Rethinking reliability engineering using machine vision systems

Condition monitoring of rolling stock has been part of the railway landscape for the past decade; thermal, acoustic and light sensors are used to capture fleet information on bearings, brakes, wheels, under-carriage and body. The data generated from these sensors has helped rail operators direct their maintenance efforts to the components of highest priority; however, in most cases it has not triggered a fundamental philosophical change in the maintenance strategy. Rail operators generally use time-based maintenance interventions to ensure fleet reliability and safety. Technology is enabling rail operators to take advantage of the benefits of moving from a time-based maintenance regime to a condition-based regime. However, technology simply provides the tools. The benefits of higher reliability and lower costs can only be realized when the organizational culture adapts to make best use of the technology. This paper will discuss the technology that can be deployed to enable a condition-based culture and some of the organizational issues that need to adapt to the new technology. The observations in this paper are not based on a structured research program, but rather are observational, based on out daily dealings with several rail operators around the world.