Page 394 - Six Sigma Advanced Tools for Black Belts and Master Black Belts
P. 394

OTE/SPH
 OTE/SPH
                         3:8
          August 31, 2006
                              Char Count= 0
 JWBK119-24
                                       References                            379
      is only limited to the origin point whereas the two-chart V-mask’s FIR is added into
      the consecutive CUSUM.
      24.2.4  ARL comparisons
      Using the Markov chain approach, Lucas and Crosier 11  analyzed the average run
      length properties of the tabular FIR CUSUM scheme. Similarly, by comparing ARL
      properties, Atienza et al. 12  further established the equivalence of the tabular FIR
      CUSUM and the V-mask FIR CUSUM. The ARL values for the tabular FIR CUSUM are
                              11
      given by Lucas and Crosier, while those for the V-mask are obtained by Monte Carlo
                                                         12
      simulation. The results of the comparison by Atienza et al. are replicated here in Ta-
      ble 24.2. It is clear that, in addition to the analysis concerning the tabular and V-mask
      FIR CUSUMs, the ARL comparison in Table 24.2 further supports the equivalence of
      the two FIR CUSUM schemes.



                                24.3 CONCLUSIONS

      The FIR feature is found to be useful in implementing the CUSUM chart in its tabular
      form. In this chapter, we have discussed a procedure for developing an equivalent
      scheme to the V-mask CUSUM. The V-mask FIR CUSUM requires only a simple
      modification of the first point in calculating the CUSUM but requires plotting two
                                                 n
      separate charts. An alternative CUSUM, CUSUM , is thus proposed to overcome the
      need for this. That is, instead of using the point at the origin (0,0), two points are
      plotted, namely (0,−FIR) and (0,+FIR). This makes the V-mask FIR CUSUM looks
      like a fork at the start-up stage. This procedure of implementing the FIR feature in
      V-mask form is also applicable to all other forms of CUSUM masks.



                                    REFERENCES

       1. Champ, C.W. and Woodall, W.H. (1987) Exact results for Shewhart control charts with
         supplementary run rules. Technometrics, 29, 393--399.
       2. Page, E.S. (1954) Continuous inspection schemes. Biometrika, 41, 100--115.
       3. Gan, F.F. (1991) An optimal design of CUSUM quality control charts. Journal of Quality
         Technology, 23, 278--286.
       4. Lucas, J.M. (1976) The design and use of V-mask control schemes. Journal of Quality Tech-
         nology, 8, 1--12.
       5. Hawkins, D.M. and Owell, D.H. (1998) Cumulative Sum Charts and Charting for Quality
         Improvement. New York: Springer-Verlag.
       6. Kemp, K.W. (1961) The average run length of the cumulative sum chart when a V-mask is
         used. Journal of the Royal Statistical Society B, 23, 149--153.
       7. Bissel, A.F. (1969) CUSUM techniques for quality control (with discussion). Applied Statistics,
         18, 1--30.
       8. Montgomery, D.C. (2005) Introduction to Statistical Quality Control, 5th edition. Hoboken,
         NJ: John Wiley & Sons Inc.
       9. Barnard, G.A. (1959) Control charts and stochastic processes. Journal of the Royal Statistical
         Society B, 21, 239--257.
   389   390   391   392   393   394   395   396   397   398   399