Expecting the Unexpected

February 21, 2008 at 10:41 am | In Six Sigma, quality | Leave a Comment
Tags: ,

Most of us do pretty well in handling the various crises that come our way, especially those that are similar to those we have seen before. If I get pulled over for speeding by a highway patrolman, I probably won’t panic, because it’s happened before. I smile and the officer apologetically, and hope she lets me off with a warning, but am prepared to accept the consequences. If the battery is dead and my truck won’t start, I shrug and mutter a few imprecations, but from past history I know what steps to take in order to get my truck back on the road.

So it is in the quality world. The question is not, “will something go wrong?” The questions actually are, “when will something go wrong?” and “do I know how to respond?” Knowing what can go wrong and planning for it can help you avoid some problems, and can also help you mitigate the issues that do arise.

One of the best tools for “expecting the unexpected” is the “Failure Mode and Effects Analysis,” or “FMEA” for short. The FMEA is a systematic approach to stepping through each part of a process or a design, trying to anticipate what could go wrong, and developing a plan to address it.   

For example, if I am making spaghetti, part of the process is to boil the pasta. In the FMEA approach, I would evaluate what could go wrong in the “boil the pasta process.” I would identify some potential failure modes:

  1. Not enough water in the pot
  2. Not enough heat under the pot
  3. Not enough time at the boiling point
  4. Too much time at the boiling point

I would then review each of these failure modes for what the effect of that failure would be:

  1. Not enough water in the pot – pasta will clump and not cook well
  2. Not enough heat under the pot – pasta may never come to a boil
  3. Not enough time at the boiling point – pasta may not cook completely

Too much time at the boiling point – pasta may overcook and be mushy

Each of these modes and the resultant effect are then analyzed to determine the severity of the failure, the likelihood of its occurrence, and the probability that the occurrence could escape detection, ranked on a scale of 1-10, with the higher number representing the worse case (obviously, the numbering scheme can be somewhat subjective…):

  1. Not enough water – Severity=4, Likelihood=4, Escape potential=3
  2. Not enough heat – Severity=4, Likelihood=2, Escape potential=1
  3. Not enough time – Severity=4, Likelihood=2, Escape potential=3
  4. Too much time – Severity=4, Likelihood=4, Escape potential=4 

        The next step in the FMEA process is to multiply these numbers. The resulting product is called the “Risk Priority Number,” or RPN. The higher the RPN, the more significant the failure mode.  From the example above, my highest RPN is 64 (4*4*4), from failure mode #4. So I would want to be sure to develop a plan to minimize the risk by perhaps monitoring the amount of time the pasta has been cooking, or performing an inspection of the pasta at a set point to determine if it is ready.

       The FMEA process isn’t perfect, and it really can be a time-consuming, pain-in-the-butt process to go through, especially for a complex product or process. It also doesn’t stand alone in the crisis avoidance process, but should be combined with other tools to analyze the situation at hand.
       But it is the wise quality manager, Grasshopper, who can use the FMEA process to her advantage. You can learn more about the FMEA process at the American Society of Quality website.

Letters to my younger self, or “Things I wish I had known”

February 6, 2008 at 5:26 pm | In Six Sigma, quality | Leave a Comment
Tags:

 Today I start a semi-regular look at the art of being a quality professional. This will be framed as information that I could have used as a young quality engineer and then quality manager. I learned my craft through on-the-job trial and error, with mentors who had learned it the same way. As I went along, I learned new lessons that could have made my life easier had the knowledge come earlier.  

At least, I think it would have been easier. It’s possible that I didn’t know enough to have used the information at an earlier time. Whatever the case, I will try to look at the art of the quality life, at least as seen through my own filter.

 

I wish that I had understood what “quality” is, back when I first came across it. When I was a young production planner with SCI Systems, the quality folks were my enemies. I saw them as the road blocks to successful product shipment. I had a goal: I was supposed to ship as much product on time as I could, and earlier was better. Every delay caused by an inspector or a quality engineer just meant I had to work a bit harder to get my product out.  I thought the quality team was arbitrary and temperamental—and I was right in many ways.

 

The problem was, I didn’t know enough about the ways my product was supposed to work, so I couldn’t appreciate the potential effects of quality issues with my products. The funny thing was, I was working on items that could save lives: voice alert systems for military aircraft, data acquisition controls for missile tracking satellites, computers that controlled the space shuttle. Any defect that caused one of my products to fail could kill a pilot or an astronaut. And while I knew that in an abstract way, it was hard for me to see where this faulty solder joint or that faulty coating could cause disastrous failure.

 

In this case, a definition of quality was easy: “Quality” was the absence of a defect.

Unfortunately, realizing that definition was difficult. The system in which I worked relied upon visual inspection to identify potential defects. This was followed by extensive tests of the product to verify that it worked properly. The proof of the quality pudding was hard to find, and I had no one to explain it to me. On top of that, my thoughts that sometimes the quality group acted in a capricious manner were accurate: the supervisor of the inspection team sometimes used his group to reward or punish those manufacturing managers whom he felt had interfered with his operation, or had erred in some way. The wispy nature of the acceptance specifications allowed him to tell his employees to “tighten” or to “loosen” their inspection approach.

 

So from this I get my first quality truth that I wish I had known:

Product expectations should be well documented and clearly communicated.

In this case, “quality” was what Bobby said it was, regardless of the specific requirements.

 

There are several text-book definitions of quality. One of the most common is that quality is “the characteristics of a product or service that bear on its ability to satisfy stated or implied needs.”

Or how about, “a product or service free of deficiencies.”  Quality guru Joseph Juran defined quality as “fitness for use,” while Philip Crosby said that quality is “conformance to requirements.”

 

If you find this confusing, join the club.

Blog at WordPress.com. | Theme: Pool by Borja Fernandez.
Entries and comments feeds.