Industrial Problem Solving

Sample Size for Process Capability

2010-11-30T21:45:00.000-08:00

It's important, no, necessary that new processes be developed with high process capabilities because a process that operates close to a specification limit is an expensive process. If you can develop the process such that it is six sigma capable, you probably won't hear much from it ever again. And that's great.

A colleague and I were reviewing a Cpk study at work recently where the result was 1.2 and the sample size was 25. Note that a Cpk statistic is a point estimate and it has confidence limits. The larger the sample, the narrower the limits. So if we put 95% confidence limits around the calculated 1.2 with n = 25, we get limits of 0.84 and 1.56. That's quite a risky process if it is as low as 0.89.

Sample sizes for process capability studies aren't well understood. I recommend 100+ and the sample needs to be taken over several runs, not all at once. Here's a formula for calculating the confidence interval for a Cpk statistic. If you are responsible for a new process, make sure your calculated Cpk statistics include the confidence limits.

Measurement System Variability

2009-04-05T21:01:00.000-07:00

I am working with a client who has experienced an increase in variability in the thickness of their product. I was able to get my hands on just over a year's worth of recorded data and the plot was quite revealing. the chart is attached - please ponder it for a moment.

Their process has some sources of variation between batches but look at how the data points appear after June 2008 (middle of the graph). At this point in time, someone replaced the measuring tool. The older digital indicator was able to provide five decimal places whereas the new digital indicator provided only three. This shows up as an increase in the short term variability in the data.

The measurement system is one of the sources of variability in a data set and in many cases, it is the largest source. It is the easiest to correct and easy to prevent.

I took the original data set and plotted the differences between each data point. Here, it is clear that the new tool increases the variability of the measurement system.

The wrong tool can be very costly - rejecting good product just because of measurement error.

Measuring Tools

2008-11-12T13:33:00.000-08:00

I saw a video clip for a high technology organization that described the importance of precision in measuring their parts. But the video showed a fellow using a dial indicator. It seems that a lot of people are not aware of the variability in this rather simple tool. If precision is really important, you don't use a tool that introduces 0.005" to 0.010" variability to the measurement process.

Measuring tools are used to assign a number to some feature of a part. In assigning this number, the tools vary. Sometimes a little and sometimes a lot. We can all appreciate the fact that if a lot of people measured the same part with the same measurement tool, there would be some disagreement in the responses. And if the same person measured the same part with the same measuring tool many times over, there would still be some disagreement. These are the sources of variation in the measuring process.

It is important, indeed necessary that the variability (repeatability and reproducibility) introduced by the measuring tool be known. Otherwise you make too many of the two types of errors - you accept product that should be rejected and you reject product that should be accepted.

Data Ends the Debate

2008-10-28T22:34:00.000-07:00

A few years ago we were having a discussion on who's product caused a problem. A wood products manufacturer created the box and a plastics supplier moulded the cap. The cap and box did not fit very well. The wood products guy said the plastic caps were distorted and each one was different. His stance was "not my problem." The plastics fellow said the plastic parts would distort out of the mould but not enough to cause the problem we were seeing. Going nowhere.

A couple of fellows took a load of the wood boxes and started measuring them. One particular dimension was tied to the problem. The Excel chart shows the distribution of this dimension over dozens of boxes. Clearly, one of the holes was being drilled too far from the edge.

With data, the debate was over and we found the root cause quickly.

A simple example but it is not common. Many of our problem solving efforts fail because we just don't have the right data.

Why We Don't Solve Quality Problems

2008-10-16T23:42:00.000-07:00

The reason why problems do not get solved is that the decision makers do not have the right data to work with. We sit in a meeting room with a team of people discussing the problem and all we get are opinions. Mostly, they're wrong. Opinions are formed in the absence of data.

Other times we have someone in the discussion who may have authority or who may just be assertive. They will present a good case but it's just another opinion. Assertiveness isn't always a good thing. Organizations have spent lots on solutions that weren't.

The name we give to a problem often suggests a root cause. One example comes from a high volume packaging line. A particular problem was termed "feed problem" which suggests that the feeder was the root cause. But the real problem turned out to be orientation of the item. Because the center of gravity was not the center of the item, some items wobbled. Once this was identified and shown to be true, the fix was easy. But for months, the "feeder problem" was not solved because people were looking at the conveyor system and not the item on the conveyor.

Labelling a defect incorrectly is common and it causes delays.
Another reason why we misdiagnose problems is because we use the wrong sample size. This happens a lot. If you find two defects and trace them back to supplier A and not supplier B, we all know that isn't enough confidence to call it a supplier defect. But what sample size is right? How many do we need to be 95% confident? 99% confident. Most people don't know how to determine the required sample size. And that also causes delays as solutions are implemented that don't solve the problem.

There are a lot of reasons why we are not so effective at problem solving and we've all experienced solutions that proved to be wrong. Collecting evidence and using statistical methods can make your problem solving efforts a lot more successful.

Productivity in Problem Solving

2008-10-14T22:42:00.000-07:00

I've been a black belt for many years now and have been involved with a lot of problem solving assignments. These projects require charters, meetings, checklists and reports. You tend to have conversations about the problems with dozens of people, one at a time. There's a lot of routine work involved, as with any project.

But nearly everything I learned about the problem came from analysis of data. Collecting evidence about process behaviour, or running a designed experiment provides new insights almost every time. So if you're tackling a difficult problem, you need to be on the floor, or in the lab, measuring parts and collecting data.

Solving quality problems is a scientific endeavour. If there's no data, it isn't science.

Know Your Process

2008-09-18T21:24:00.000-07:00

In an injection molding factory, most parts will have a mold and cavity ID molded on the back of the part. In this way, when there is a quality problem we can quickly know where the parts came from. Not many processes are set up with such convenience.

I was able to work with a high tech company that had a high reject rate, 2.7% coming from a long and complex process. There were several workstations for each process step so the number of possible paths that each product could have taken was in the hundreds. This is a problem in many high volume manufacturing companies. Not many can run one-at-a-time serial production.

One of the process steps was a heating cycle. Thousands of parts were soaked at an elevated temperature for 24 hours. When they were removed from the ovens, the parts were handed to several people for more work before making their way to a final test station. It was here at final test where the 2.7% failed.

In this assignment I had each product labelled (Sharpie marker) with the shelf number from the oven (from 1 to 12). We were very surprised to see that nearly all failures were labelled 1, 2 or 3. The bottom three shelves in the oven were responsible for almost all the failures! The temperature in the oven dropped of sharply near the bottom and these parts were not getting processed adequately.

There were so many opinions on a possible root cause and the real root cause wasn't on the list! Collecting data may be tedious but there's no other way to know your process.

An Expensive Opinion

2008-09-04T22:43:00.000-07:00

Over a year ago, I was involved in a project to solve a weld problem. This company performed thousands of spot welds every day and some (approx one in 300) just didn't take. The failure wasn't obvious unless you tugged at the weld and even that wasn't a reliable test. But a poor weld could be serious if it failed in the field so this had to be solved.

A department manager was convinced that a current monitor would take care of things. In his opinion, the weld fails when the current drops below some critical number. So when the current is too low, the operator discards that last welded part. My argument was that there are many things that could cause a weld to not take; part cleanliness, surface geometry, applied force, squeeze time and more factors that are part of the welding process. The project required much more effort and some data. But he was adamant and he told us that this was what they do in the automotive industry. So we backed off and accepted this as a problem nearly solved.

A year later, the process has a reject rate of one in 300.

Collecting evidence to learn about a root cause takes so little time. Solving a problem based on an opinion can set you back months.

Clue Generating Tools

2008-09-04T21:56:00.000-07:00

A client had a high reject rate on one step in the manufacturing process. This is a complex item with several parts and at this stage, near the end, rejects were costing just over $15,000 every month. The defect was causing a leak at a particular location on the part and many efforts over many months were not successful.

I was invited to assist in this problem solving project and in my line of work, I never accept things as facts without evidence. I've seen too much waste. We rented a high speed camera to record what was happening at the point in the process where the defect was created. To everyone's surprise, we saw the leak start in a different location. It was centrifugal force that moved the fluid to the spot that everyone had thought was the leak.

The high speed camera showed us where the problem started. Tools that enhance human perception can often provide clues.

Old Machinery

2008-08-27T19:28:00.000-07:00

We often blame the old machine when we have a quality problem. This old thing doesn't work like it used to. At some point, we end up buying a new machine. But the problem wasn't the entire machine. Some part of it had failed or worn out. The machine ought to have a much longer life when you replace worn parts. The trouble is, a lot of people give up the hunt for the elusive broken part of the machine.

A company I worked with was dispensing fine powders as a part of one of their processes. The reject rate was too high because the amount of powder dispensed was sometimes too much, other times too little. The problem was in the variability of the dispensing process. A new, half million-dollar machine was purchased to replace the old machine and for a while, both machines worked side-by-side. Now get this. The new machine was worse than the old one! More overweight rejects. More underweight rejects.

I was able to work with the company on this specific project and we determined the root cause of the old machine's variability problem. It was corrected and was producing over 99% quality product (up from 85%!) in a few weeks. The new machine was significantly different and required some pretty serious redesign work. My contribution was over and I moved onto other assignments but I hear from people still there that the new machine is in the corner. No longer needed.

Old machines don't fail. Parts of it wear. The root cause of many old machine problems turns out to be a hundred dollar part. It's well worth spending the time to figure this out.

Collect the Evidence

2008-08-27T19:11:00.000-07:00

All manufacturing companies have quality problems. These problems typically consume 15 to 25% of sales revenue. Problems of this magnitude must not be treated lightly. A sound, structured approach is necessary or the problem will just continue to bleed.

The first task is to meet with the smart people in the company and discuss what could possibly be going wrong. There's value in this because it may be a known problem, something that has happened in the past. But once you have checked the obvious, other ideas are just opinions and best guesses and they are likely to waste more time. It's important to keep in mind that the meeting can only bring up existing knowledge. If there's something new or unexpected happening, it's not likely to be identified in a brainstorming session. There are hundreds of things that could be going wrong but only one or two actually are going wrong.

All processes communicate their behaviour. The problem is that we do not speak the same language. We need to use tools to tap into this 'voice of the process.' In my experience, when you tap into this voice, you always learn something new.

As an example, if your process is a dispensing process, start taking before and after weights to better understand the dispensing process. Learn how the process varies from item-to-item, shift-to-shift, week-to-week. If you use ovens, use thermocouples to see how temperature varies from top-to-bottom, front-to-back and over time. I've used tachometers before to measure the speed of a moving conveyor. The belt speed was set to 30 feet per minute but the tachometer showed us that the belt was getting hung up somewhere and the belt speed would stagger a little. Not visible by eye, but measureable with the right tool. And the stagger was the root cause of a quality problem we were chasing. This was not identified in the initial brainstorming session.

Use measuring tools to learn about process behaviour. Evidence trumps opinion when you're solving quality problems.

Sample Size

2008-08-18T20:45:00.000-07:00

A manufacturer recently recorded 14 test failures out of a total of 318 products submitted for testing. This failure caused some panic and with it, some hasty decisions. Root cause analysis was performed in a meeting room. Wrong place. A decision was made to modify some tooling on the assumption that this was the root cause. With the new tooling in place another 30 samples were submitted and all passed the test. And there was much rejoicing.

The sample size was wrong. The problem still existed and there was a lot more product in inventory because the group celebrated early, producing a lot more units with both the modified tool and the real root cause.

The sample size chosen only provided 75% confidence, not enough for such an important decision.

There are many reasons why problem solving efforts fail. Knowing the minimum sample size required is one of them and it shouldn't be.

Quality Resources

2008-08-18T20:05:00.000-07:00

Most manufacturers that I have worked with over the years have a small band of Quality department employees. A Manager, one or two engineers, an auditor and a couple of technicians and inspectors. The output from such a department tends to be a fair amount of Quality administration such as ISO 9001 registration documentation.

But the job of each department in the organization is to prioritize economic performance and Quality administration doesn't cut it. The Quality resources must be responsible for problem solving. Not just identifying them and issuing additional documentation (CARs, NCRs, etc.) but actually resolving the company's top problems. Solving the top few problems in a company will put at least 10 points back to the bottom line. That's how to allocate your Quality resources.

Similarities in Quality Problems

2008-08-12T00:03:00.000-07:00

Manufacturers view their processes as unique and so they see their problems as unique. But when you step back from this focused viewpoint and see these activities from a cross-industry perspective, they’re very much alike.

I’ve worked in a lot of manufacturing companies and their processes are all the same. Raw materials are purchased according to schedule. Incoming inspection approves them for use. Parts are fitted, formed, welded, trimmed, heated, cooled, measured, assembled, tested, packaged and shipped against orders. Even at the detailed level of the processes we see many similarities. Vibratory feeders, ovens, conveyors, dispensers, laminators, presses, welders, adhesives, rotary tables, lasers and more. Similar processes show up in companies manufacturing very different products. Quality problems that show up in different companies also tend to be similar.

Develop your problem solving capabilities. When you are good at problem solving, you can make a huge difference in your company.

Solving a Quality Problem

2008-07-27T23:12:00.000-07:00

Most efforts at solving a quality problem start in the boardroom and unfortunately spend too much time there. This is where we brainstorm and throw our ideas on the table for review. The place to be when you're trying to solve a quality problem is on the factory floor where the problem is created. Like a detective trying to solve a crime, you need to be at the scene of the crime and collecting clues. This sounds obvious but it is surprising how infrequently people take this approach.

I was working with a group in China a few years back and there was a quality problem that caused significant numbers of their product to fail and they had not been able to determine the root cause. Upstream in their process was a 12-station rotary table that was used for swaging (forming metal). There were no labels on this rotary table identifying station 1 vs 2 vs 3 etc. So we labelled them 1 to 12 and then took items off the process and lined them up in several rows of 12. It was there right in front of us. Station #5 produced terrible product. The company management was impressed with how quickly we resolved a long-standing quality problem.

Brainstorming doesn't work. Gathering evidence does.

Quality Problems in Manufacturing

2008-07-21T18:46:00.000-07:00

In 2005, CME's 20/20 Magazine published results from a Management Issues Survey. The least satisfactory skill set among employees in Manufacturing was Problem Solving. All manufacturers have quality problems. Plenty of them. We tend to assign these to our technical employees assuming that they possess the required skills. Engineers will certainly have strong math skills and engineering intuition. But this isn’t enough. There are other elements of problem solving that need to be known such as nonparametric statistics, variation reduction strategies and components of variance techniques. Problem solving is a skill that can be learned. It’s important, indeed necessary, that manufacturers develop problem solving skills among their employees and dedicate a small team to doing nothing else.

Quantify and Rank Your Quality Problems

2008-07-10T22:45:00.000-07:00

A lot of folks tend to dismiss the Pareto chart as a simple bar chart with little value. But the pareto chart shows us the distribution of many phenomena including earnings and losses. The chart posted here shows athletes' salaries in June 2008. I chose this one rather than an industrial example to show the universality of the pareto principle. When we are talking about solving problems in manufacturing, knowing which problem ranks #1 is highly valuable. Here's why. The difference between successive entries on a pareto is alway a trivial amount except the top one or two. So your top problem, measured by cost, will outrank your second problem not by 2% or 5% but 50% or 100% or more. It is well worth collecting data on all your top problems to see which one is top dog.