On a Mathematical Design System: Maximum Reliability, Minimum Cost

2088's picture
Saleh Sadeq Afaneh

In this thesis, reliability concepts, measures of reliability and static models have been studied. Also, comparisons between exponential and logistic distributions have been discussed. We used two methods; dynamic programming and heuristic approach to maximize the reliability of an electronic device systems, where the optimum structure of components and units of the assumed system have been determined. Examples are given to show the optimum structure ofthe system with the maximum reliability and minimum cost. Comparison between dynamic programming and heuristic approach shows that the dynamic programming results are better than the results obtained by the heuristic approach. Finally, our objective is to characterize marginal cost and minimize cost capacity plans for a typical service delivery system. Results indicate that marginal costs are convex with respect to reliability of service, while changes in the demand distribution’s variability may impact optimal capacity by either increasing or decreasing required capacity. Two demand distributions are assumed; uniform and logistic distributions. The results show that the logistic demand distribution gives an optimum criterion which are more realistic. Also, the optimum capacity using logistic is greater under the condition b/B> 0.5.