A Cost Analysis of Systems Subject to Random Field Environments and Reliability

We present a generalized cost model subject to random field environments with considerations of cost to remove failures during testing and warranty periods, and penalty cost due to the system failures. We also determine the optimal release time policies that minimize the expected system cost. Many scientific contributions have been developed in software reliability modeling, while none has studied the manufacturing or industrial system reliability growth yet, considering also the differences between testing and operating environments. The application of the proposed model, in comparison to the nonhomogenous Poisson process Goel-Okumoto model, to an industrial application is discussed to illustrate how the proposed model can be used in practice.