Capacity Planning For Mixed-Load Tester Under Demand And Testing Time Uncertainty

• Main Author: Asih, Hayati Mukti
• Format: Thesis
• Language: English; English
• Published: 2018
• Subjects: T Technology (General); TA Engineering (General). Civil engineering (General)
• Online Access: http://eprints.utem.edu.my/id/eprint/23390/1/Capacity%20Planning%20For%20Mixed-Load%20Tester%20Under%20Demand%20And%20Testing%20Time%20Uncertainty.pdf; http://eprints.utem.edu.my/id/eprint/23390/2/Capacity%20Planning%20For%20Mixed-Load%20Tester%20Under%20Demand%20And%20Testing%20Time%20Uncertainty.pdf; http://eprints.utem.edu.my/id/eprint/23390/; https://plh.utem.edu.my/cgi-bin/koha/opac-detail.pl?biblionumber=112887
• Institution: Universiti Teknikal Malaysia Melaka

Capacity planning is an important decision in production planning as it determines the capacity to install in order to satisfy customer demands and also to allocate products to those capacities.This research is based on mixed-load machine problem which is categorized by multiple products that can be processed simultaneously with different processing time.The problem is further complicated with high product varieties and high demand variabilities.This research was conducted based on a case company from a multinational manufacturing company in Malaysia that produces hard disk drives.The study focused on the automated testing process characterized by long lead time and high product variability.Each testing machine with 2880 slots is a mixed load tester with the ability to load and test multiple product families simultaneously.In addition,the uncertain demand and testing time makes the problem more challenging. Currently,the company’s issue is low tester utilization of about 71%,well below the target of 96%.The objective of this research is to improve tester utilization while achieving the production target under uncertain demand and testing time and also to determine the break-even point on the testers required.A novel approach of integrating a mathematical model,robust optimization model,genetic algorithm,simulation model and cost–volume –profit analysis was developed.Firstly,a mathematical model of mixed-load tester was formulated.Next,a set of discrete scenarios was proposed to address uncertain demand and testing time.A robust optimization and genetic algorithm model was developed to optimize the number of testers under the described uncertainties.Next,these scenarios were simulated using the Pro Model simulation software to validate the proposed models and to evaluate throughput and tester utilization.Finally,the cost–volume–profit analysis was performed for scenarios that require additional testers at various levels of uncertainties.The results showed that the proposed solution improved tester utilization by 25% compared to the current system.This research has contribution by developing novel hybrid methodology and able to provide useful insights to assist company’s managers to plan and allocate resources according to variations in customers’ demands and testing time.