Thesis (Ph.D.)-- Western Michigan University, 2004. / Includes bibliographical references (leaves 114-116).

Thesis (MScEng (Mechanical and Mechatronic Engineering))--University of Stellenbosch, 2011. / ENGLISH ABSTRACT: Vacuum assisted resin transfer moulding (VARTM) belongs to the category of resin infusion techniques that use lower than atmospheric pressure to infiltrate a reinforced cavity. This technique has various advantages; however, manufacturing costs can be relatively high due to more difficult shapes fabricated and a lack of knowledge regarding cost driving factors. The objective of this study was to develop a cost model for composite parts. Such a model allows the estimation of manufacturing costs of shapes of different geometries. Therefore, it provides a comparison to alternative manufacturing techniques, such as metal forming or composite spray lay-up and helps to avoid unnecessarily expensive design features. The proposal was made to split complex shaped composite parts into individual basic shapes, which are further investigated here. For the basic shapes, an experimental approach was used where the manufacturing times of each process step are measured and then statistically analysed. Infusion simulation software was used to obtain additional filling times to complete the design of experiments. This method allows the estimation of manufacturing times of composite parts with different geometries. The manufacturing times were validated to that of a complex shaped industrial part, with reasonable results. Finally, a flexible cost model was developed to compare different manufacturing techniques and to estimate the manufacturing costs. / AFRIKAANSE OPSOMMING: Vakuumgesteunde harsinspuitingsgietwerk (VARTM) behoort tot die kategorie harsinspuitingstegnieke wat laer-as-atmosferiese druk gebruik om ʼn versterkte holte binne te dring. Hierdie tegniek hou verskeie voordele in. Tog kan vervaardigingskoste betreklik hoog wees wanneer dit by ingewikkelder vorms en ʼn gebrek aan kennis met betrekking tot kostesnellers kom. Die doelwit van hierdie studie was om ʼn kostemodel vir saamgestelde onderdele te ontwikkel. Die model maak voorsiening vir die raming van die vervaardigingskoste vir verskillende afmetings. Sodoende bied dit ʼn vergelyking met alternatiewe tegnieke, en help voorkom onnodig duur ontwerpkenmerke. Daar is voorgestel dat dele met ingewikkelde vorms in individuele basiese vorms verdeel word, wat dan hier verder ondersoek word. Vir die basiese vorms word ʼn eksperimentele benadering gebruik waar die vervaardigingstye in elke prosesstap gemeet en statisties ontleed word. Voorts word inspuitingsimulasiesagteware gebruik om komplementêre inspuitingstye te bepaal ten einde die eksperimentele ontwerp te voltooi. Hierdie metode maak dit ook moontlik om die vervaardigingstye vir saamgestelde materiaal onderdele van verskillende afmetings te raam. Die vervaardigingstye word dan bevestig aan die hand van dié van ʼn kompleks gevormde industriële onderdeel, met redelike resultate. Uiteindelik word ʼn buigsame kostemodel ontwikkel om verskillende vervaardigingstegnieke te vergelyk en die vervaardigingskoste te raam.

A dynamic job shop control system with a combined input/output control mechanism is developed for achieving improved shop performance. The problem is modelled such that at periodic intervals, the best combination of input and output variables for the forthcoming period are identified. The purpose of the control system is to determine the set of jobs to be released into the shop (input variables) and the capacity levels of machines (output variables) for a planning period such that a composite cost function is minimized. The problem is mathematically formulated as a 0-1 linear mixed integer program (MIP). An iterative based heuristic optimizing algorithm incorporating the MIP is developed. The control algorithm is compared with another job shop control system where only the input is assumed variable. The two systems are compared by computer simulation and results indicate significant improvements for most of the performance measures evaluated. Significant reductions in the mean and variance of the manufacturing lead time with a better distribution among its parts were also achieved. / Ph. D.