A simulation study in an automotive part manufacturing company

Abstract

Productivity improvement plays an important role in determining whether a company can survive in the future based on customer demands that it can fulfil. In this respect, simulation can be utilised as a what-if analysis tool to decide which productivity improvement strategies to be adopted. This thesis presents a simulation of the current performance of outputs and profits using WITNESS simulation software. The main objective of this project is to demonstrate the use of simulation in analyzing the existing production floor performance and in evaluating various alternatives to overcome the existing problem. This project is conducted in an automotive part manufacturing company located in Kawasan Perindustrian Nilai, Negeri Sembilan. The production line involved in this project is the main assembly line K2 while the product involved in this study is the rear door frame. First, a simulation study on the existing production line is carried out to evaluate the production floor performance. Results of the study revealed that the current production floor was suffering from product scheduling problems. From here, a total of 8 dispatching rules were proposed and they were simulated to determine their effect on the production performance. Kruskal – Wallis nonparametric test was conducted and the results showed a significant difference between each of the rules. This test is chosen because the numbers of experiments conducted are too small and not suitable to be tested using parametric test. To further supplement the finding, one way ANOVA test with multiple comparisons was conducted to select the best result. The results again showed a significant difference between each rule and the best alternative to improve the existing production line is the shortest processing time rule with a productivity increment of 10.27%.