A tool to balance cost and schedule while maintaining facilities readiness.

In the October issue of Controlled Environments, Richard Bilodeau, “Ask the Facilities Guy” wrote about establishing an “Equipment Reliability” program. While clearly an important issue, it is one that often facilities departments have a hard time getting their arms around—as Richard points out. One tool that we have found quite useful in supporting high facility onstream time and process yield factors, as well as sustainability, is the equipment or hardware FMEA (Failure Modes and Effect Analysis). The FMEA exercise will provide the facilities team with a prioritized “risk burn-down” plan for ensuring readiness and can serve as a convenient basis for capital and operating expense budget creation and execution.

We have recently performed FMEA exercises for aerospace assembly, integration, and test facilities, aseptic filling laminar flow units and accompanying HVAC systems, thermal vacuum test chambers, powder metallurgy processing lines, precision cleaning equipment, and continuous web processing machinery. In many cases not only were predictive and preventive maintenance issues uncovered and addressed with corrective action plans developed as a consensus among customers, users, service providers, and subject matter experts, but in a few cases, serious life safety and product safety issues were brought to light and effectively dealt with before a catastrophe— likely one without warning—could occur.

An FMEA identifies the severity, occurrence, and detection of failure effects and then establishes priority- ranked corrective action plans. A cross-functional team including the customer or process owner, subject matter experts, facilities and maintenance specialists, quality assurance, and design engineering participate in a brain-storming exercise that identifies each potential failure and ranks the possible effects of each failure and develops a resulting RPN or “Risk Priority Number.” The RPN is the arithmetic product of the severity multiplied by the (probability of) occurrence multiplied by the (ability of) detection.