Failure mode and effects analysis (FMEA) is a method that examines potential failures in products or processes and has been used in many quality management systems. One important issue of FMEA is the determination of the risk priorities of failure modes. In this paper we propose an FMEA which uses data envelopment analysis (DEA), a well-known performance measurement tool, to determine the risk priorities of failure modes. The proposed FMEA measures the maximum and minimum risks of each failure mode. The two risks are then geometrically averaged to measure the overall risks of failure modes. The risk priorities are determined in terms of overall risks rather than maximum or minimum risks only. Two numerical examples are provided and examined using the proposed FMEA to show its potential applications and benefits. © 2009 Elsevier B.V. All rights reserved.