The choice is not only between overhaul and repair; equipment replacement alternatives lurk in the background and need to be considered as part of a sequential decision strategy. Possible sequences could include repair, overhaul perhaps a second overhaul replacement, repair, overhaul, and so on.
Suppose, for example, that a machine is usually overhauled or replaced after two years of service. The present machine was purchased two years ago and a decision must now be made concerning overhaul or possible replacement. The machine costs $ 9000 installed and annual operating costs (including maintenance) were $2000 for the first year and $3000 for the second year. The machine can be overhauled for $5000, but the operating costs for two years will be $2800 and $4000 after the first overhaul and $3500 and $5000 after the second overhaul.
In deciding whether to overhaul or replace at this time, we should consider the available alternative sequences of decisions. For example, we can overhaul or replace at this time. For each of these possible decisions, we have the same options two years hence, ad so no. In order to compare the alternatives, the futures costs are discounted to present value. The four alternative strategies are the following:
1. Replace now and in two years (R – R)
2. Replace now and overhaul in two years ( R – OH)
3. Overhaul now and replace in two years (OH – R)
4. Overhaul now and again in two years (OH – OH)
Because operating costs increase so rapidly in this example, it might be worthwhile in the present value analysis to see what happens with a longer horizon, perhaps six years. This of course would add more alternative branches to the tree.
This example assumes replacement with an identical machine, but it is often true that alternative machines will have rather different capital and operating costs. New machine designs often have improvements that reduce labor and maintenance costs, and these cost advantages could affect the analysis and the overhaul – replacement decisions.
Implications for the manager:
The general concepts of system reliability are important for managers to understand. When productive systems involve a network of activities with many required sequences, it will be difficult to maintain the reliability of the system as a whole. This system unreliability would exist even though each individual operation might be 99 percent reliable. Managers can improve reliability by providing parallel capabilities and slack capacity, although these remedies may be expensive.
The most important techniques available to managers for sustaining reliability are quality control and equipment maintenance systems. The quality control system functions as a primary control loop, and the maintenance system provides reliability in the longer term through a secondary control loop.
Quality control begins in the pre-production planning phases of an enterprise, when policies regarding market strategies are developed. Quality standards are then developed out of the iterative process of product/service design and productive system design. The productive system must be designed so that it is capable of producing the required quality level at reasonable cost.
Monitoring quality levels of output is necessarily a sampling process because the entire output population may not be available for screening and the cost of 100 percent inspection may be too great. The techniques of statistical quality control are often valid and cost-effective mechanisms for managers to employ.
Quality control of services is difficult for a variety of reasons related to the unique character of the services. Attempts are being made to establish a framework for control. Legislation has placed great emphasis on quality control in health care systems, and self regulation experiments are now developing. Nevertheless, quality control techniques in services operations are underdeveloped, partly because rigorous standards for the quality of services are not available.
Managers often regard the maintenance function as ancillary to operations, ignoring its crucial role in supporting the reliability system. It is important to understand when preventive maintenance is likely to be appropriate. The analysis of breakdown time distribution provides guidelines for the development of preventive maintenance policies. In general, these policies are appropriate when breakdown time distributions exhibit low variability and when the average time for preventive maintenance is less than the average repair time following breakdown. Also, when down time costs are large, preventive maintenance is preferable to repair if it can be performed when the facilities re normally down anyway.
The maintenance function extends into decisions involving major overhaul and replacement. These kinds of decisions involve a longer time horizon and the proper handling of capital costs. Present value techniques for the evaluation of alternative strategies may be used, and these strategies need to consider sequences of decisions involving repair, overhaul, and replacement.