Process Equipment Reliability Database (PERD) | AIChE

Process Equipment Reliability Database (PERD)

Last updated March 23, 2020

The CCPS PERD Project’s mission was to help processing and manufacturing industries minimize hazards and improve equipment availability with high-quality equipment reliability and maintenance data.  High-quality, structured data give companies better insight into equipment reliability and performance and thus enable data-driven decision making for equipment assets.  High-quality data also support quantitative risk assessment (QRA).

With PERD methods, participating companies could use high-quality data to quickly identify equipment failure events with the greatest consequences.  They could analyze those failure events to identify failure patterns of the equipment causing them and obtain details necessary to take corrective action.  They could then resolve the reliability issues by implementing corrective maintenance measures.

PERD design principals were to simplify, standardize and automate equipment reliability data processes. They standardized by using methods consistent with industry and international standard processes, by representing facilities in a logical and standard manner, and by using standard failure data collection, merging, and assessment processes. 

The Center for Chemical Process Safety had been providing quantitative risk assessment data to the chemical process industries since the 1980s, and published the first of its “CCPS Guidelines for Process Equipment Reliability Data” in 1989.  Subsequent enhancements to this information-collection effort led to PERD, which provided deeper and more specific analyses of equipment availability, reliability, design improvements, maintenance strategies, and life cycle cost determination.

The PERD software component is being phased out and managed by Tony Ciliberti on a personal basis. The PERD taxonomy and publications are still available for purchase. For more information about PERD, please contact Tony at For taxonomy purchases, please contact