-case scenarios. The current factory performance was plotted on the graph and found to be in the ‘bad’ region. The practical worst-case performance was calculated to be 7492.95 units/hr whereas the current performance was only 166.23 units/hr. The actual factory analysis also showed that the bottleneck occurred at station 1. A model was created for simulation using SIMUL8; it was then validated using actual factory values with a small difference of 3.47%. The simulation was used to find methods to enhance factory performance. The author found that additional machines at workstations would increase the throughput for the manufacturing system. The simulation also experimented with the statistical distribution of arrival for parts. The experiments demonstrated that varying parameters such as operating hours, machine efficiency and number of machines had effect on the overall factory performance. Finally, recommendations were made based from model experimentation results to improve factory performance. The author found that in order to increase the throughput value significantly, additional machines must be purchased. The main limitation for the project was the lack of data for the NMPCC package. All manual calculations for the NMPCC package were taken as estimates for the whole production line; this consisted of several packages. The author suggests that a more in-depth study can be performed if time permits and relevant data is sourced.