Page 245 - Lean six sigma demystified

Chapter 6 Tr an S a C T iona L Six Sigm a 223

correction people, and service order entry personnel. To attempt to do all six at
one time with the same people would have been foolish. The errors required
different subject matter experts, and the root causes were too different. By
restricting ourselves to just one error type per team, we were able to find the
root causes in just one half-day meeting per team.
To prepare for the meeting, we printed out 50 to 100 examples of each error
(it helps to have more than 30 when you start, because you’ll find some that
don’t actually belong in the category). Then, we gathered around a computer
terminal and investigated each error.
1. Using all of the online systems, we investigated the root cause of each
rejected transaction. Again, we restricted ourselves to analyzing just one
transaction at a time. We had one person who really knew how to drive all
of the information systems look up the transactions and all related infor-
mation (e.g., customer records).

2. As the team reviewed all of the information and agreed on the cause of
the rejected transaction, I kept a stroke tally for each root cause (Fig. 6-5).
Gradually, as we looked at more and more transactions, a pattern would
reveal itself. Sometimes it only took 25 transactions, sometimes it took 50,
but a pattern would reveal itself clustered around one or more root causes.

Figure 6-5 • Check sheet of cause data.