Page 288 - Design for Six Sigma a Roadmap for Product Development
P. 288

258   Chapter Eight



             { } [ ]     { } ➔         { } [ ]{ }
                                        y 1
                    
 0
              y 1
                                              
 0
                                                  0
                                                      x 1
                          x
                           1
                                        y 2
              y 2
                                              

 0
                                                      x 2

                          x
                           2

              y 3
                                                      x 3

                                        y 3
           Figure 8.13 Decoupling by adding extra design parameters.
              parameters so that the number of FRs equals the number of design
              parameters, if a subset of the design matrix containing m 
 m
              elements constitutes a triangular matrix (Fig. 8.13*).
           2. Perform decoupling by utilizing the system’s sensitivity. In this
              case, the designer is seeking parameters that have a minimal effect
              on FRs other than the targeted FR. This can be done by analyzing
              the magnitude of the off-diagonal elements in the design matrix by
              varying the x values over an extreme design range.
           Methods 1 and 2 seek decoupling or uncoupling by adding, replacing,
           or changing the sensitivity of design parameters. These methods may
           greatly benefit from other axiomatic design theorems and corollaries
           (Suh 1990). In addition, a great solution synergy can be gained using
           TRIZ contradiction elimination principles (Chap. 9) to reduce or elim-
           inate coupling vulnerability:
             Functional requirement (y 1 )  A 11 A 12  design parameter (DP 1 )
                                                                           }
           {Functional requirement (y 2 )         ]{design parameter (DP 2 )
                                      } [A 21 A 22
           Assume the coupled design matrix above. The DFSS team can use
           TRIZ to make the sensitivities A 12 , A 21 , or both negligibly small by the
           right choice of the DPs using TRIZ Altschuller’s contradiction matrix
           (Sec. 9.8).
             TRIZ is based on principles extracted from international patents
           showing how people have invented solutions for different categories
           of technical problems in the past. The principles are organized in prob-
           lem categories for selective retrieval, and the methods include proce-
           dural algorithms. Because the principles are associated with similar
           problems successfully solved in the past the likelihood of success is
           enhanced. A simplified TRIZ process description is given in the follow-
           ing steps:
           a. Convert the design problem statement into one of a conflict
              between two FRs considerations


             *The entry 
 in the Fig. 8.13 design matrices is a shorthand notation for nonzero
           sensitivities (∂y i /∂x i   A ji ).
   283   284   285   286   287   288   289   290   291   292   293