The big buzzword with the government right now, in terms of quality assurance and process, is Lean Six Sigma.
Which is actually two techniques that have been mooshed together.
Lean is a process by which you eliminate (or at least reduce) all the unnecessary parts of a process. It sounds obvious, but it can lead to some really unexpected results. A lot of things that you would think are necessary are, in point of fact, not -- even if they are needed. The famous example is a car assembly plant that Leaned its processes and discovered that every turn of a screw except the last one is unnecessary. So they shortened the bolts they were using, cutting down on both time needed to assemble the cars and cost of materials.
Six Sigma is a defect-reduction strategy originally conceived for parts manufacture that employs some really high-level statistical analysis to your data. (The name comes from that analysis -- a sigma is a measure of standard deviation; if you have six of them, then that means that you have a defect rate of less than one in a million.) Whether you actually want to achieve a level of six sigma depends on the process you're looking at: if you're talking about, say, pacemakers or airplane landing gear, then yes, you really do want a failure rate that low. Software that doesn't hold people's lives in the balance, however, should probably not strive for quite that level of perfection -- eliminating defects gets more and more expensive the further down that path you go, so there's a point at which improvement just isn't worth it.
Smooshed together as Lean Six Sigma, you get a system for improving (usually shrinking) processes by statistically analyzing their defects. It's a nice tool to have in the bag, but like all tools, it's not always the best one.
But the government, like large corporations, isn't really interested in fitting the right tool to the job. It's interested in being buzzword-compliant, and Lean Six Sigma is the current buzzword of choice in the quality community.
So I've been working, in my copious free time (hahaha!) over the last six months, on a project which will certify me in Lean Six Sigma. Both so that I can add this tool to my kit, and so that I have something to tack onto my resume and make me worth a little more to the company. The project is one of those where Lean Six Sigma doesn't really fit very well, but well, the problem we had was a nail, so this set of calipers begins to look awfully like a hammer. I made it work, if I do say so myself. I even managed to make our data look reasonable (we got our sigma from 2.3 to 2.8, which is about a 10% improvement in defects -- not bad, really).
My final presentation for the project is this morning, after which I'll have my certification and can update my resume. Go, me!
No comments:
Post a Comment