Doctor of Philosophy
Department of Mechanical Engineering
Dwight, Richard, Frameworks for measuring the performance of the maintenance system in a capital intensive organisation, Doctor of Philosophy thesis, Department of Mechanical Engineering, University of Wollongong, 1999. http://ro.uow.edu.au/theses/1582
The measurement of performance can be viewed at vastly different levels of complexity. Performance can be defined as the difference between what is deemed possible to achieve relative to the organisation's goals and what is actually achieved. This implies that the standard of performance must be in terms of the organisation's goals and what is possible. The measurement must connect actions undertaken in a period of interest with their immediate and long-term outcomes. The concept of measuring changes in value to determine performance is thus proposed, where value is normally expressed as the net present value of future cash flows.
Traditional measures have been found to ignore this fundamental requirement, leading to an inability to account for lags between harmful actions and their effects. This link is critical to pursuing logical approaches to the management of the maintenance system. It is missing from all currently advocated maintenance performance measurement approaches. The use of ratios to measure performance, whilst promising a catalogue of measures to be used and benchmarked, does not allow appropriate standards of performance to be determined.
Key to the measurement of performance is the establishment of a standard based on established goals. The subjective nature of the process of goal determination has been found to limit the possibility for objectivity in performance measurement. Similarly the perceived level of control of the subject system over the organisation's goals establishes the boundaries of the search for performance measures.
The measurement of performance in terms of changes in value accommodates these concerns. It also introduces some difficulties and constraints on measurement approaches. Determination of future operating requirements, changes in use, failure behaviour given a particular set of actions, and the organizations perception of risk are some of the variables that prove difficult to quantify. Each has a significant impact on the definition of the standard of performance. Nonetheless decisions, and so decision quality, depend on estimates of these future events. Performance measurement must be concerned with likely outcomes of current decisions and so with prediction of future outcomes.
Although the fundamental basis for performance measurement has been identified, practical approaches to its determination must be found. Based on the establishment of fundamental equations for value-based performance measurement, frameworks for two new measurement approaches have been established and their practical application explored. The systems-auditing approach concentrates on the alignment of maintenance system activities with the organisation's goals. The relative performance of these activities is established as a surrogate for performance in terms of value. The approach calls for the comparison of the current approaches to individual functions with alternatives. This introduces the search for alternatives as an integral part of performance measurement. It fits with a quest for improvement.
A second new approach to performance measurement has been labelled the event analysis approach. It focuses on understanding the impact of specific actions on the value of the organisation. Alternate actions are considered in terms comparing future values. The standards set are the best known actions given the situations arising during the period. Both new approaches focus on a standard of performance related to what could be achieved as opposed to achievement of a budget or benchmark.
The new approaches have been applied, allowing establishment of their nature and applicability. They establish a high scientific rigour to the measurement of the performance of management systems, including in particular maintenance management systems. Both new approaches will benefit from detailed modelling of the maintenance process.
02Whole.pdf (3709 kB)