Reliability Testing and Semi-Infinite Linear Programming
A typical approach in reliability testing of a complex system is to assign to the components an allocated level of reliability, and then figure out the number of component tests which guarantees component reliabilities with a certain level of confidence. Another approach is to test the system as a whole, and base the test plans on the desired value of system reliability. Both approaches have advantages and disadvantages. There is also a third method which is known as system based component testing. It is based on the idea of combining the advantages of component and system tests. The determination of minimum cost component test plans according to this new approach can be formulated as a parameterized semi-infinite linear programming problem. In this paper we explain the mathematical model and describe the solution procedure, which is based on the well known cutting plane idea and column generation technique.
KeywordsSystem Reliability Column Generation Series System Test Plan Column Generation Algorithm
Unable to display preview. Download preview PDF.
- I. K. Altinel. System Based Component Test Problem: The Design of Optimum Component Test Plans. PhD thesis, University of Pittsburgh, Pittsburgh, PA 15261, December 1990.Google Scholar
- I. K. Altmel. A note on the parameter of Poisson probability distribution. Technical Report 90–21, Department of Industrial Engineering, University of Pittsburgh, Pittsburgh, PA 15261, September 1990.Google Scholar
- I. K. Altinel. The design of optimum component test plans in the demonstration of system reliability. European Journal of Operational Research, 78:97–115, 1994.Google Scholar
- I. K. Altmel and S. Ozekici. Optimum component test plans for systems with dependent components. Technical Report FBE-IE-01/95–03, Department of Industrial Engineering, Bog͂aziçi University, Bebek, İstanbul 80815, May 1995.Google Scholar
- T. R. Rockafellar. Convex Analysis, 2 nd printing. New Jersey: Princeton University Press, 1972.Google Scholar