Page 297 - Design for Six Sigma a Roadmap for Product Development
P. 297

Axiomatic Design  267


             of a pair of mating gears are tangential to each other. However, the pitch
             diameters are hypothetical circles that can’t be measured directly. The func-
             tion of the gear is to “transfer speed.” This function can be mapped to many
             design parameters that can be grouped basically into two subsets of DPs: a
             geometric subset and a material property (stiffness, hardness, etc.) subset.
             The diameter of the pitch circle PD follows a normal distribution as a result
                                                                         2
                                                                    2
             of manufacturing process variability: PD    f(PD)    (1/ 2   )  e  PD /2  2 .
             Then, we have
                          h(f)     f(PD) ln f(PD)
                                          PD 2
                                  f(PD)          ln 2    2
                                          2  2
                                 1       2
                                  2  E (PD )   ln  2    2
                                2
                                        2
                                ln  2 e   nats                         (8.14)
             This equation is depicted in Fig. 8.18.
               In the case of a normal source of information, this example shows that
             information and complexity are both functions of variability. A reduction in
             the variance will reduce not only the probability of manufacturing noncon-
             firming gears but also the required information needed to manufacture the
             part. This is the power of design axioms.
               Equation (8.14) states that, in the case of normal information source, a
             random variable, complexity, and information are functions of variability.






              3.5

                3
              2.5

           σ 2  2

              1.5
                1

              0.5

                0
                         0.6    1.2    1.8     2.4    3.0     3.6
                                           h
           Figure 8.18 Variance as a function of complexity.
   292   293   294   295   296   297   298   299   300   301   302