COMPUTERISED MAINTENANCE SCHEDULING FOR SMALL UTILITIES

Abstract

In this study, a computerised maintenance scheduling system which caters to the conditions and constraints of small utilities consisting solely of thermal generating plants is developed. The general approach taken is to compare the performance of a few commonly used methods. Two deterministic methods and a stochastic method were examined. The deterministic methods employ the levelling of reserve capacity criterion, of which one uses a heuristic extension to level the equivalent reserve capacity computed from the static probabilistic case. The stochastic method simulates the levelling of risk criterion by using the peak load carry capacity (PLCC) of the available units. The results indicate that for the size and type of maintenance scheduling problem modelled, the PLCC method does not produce a schedule which is significantly better than the deterministic methods tested. The small amount of computer CPU time required to implement the implicit enumeration algorithm to determine the optimal schedule with the deterministic criteria also indicates the feasibility to apply the maintenance scheduling program on a microcomputer system.