Browsing by Subject "Replacement Policies"

Now showing 1 - 2 of 2

Open Access
Modified block replacement models in discrete and continuous time
(2000) Arun, Pelin
In this study, we present modified multi-component block replacement policies. Units (items) are replaced only at prescribed times j = 1,2,... A failed unit is changed with a good one with probability a. Replacement time is negligible. Three replacement policies for models that are not represented as renewal processes are provided under this setup. Some reliability characteristics are discussed. In the first model, total control is considered. All units are controlled at time jT, j = 1,2,.... In the second model, a partial (group) control is studied in which a sample of size n, (0 < n < A) is taken from all units to inspect. And the last model deals with cyclic control: units are divided into r parties. Part}' k is controlled at time jT , j — 1,2,... where j = k (modulus r), k = l,2 ,...,r — 1 and if k is equal to zero then party r is controlled. A comparison between the partial (group) control and cyclic control is provided. We also introduced cyclic partial control which combines the partial and cyclic control policies. The cyclic partial control and cyclic control is compared as well. Cost type of functionals are considered and optimal replacement interval T* is studied as well.
Open Access
Optimal replacement policies with minimal repair and random cost
(1993) Demirel, Hakan Levent
When a system fails usually two actions take place; either replacement of system with a brand new one or repairing it if possible. In this study, it is assumed that system under consideration is repairable and is minimally repaired at failures with a random repair cost. Two replacement models are provided under this set-up. First model assumes that the system is replaced when the total cost of minimal repairs exceeds a total cost limit. Second model incorporates the number of failures into replacement decision. Here the concept of critical failure is introduced and used by means of two sub-models. In the first sub-model it is assumed that the system is replaced at the kth critical failure or at cige T. And the second sub-model assumes that the system is replaced at the first critical failure occurs after age T. The first model is just constructed but cannot be solved due to complexity of the resultant function. But, solution methods of the sub-models of second model are provided.