Page 467 - Design for Six Sigma for Service (Six SIGMA Operational Methods)
P. 467

Theory of Constraints  425

        The theory of constraints also believes that the balanced capacity may not
        be a good approach for maintaining high throughput. The most important
        issue is to balance flow. This is illustrated by Example 12.4.

            Example 12.4: Capacity Balance versus Flow Balance
            Table 12.1 describes a sequential production process with five workstations.
            Specifically, the incoming job is first worked on station 1, then on each of
            stations 2 to 4, and finally on station 5. The second column in Table 12.1 lists
            the current task completion times for each workstation. Clearly, workstation 4
            takes the longest time to complete its work, so it is the slowest workstation and
            the bottleneck of the whole production flow. A proposed improvement plan is
            improvement option 1, which balances the capacity of each workstation to be
            53 seconds, in terms of work completion time. This is the balanced capacity
            approach. However, from the viewpoint of the theory of constraints, option 1 is
            not the desirable approach. Because there will always be fluctuations in the
            production that will affect the actual capacity of the process steps, if an
            upstream process is clogged, it will immediately become a bottleneck for the
            whole production system, as illustrated in Fig. 12.10. Improvement option 2 is
            a better approach, because it gives downstream process steps progressively
            bigger capacities so the clogs are less likely to form (Fig. 12.10).


        12.3.4 Elevate the System’s Constraint
        After identification of the constraint, efforts should be made to increase
        the capacity of the constraint. For a policy constraint, a low cost or even a no-cost
        fix is possible. For a resource or material constraint, additional investment is
        usually needed. Elevation of the system constraint will usually increase the
        throughput. However, when the old constraint (the weakest link) is gone,



        Table 12.1 Workstation Task Completion Times

                        Work Completion   Improvement   Improvement
          Workstation   Time (Seconds)   Option 1     Option 2
          1              45              53            59

          2              40              53            56
          3              60              53            53
          4              70              53            50

          5              50              53            47
          Total time    265              265          265
   462   463   464   465   466   467   468   469   470   471   472