Solving Special Cause & Common Cause Problems

Retired course

This course has been retired and is no longer supported.

About this lesson

Special cause problems should be resolved first in order to achieve process stability. Then common cause problems are addressed to reduce process variation.

Quick reference

Solving Special Cause & Common Cause Problems

Special cause problems should be resolved first in order to achieve process stability. Then common cause problems are addressed to reduce process variation.

When to use

These problem solving guidelines should be followed when the team starts to create solutions in the Improve stage of a Lean Six Sigma project. If both types of causes of the problem are present, all special cause solutions should be implemented and a new process baseline measured before starting to implement common cause solutions. In that manner, the true effect of the common cause solution can be measured and understood.

Instructions

Special cause root causes and common cause root causes are fundamentally different and require fundamentally different approaches for solution. Special cause root causes are unique events. Their occurrence is random and their magnitude is unpredictable. In many cases, they can be totally eliminated if the enabling factors are controlled. Common cause root causes are always present. The magnitude of their impact is random within some predictable boundaries but their occurrence is continuous. They can only be controlled or eliminated by making a change to the basic characteristics of the process or system.

When special cause root causes are present, they should always be solved first. Their solution could impact the overall process performance. Once their solution is in place, the process performance should again be assessed and if there is still an issue, then common cause aspects of the process must be addressed.

Solving Special Cause Problems

Special cause solutions will either consist of actions to control or eliminate the factors that enable the special cause condition to occur or by monitoring for the special cause occurrence and establishing workarounds for immediate implementation when a special cause is detected. Essentially a preventive action approach and a corrective action approach.

The preventive action approach of eliminating the enabling factors usually requires a 6 M (material, machine, manpower, methods, measurement, Mother Nature) analysis of contributing factors to decide which ones are controllable. The corrective action approach often requires the implementation of a monitoring system and the prepositioning of resources or procedures in the event of a special cause occurrence. Trying to solve a special cause problem with common cause solutions is normally ineffective because the random event will still occur randomly.

Solving Common Cause Problems

Common cause solutions must follow one of three approaches:

  • Either a major change to the existing process to reduce the common cause variation,
  • A change to a totally different process with inherently lower common cause variation,
  • Or in some cases, recalibrating the existing process so that the common cause variation occurs within the allowable performance tolerances.

In each case the variation is continuing, it is just that the magnitude or distribution is no longer an issue. Trying to solve common cause problems with special cause methods will often increase the amount of inherent variations since the process monitoring system is now chasing and over-correcting for normal random variation.

Hints & tips

  • There is a tendency for the organization to want to treat everything like it is a special cause and start to chase normal random variation. Take the time in the analysis phase to understand the type of causes you are facing and then take the appropriate action.
  • Special cause monitoring solutions should be periodically tested to ensure they are able to detect the occurrence of a special cause.
  • Work with process operators and supervisors to ensure their expectations for process performance is reasonable. If solving special cause with a monitoring approach, they should expect that the special cause condition will occasionally occur. When solving common cause problems, they should know that there will still be variation in the system, but that the variation is now at an acceptable level.
Login to download
  • 00:05 Hi, I'm Ray Sheen.
  • 00:06 When determining a solution approach, you need to know if you're fixing a special
  • 00:10 cause problem or a common cause problem.
  • 00:13 The solution strategy is quite different
  • 00:15 depending upon the nature of the root cause or causes.
  • 00:20 Let me explain the strategy to use for each of those conditions.
  • 00:24 You should always address special cause problems first.
  • 00:27 You can't begin to solve common cause problems when the process is unstable and
  • 00:31 unpredictable.
  • 00:33 So first we go for stability and that means eliminating special cause problems.
  • 00:38 This is the approach embedded in many of the basic problem solving methodologies.
  • 00:42 They assume every problem special cause and try to solve it.
  • 00:46 The approach is easy to understand.
  • 00:48 Analyze the problem until you find the root cause and then fix that root cause.
  • 00:53 Either eliminating it, or screening for that root cause occurrence factors and
  • 00:57 taking appropriate action when they are present.
  • 01:00 Once the special cause problems are resolved and
  • 01:03 the process is stable, if there is still a performance problem,
  • 01:07 then it is time to go after the common cause issues.
  • 01:10 Recall the common cause variation is the normal variation inherit in the process.
  • 01:15 So to reduce common cause variation,
  • 01:17 you can never totally eliminate it .You must make a change in the basic process.
  • 01:22 So once the analysis is determined, the contributing factors or
  • 01:25 drivers of variation, your approach should be to determine which factors you
  • 01:29 can reasonably or realistically control or change.
  • 01:32 In other words, what types of solutions are feasible?
  • 01:35 And then select the change or
  • 01:36 changes needed to reduce variation to an acceptable level.
  • 01:40 Let's look a little deeper at what it takes to fix special causes.
  • 01:44 Of course, the actual solution will depend on the actual problem.
  • 01:48 But there are two approaches you can take to create a solution for
  • 01:51 special cause problems.
  • 01:53 The first approach is to control or eliminate the enabling factors.
  • 01:57 When you do this,
  • 01:58 the situation that creates a special cause problem is prevented.
  • 02:02 For instance, let's say your special cause was due to the change to the new vendor
  • 02:05 and how their component performs in your product.
  • 02:08 Eliminate the new vendor, going back to the old one and
  • 02:11 the special cause problem is solved.
  • 02:14 To use this approach, you must have completed a thorough analysis of
  • 02:17 the problem and determined the root cause or causes.
  • 02:20 Then decide which of these causes you can impact.
  • 02:23 That will usually mean doing something in one of the M categories of material,
  • 02:28 manpower, methods, machines, measurements, or mother nature.
  • 02:32 Design your change or improvement and put it in place.
  • 02:36 The goal is to eliminate or prevent the root cause from occurring.
  • 02:41 The other approach used is when the root cause is not controllable.
  • 02:44 Let's say our root cause was due to a power outage.
  • 02:47 That is an infrequent occurrence.
  • 02:49 It's unpredictable and uncontrollable.
  • 02:52 While it can't stop it, I can put a place, an instant warning signal and
  • 02:56 the actions needed to compensate for this root cause problem.
  • 03:00 So in my example, I get a back up generator.
  • 03:03 Whenever the power goes out, the generator automatically starts.
  • 03:07 I didn't prevent the root cause, but I have a design that recognizes and
  • 03:10 compensates for it.
  • 03:12 This approach requires an active monitoring system and
  • 03:15 I recommend that it also include a periodic test of that system.
  • 03:20 Okay, now lets switch and look at the approach to use for common cause problems.
  • 03:25 Now as we've said many times, there is always going to be common cause variation.
  • 03:30 The problem is when the variation causes our product or
  • 03:33 process to miss its performance goals.
  • 03:36 To reduce or modify the inherent common cause variation,
  • 03:39 will normally require a fundamental change to the system or process.
  • 03:44 There are three tactics that we can use to accomplish this.
  • 03:47 One tactic is to work with the existing product or
  • 03:49 process to reduce the sources of variation.
  • 03:52 If there's only one or
  • 03:53 two sources, then change how those parts of the process work.
  • 03:57 Think through the six Ms again when considering this change.
  • 04:01 If there's several interactive factors, you'll probably need
  • 04:04 to do a design of experiment study and more about that in another module.
  • 04:08 Of course, many times it's just easier to switch to a different process or
  • 04:12 system altogether, than trying to fix the existing one.
  • 04:15 That's especially true if the current system is old or uses outdated technology.
  • 04:20 This will likely take some re-engineering or system design.
  • 04:24 Again, a design of experiments can be helpful here.
  • 04:27 The third approach is to shift the existing process performance so
  • 04:30 that the amount of inherit common cause variation
  • 04:33 all falls within an acceptable level of process performance.
  • 04:37 Many processes in systems have a tendency to drift over time.
  • 04:40 The system may just need to be recalibrated.
  • 04:43 This is the easiest of the three approaches, but it's not always variable.
  • 04:47 So when designing you're improvement, be sure you are using the correct approach.
  • 04:52 Trying to fix a common cause problem with a special cause solution will normally
  • 04:56 just make the inherent variation larger and the problem even bigger.
  • 05:00 And trying to fix a special cause problem with a common cause
  • 05:03 solution will not prevent the power outage from ever happening again.
  • 05:08 I spent a lot of time discussing the difference between special cause and
  • 05:12 common cause in earlier modules.
  • 05:15 The reason was so that the improvement approach selected at
  • 05:18 this point in the project, will be one that is truly effective for the problem.

Lesson notes are only available for subscribers.

PMI, PMP, CAPM and PMBOK are registered marks of the Project Management Institute, Inc.

Gift this course