Franz, Roger L. (1992, ASQC) Motorola, Inc., Northbrook, IL 60062
Failure analysis applies to more than just products delivered to customers and means more than simply waiting for a failure and doing a post mortem. It also includes internal services and processes and can be applied before failures actually occur.
During the design phase of customer products, samples and prototypes can be tested in a lab to determine failure rates that in turn can be projected to the actual products. Factorial experiments, Taguchi techniques, and other design of experiment (DOE) techniques apply.
Failures can be examined for root causes and defects listed in Pareto charts. Process quality control indicates whether results are within limits for the process. Continuous improvement can be achieved by fine tuning the process to reduce failure rates. Quality goals can be set by benchmarking competitive products.Another source of information is customer complaints, whether the customers be internal or external.
Failure analysis contributes to many aspects of the Baldrige quality system model. The techniques described here can be extended even further into the design phase for even earlier detection.
Continuous improvement (CI),Design of experiments (DOE),Factorial experiments,Failure analysis,Pareto charts,Root cause analysis (RCA),Taguchi method