The development of an intelligent maintenance optimisation system

Proudlove, NC

Abstract

This thesis describes the background to and development of a computer-based
decision support system (DSS) known as IMOS, the intelligent maintenance
optimisation system. The aim of the system is to help industrial maintenance
engineers improve the planned preventive maintenance policies applied to large and
complex technical systems. IMOS attempts to achieve this by providing some
automated analysis of the huge amounts of maintenance history information which is
accumulating in the computerised data bases of many large industrial companies. The
keys to this analysis are a set of mathematical models of the effects of maintenance
activities and expert judgement about which of the models is most suitable under a
particular set of circumstances. These features are incorporated in the IMOS
software as a 'model base1 module, consisting of a set of routines for each
mathematical model, and a 'rule base' module which selects the most appropriate
models by recognising characteristic patterns in the historical data for each item of
equipment. There are no previous attempts in the maintenance literature to formulate
such a list of rules to guide model selection.
The study and modelling of industrial maintenance is reviewed, as is relevant work on
the support of management decision making and the features and evolution of DSS is
also discussed. The need for and benefits of a system such as IMOS are described
and the suitability of the intelligent decision support system approach is discussed.
The mathematical models, the selection rules, and optimisation criteria and
techniques are detailed, and the development of the software, written in C for an
IBM compatible PC, is described. The research was conducted in collaboration with
two major oil exploration and production companies and data from several North Sea
oil-production platforms are analysed and discussed. Finally, achievements and
shortcomings of the system are discussed and some suggestions for further research
outlined.

Citation

Proudlove, N. The development of an intelligent maintenance optimisation system. (Dissertation). University of Salford

• Thesis Type: Dissertation
• Deposit Date: Jul 23, 2021
• Award Date: Jan 1, 1995