Optimization of a Flexible Manufacturing System (FMS) for High Technology Product Producing Factory via the Modularization of Production Systems

• Main Author: Prashobh, Karunakaran
• Format: Thesis
• Language: English; English
• Published: Universiti Malaysia Sarawak (UNIMAS) 2011
• Subjects: TA Engineering (General). Civil engineering (General)
• Online Access: http://ir.unimas.my/id/eprint/32794/1/20201118115416.pdf; http://ir.unimas.my/id/eprint/32794/4/Prashobh%20Karunakaran%20ft.pdf; http://ir.unimas.my/id/eprint/32794/

The objective of this research is to justify the utilization of buffers in High Technology Product Producing Factories (HTPPF) especially in the hard disk industry. This is significantly different from the trend in other industries such as canned food or car, where machines are joined and the process are continuous from start to end. In the latter industries, buffers are counted as stock and a waste, but in HTPPF, buffers are critically important to enable research activities to achieve daily upgrade of the Key Quality Characteristics (KQC) of the product. This has enabled the hard disk industry to increase the data capacity of a 95mm diameter Al-NiP disk from 1 GB in 1995 to 800 GB in 2012. A common trend among hard disk manufacturers, especially in Japan, Taiwan and also the Western Digital factory in Johor, was to join up all the production machines, which is preventing research activities. The assumption made by HTPPF is that dedicated research lines should be a good enough test bed for research activities required to achieve the technological improvements to the products. This is a wrong assumption because research lines are not always running and it is a fact that upon startup of a line, the yield is bad and this has to stabilize before any research changes can be made. Therefore the best test bed for research is a production line, which already has a good yield and a small change in chemical for example, can show up at the final test point to indicate if improvement to the KQC has occurred. Lack of awareness of this technique has caused a number of HTPPF to have decreased market share or close down. The few successful HTPPF that utilizes buffers operates them manually. This introduces human made defects, contamination and also a decreased throughput. In this research, designs were developed to automate these buffers such that humans need not handle the products, while at the same time enabling researchers to stop the production line to test out their hypotheses. Having designed these improvements, the next problem to solve is the throughput of HTPPF. So designs were made to solve the defect detection system (DDS) which is always the bottleneck in HTPPF. The next problem is to solve the wastage in the current system due to the inaccurate demand predictions at the factory, wholesaler and supermarket. To solve this problem a feasibility study was made to enable a customer ordering a product with a credit card to initiate factory machines to immediately manufacture it.