Date of Award


Document Type


Degree Name

Doctor of Philosophy (PhD)

First Advisor

Lawrence Mann, Jr


In this research, a model for fault diagnosis of equipment is discussed. The model discusses the use of multi-variable monitoring for a more precise means of equipment malfunction trouble shooting. Also within this model means are discussed for automatically generating work-orders. Equipments are divided into subsystems. The subsystems are broken down into sub-subgroups and further down to individual parts. Preventive maintenance techniques, along with expert system technology, are used to develope the maintenance expert system (MES). Statistical pattern recognition techniques are used as a way to describe the relationship between categories of equipment. The relationships shown here are mostly that of E. A. Patrick. These relationships give an example of how different categories within each subsystem could be effected by each other. In order to achieve accuracy when this model is applied, new and original probability density functions applicable for that particular case should be used. A prototype MES is built.