Page 234 - Lean six sigma demystified
P. 234

212        Lean Six Sigma  DemystifieD


                 Process or Technology


                        I was on a plane from Denver to Knoxville to train a hospital on control charts
                        when I opened up the American Way magazine and found an interview with Larry
                        Ellison, CEO of Oracle, the second-largest software company. The article tried to
                        show that using its own software helped Oracle save $1 billion dollars, but Ellison
                        said something even more important: “The way you get quality is to define a set
                        of processes and procedures and make sure they are implemented everywhere.”
                          I was stunned! Here’s a tech-CEO saying the key was consistent processes.
                        And what he said next resonated with my two decades of software develop-
                        ment and maintenance experience: “Before we could automate anything, we
                        had to standardize the new processes we would need. It meant simplifying and
                        modernizing every procedure.”
                          “People ask the wrong question when they automate a company or process:
                        Will this bunch of software allow us to [do] things the way we [do] them
                        today? The right question is Will this bunch of software allow us to [do] things
                        the way we should do them?
                          After I graduated from the University of Arizona with a B.S. in Systems
                        Engineering (the high art of optimizing systems), I got hired as a COBOL pro-
                        grammer for a telephone company. There I was tasked with writing programs
                        to automate existing manual processes that were so cumbersome and error
                        prone that I often wondered what we hoped to gain by automating them.
                        Here’s what I learned: When you automate a poor process, you make it difficult
                        and time-consuming to change. Things you might have changed on the fly now
                        had to go through screening, prioritization, requirements, design, code, and test.
                        Most changes took months, even years.

                          Years later, it seemed we were still doing the same things, but even dumber
                        stuff. If an existing system caused too many errors, we’d write a mechanized
                        system to fix the errors caused by the first system because the first one was
                        deemed too complex to fix! There were systems that fixed addresses on outgo-
                        ing bills (150,000 per month were undeliverable). Why didn’t we go back into
                        the service order system and prevent the input of incorrect addresses? It might
                        slow down our service reps. Silly huh?
                          So, if you want to maximize the benefit of your new information systems,
                        use Ellison’s and my advice.

                          1. Simplify, streamline, or reengineer your processes first.
                          2. Then choose or build a system that reflects the streamlined flow, not the old
                            flow.
   229   230   231   232   233   234   235   236   237   238   239