Archibald, T.W., Ansell, J.I. and Thomas, L.C.
The stability of an optimal maintenance strategy for repairable assets
Proceedings of the Institution of Mechanical Engineers. Part E, Journal of Process Mechanical Engineering, 218, (2), . (doi:10.1243/095440804774134253).
Full text not available from this repository.
Equipment in the process industry is often subject to decay and requires maintenance, repair and eventual replacement. The challenge of competition and the accompanying regulatory regime requires that actions be integrated and cost effective. Ansell and colleagues in 2001 explored an approach to the assessment of asset life of maintained equipment in the process industry using a semiparametric approach. Using stochastic dynamic programming techniques an approach to find the optimal strategy was also developed by Ansell and colleagues. The approach was illustrated using data from the water industry. A major aspect in the development of optimal strategy for repair and replacement is the cost of these activities. Often the costs can only be ascertained roughly in terms of hours expended on the activity. Detailed costings are rarely available. The discount factor will depend on the interest rate in place. Generally a conservatively high level has been taken but with current low rates one needs to explore the sensitivity of solution to the discount factor. Also there is a need to explore the sensitivity of the solution to changes in the costs of the maintenance activities involved. In this paper we explore the stability of the results to changes in the relative costs. It is seen that two costs seem to be more important than the others; hence the accounting effort appears to be best directed towards these costs. One concern that arises from these results is the impact of the maintenance events. In past modelling the impact of maintenance has been assumed to have a fixed effect. This was derived from study of data. It is felt that this assumption may be too simple and so a differing model is explored to examine this issue.
Actions (login required)