Economic-based design of engineering systems with degrading components using probabilistic loss of quality

The allocation of means and tolerances to provide quality, functional reliability and performance reliability in engineering systems is a challenging problem. Traditional measures to help select the best means and tolerances include mean time to failure and its variance; however, they have some shortcomings. In this paper, a monetary measure based on present worth is invoked as a more inclusive metric. We consider the sum of the production cost and the expected loss of quality cost over a planned horizon at the customer’s discount rates. Key to the approach is a probabilistic loss of quality cost that incorporates the cumulative distribution function that arises from time-variant distributions of system performance measures due to degrading components. The proposed design approach investigates both degradation and uncertainty in component. Moreover, it tries to obviate problems of current Taguchi’s loss function-based design approaches. Case studies show the practicality and promise of the approach.