As pipelines age, maintenance is the primary method of restoring performance an prolonging equipment life. According to industry analyst firm ARC Advisory Group, pipelines are relatively easy to maintain. While pipeline related incidents still present significant risks to operators and the environment, more advanced technology is now available to detect pipeline degradation and leaks. Applying data mining techniques to the data generated from pipeline maintenance and monitoring technology, as part of a larger asset management strategy, will help drive asset enhancement and improve reliability. Data mining is a combination of data management, statistical techniques, data analysis and pattern recognition. From a reliability engineer's perspective, it may be said that "the secrets of life are contained in the data," and the reliability engineer is challenged to extract these secrets and use them to inform knowledge based decisions. Asset strategy decisions are made using a combination of knowledge and judgment from various sources. However, in today's pipeline environment, systems are complex and there is a great deal of interaction and interdependence between systems. Pipeline requirements, operating windows, maintenance strategies, process changes and operating procedures are among the factors that contribute to varying degrees and rates of equipment degradation, and eventually lead to a functional failure.
展开▼
