Data Collection

Appropriate data collection, after identifying the problem is one of the most important steps in reliability analysis. A data collection must be designed, developed and maintained to facilitate correct and effective reliability analysis. The uses of the collected data may be one or more of the following:

• Detection of in-service problems leading to production losses.
• Evaluation of reliability and maintainability characteristics.
• Studying the economics of system breakdowns.
• Feed back to manufacturer for design modification if needed.

Data on reliability characteristics can be obtained by two methods, either directly from the field or from the sample testing in laboratories. For an effective reliability program, the collection and analysis of the failure statistics such as failure modes, their frequencies and possible causes of failures for each component and equipment type from a representative population operating under identical condition is needed.

Data collection in field is an expensive and time consuming work. Nevertheless, data acquired from field is the best representative of the system reliability and maintainability characteristics and all efforts should be made to collect data from the field. Some of the points which are of practical importance for development of field data collection systems are given here (Mahon, 1972; Ireson, 1988):

• The depth of reporting and data collection should vary according to the importance of the equipment or its components.
• Reporting should not be confined to facts. Opinions in the "engineering judgment" category should also be included.
• Attempt should be made to collect only those data that are likely to be useful.

During the last few years, various companies with high degree of mechanization have realized the potential benefit of collection of machine performance statistics.