Reliability development and improvement of a medical instrument

A new medical blood analyzer (laboratory instrument) was designed, developed and readied for production. Near the end of this series of activities, ten prototype systems exhibited lower than desired reliability in a development test. It was unclear to the designers why this had occurred as "the best design criteria" were employed. A short, controlled, full life test was proposed as a means to quantify the probable types and frequency of failure. This paper details the reliability findings at this point as well as methods for improvement of the typical product. Some of the technical choices and tradeoffs for reliability, maintainability, field performance, costs and quality as well as engineering decisions associated with reliability and field support are identified. Close attention is paid to the identification of a small number of important reliability measures. The reliability, maintainability, support and improvement data should prove highly instructive for any commercial or consumer company wishing to justify starting or continuing the reliability improvement process. Benchmark data is presented to aid others in establishing progress points during development. The data represents a blend of several similar systems.