So you have planned out some improvement for an existing process, product or service but you want to ensure that it is a success.  One of the best ways to make sure improvement goals are a success is to plan for possible failures.  FMEA or failure mode effects analysis will do just that for you.  FMEA is a method of identifying possible failures in a process.  As the name implies "failure modes" hints that it will help us to understand the possibilities of failure.  Sound like a valuable activity?  It certainly is,  so how is it performed?
  Before we get started with an overview of FMEA you need to know that there are a few different FMEA methods.  One is for process oriented analysis.  You may have heard it referred to as PFMEA.  The other type is used in the design or redesign of a process, product or service this type of FMEA is referred to as DFMEA.  Today we will not focus on either one specifically.  Remember you can find templates here.
How to Perform FMEA

1. As with almost any other project or initiative you need to define the details of the project.  This would include such elements as the objective of the activity, estimated timeline, history, resources you might need and why exactly you are undertaking the initiative.  A project charter can be used to lay out and define the the initiative.
2. Now that you have defined your project and assembled together a cross-functional team of SME's we can begin gathering measurements and understanding the current state of the process.  While there are a few different methods of understanding the current state or as is process or product the absolute best method is usually to map the process out.  However if there is no process in place using a quality function deployment approach may be helpful.  Tools like the house of quality can help you to ensure you are aligning your abilities with the needs of the customer.  
3. After the team has gained a good understanding of what is needed or what exactly is going on you will be ready to begin identifying potential modes of failure for each step in the process.  The current state map should have identified initial issues for you to focus on.  These potential failures should be listed next to the appropriate process step in your template. Once you have your potential failures link them up with how they will impact the customer. Try to keep in mind answering the question "what goes wrong when the failure occurs and how severe is it?"  This answer combined with the impact on customer requirements will give you your first score which is the severity rating of the potential failure.
4. Having now listed potential modes of failure and established an agreed upon severity score the team can now turn its focus to the actual causes or root causes of the failures.  
5. The next step in our FMEA method is to identify and rate the frequency of the potential failures occurrence.  This is an important part of the fmea process because it helps us to understand where we will focus our corrective activities.  The risk priority number that fmea establishes is based on severity, frequency and the level of detection or how well we can detect and prevent failures prior to reaching the next customer (internal or external).  One very effective way of seeing level of occurrence or frequency is through the use of check sheets and pareto analysis.
6. After a frequency score has been documented we need to document the current controls that are in place.   These controls are placed in the tab labeled "current controls."  These could be any realm of activities that the department or organization currently uses to prevent failure from happening or risk from escalating.  One thing to note is that some process steps may not have any control in place, these boxes remain blank.
7. Once the current methods of control are documented we will again attach a score to the control methods.  Although criteria varies from organization to organization ultimately you will be scoring detection based on how well the existing controls either prevent the failure from occurring or detect it.  
8. The next step is to calculate your RPN or risk priority number.  If you are using a template (hopefully) the number should populate automatically, but if you are not you will need to multiply your severity score by your occurrence or frequency score and by your detection score (S x O x D = RPN).
9. Finally we are ready to recommend actions that will reduce the possible failures severity, occurrence and improve your ability to detect or prevent the possible failure.  These recommendations should have an owner attached to the activity and should provide status as to what is happening with the recommendation.
10. Once the team agrees on the action plan move forward by implementing the agreed upon activities.
11. Last but not least calculate the results of your action plan with the RPN.  One best practice is to continue to set improvement goals targeting improvements or reductions in the potential failures.

  Keep in mind that fmea is meant to identify and completely understand a product or processes potential failure modes along with how they impact the customer.   FMEA will also establish a level of risk associated with the identified failure modes, effects and causes and will assist in prioritizing solutions and narrowing down a seemingly long list of actions.  But like any other improvement project you will need a team of experienced and skilled professionals to properly gather "failure modes" and "analyse the effects."    

  In the first three phases of the REDUCE methodology, required data is established, the value stream is engaged for a deeper understanding and internal and external activities are differentiated or separated as defined in the SMED system.  The first two phases are largely gathering information and analysis while the second begins to separate tasks and define what can be done while the machine is running and what can be done while the machine is stopped.

​  Of all the steps in a setup or changeover reduction, separating internal and external activities may in fact be the most important as well as the simplest steps you can take.  In the most general sense, performing activities like preparing tools, kits and materials as well as transporting items while your line is running can reduce setup time by as much as 50%.

Before improving anything in a setup, we must first consider the current state or common problem areas of what we are improving.  In the case of setups, changeovers or turnarounds there are traditionally 4 basic steps where problems and opportunities fall.  Those steps are:

1. Preparation.
2. Removing and Mounting Tools and Parts
3. Calibration.
4. First Articles, Trial Runs and Adjustments.

  If you travel into any machine shop, office or hospital facility in the world guaranteed you have seen a changeover before.  The changeover occurs in between the last good part or service until the next good part or service.  Some examples would be:

1. Changing tooling.
2. Patients service completion.
3. Changing paper in a printer.
4. Taking one batch of cookies out of the oven in order to start a new batch.  

This "in-between" time is usually spent performing wasteful activities which often result in long changeovers that inevitably eat up value added time.

  Almost every project will include brainstorming of some kind at some point in the project lifecycle.  Brainstorming sessions can be a powerful gateway to unlock solutions, make issues visible, prioritize actions and bring experienced minds together.  When individuals come together as a team, innovative ideas can be born.  One of the struggles of being a part of a powerful and productive brainstorming session is that they generate many great ideas and often reveal a large amount of issues.  This can leave a group feeling overwhelmed.  Often times, the wide array of ideas can be hard to organize, understand, validate and act on.  Worse yet, many members of a team might leave feeling invalidated, unheard or completely shut down. 
  

  The fishbone diagram, cause and effect, or ishikawa diagram is one of the most common quality tools used today.  Best known by its resemblance to a fish's body the fishbone diagram is used to show the many possible causes for an effect. The tool is used to help coordinating brainstorming in an effort to discover root causes.

​  The first day on the job is always an exciting experience for any new employee.  A new chance to show your skills, meet new people and grow in a new organization.  Excited to learn something new in training you get thrown to the wolves and here the phrase "you will figure it out."  That can be a bit scary to say the least.  Surprisingly as you start figuring out everything has procedures connected to it.  Three days later and you have mastered erp, assembly and every office function in the organization.  A bit surprised by this you're off to a great start!

   Have you ever gotten into a discussion of why? because. Why? because.  Last weekend I had the great pleasure of this discussion for what seemed to be the first time.  I never really put much thought into it but asking why never really was a difficulty for me.
  Oftentimes when we want to use the 5 whys to drill from issue to root cause we end up in a pattern of why, because. Why, because.  Or, we conduct a long analysis only to find out that none of our levels of causation matched at all.  One of the best known examples of a 5 why analysis was performed by the master Taiichi Ohno.  He used the example of a welding robot stopping in the middle of its operation.  Like a sensei does he naturally went from initial issue to root cause with almost no difficulty at all.   So, how do we begin developing this level of mastery with regards to root cause analysis?  Here are a few important things to keep in mind when looking for the ROOT cause.