Page 274 - Design for Six Sigma a Roadmap for Product Development
P. 274

244   Chapter Eight


               Functional Requirements            Design Parameters
             • FR : Control the flow of water  • DP : Angle of valve 1, φ
                 1
                                                   1
             • FR : Control the temperature    • DP : Angle of valve 2, φ 1
                 2
              of water                             2              2
             Hot water     Cold water         Hot water     Cold water
                       φ 1                                 φ 2
                         φ 2



                                                   φ 1

                  Coupled Design                  Uncoupled Design
             (DPs create conflicting functions)  (DPs maintain independence of functions)
               Control Flow  =  ×    ×  φ 1     Control Flow  =  ×    0  φ 1
            Control Temperature  ×    ×  φ 2  Control Temperature  0    ×  φ 2

           Figure 8.4 Example of design coupling. [From Swenson and Norlund (1996).]



           and environmental impact don’t measure function directly and are not
           delivered by any particular DP; therefore they are called “constraints.”
           A constraint is a property of a design, not something the design does.
           Typically, all elements—not only one element—in the design contribute
           to a constraint. We cannot add on a DP to improve the constraint.
           However, constraints are paced with the mapping in the zigzagging
           process; thus constraint modeling and transfer function modeling are
           both critical.
             The importance of the design mappings has many perspectives.
           Chief among them is the revelation of both the transfer functions and
           coupling among the functional requirements, the domain in physical
           mapping, and the design parameters, the codomain. Knowledge of cou-
           pling is important because it gives the DFSS team clues of where to
           find solutions, make adjustments or changes, and how to maintain
           them over the long term with minimal negative effect.
             The design matrices are obtained in a hierarchy when the zigzag-
           ging method is used (Suh 1990) as described in this chapter. At lower
           levels of hierarchy, sensitivities can be obtained mathematically as the
           FRs take the form of basic physical, mathematical, architectural, and
           engineering quantities. In some cases, the transfer functions are not
           readily available and some effort is needed to obtain them empirically
           or via modeling (e.g., CAE or discrete-event simulation). Lower levels
           represent the roots of the hierarchical tree where Six Sigma quality
           level can be achieved.
   269   270   271   272   273   274   275   276   277   278   279