Page 446 - Design for Six Sigma a Roadmap for Product Development
P. 446

Failure Mode–Effect Analysis  407


               of its intended life. The potential process failure mode is the way
               in which a processed entity in the structure may fail to deliver its
               array of DPs.
            5. Studying the failure causes and effects. The causes are generally
               categorized as weaknesses due to
               a. Process weakness because of axiom violation. In this cause of
                  failure, a component is manufactured and assembled to process
                  capability and design specifications. Nevertheless, it still can’t
                  be assembled or function as intended. Usually this will happen
                  when the design was not conceived concurrently between design
                  and manufacturing.
               b. Noise factors mean effects and their interaction with the PVs:
                  (1) Manufacturing and assembly variation and deficiencies
                      due mainly to incapable processes or material variation
                  (2) Production environment and operator error
                  (3) Machine deterioration: wear over time
               c. The effect of a failure is the direct consequence of the failure
                  mode on the next higher hierarchical level processes, and ulti-
                  mately the customer. Effects are usually noticed by the opera-
                  tor or the monitoring system at the concerned process or
                  downstream from it. Potential failure causes can be analyzed
                  at the process level by tools such as
                  (1) Fault-tree analysis (FTA)
                  (2) Cause-and-effect diagram
                  (3) Cause-and-effect matrix
               d. Two golden rules should be followed in cause identification
                  (1) The team should start with the modes with the highest
                      severity ratings from the related DFMEA(s).
                  (2) They should try to go beyond the first-level cause to second-
                      or third-level causes. The team should ask the following
                      process questions:
                      (a) What incoming source of variations could cause this
                          process to fail to deliver its array of DPs?
                      (b) What could cause the process to fail, assuming that the
                          incoming inputs are correct and to specifications?
                      (c) If the process fails, what are the consequences on
                          operator health and safety, machinery, the compo-
                          nent itself, the next downstream processes, the cus-
                          tomer, and regulations?
            6. Ranking of potential process failure modes using the RPN num-
               bers so that actions can be taken to address them. Each potential
               failure mode has been considered in terms of severity of its effect,
               detection likelihood, and occurrence of its causes.
   441   442   443   444   445   446   447   448   449   450   451