Preventing Failures Leads to Success

I was assigned a project at the Armstrong St. Helens, Oregon plant. Our objective was to reduce production line changeovers from the historical average of 25 minutes to less than 15 minutes. This project had a capital budget, a corporate and plant team, and commitment to Kaizen (improvement) events. I was the most experienced in Lean and Kaizen on our team, so I facilitated many changeover reduction events and assisted with the tools to manage the overall project.

We utilized Failure Modes Effect Analysis (FMEA) to manage the overall risk to the project, facility, and personnel from all investments and activities. FMEA utilizes Risk Priority Number (RPN) to help teams understand risk. RPN is made up of three components, scored from 1 to 10:

1.      Likelihood of Occurrence – how likely is it that an identified failure will happen? Is it never going to occur (1) or could it happen no matter what (10).

2.      Likelihood of Detection – if the failure occurs, would anyone notice it? It is much worse to have a failure that is undetectable (10) than one that is easily seen (1).

3.      Severity – if there is a failure, how severe will the impact be from it? The worst is a fatal event (10). The best is when there is no chance for damage or injury (1).

The three components are multiplied together to get a final score. The higher the RPN score, the higher the priority should be placed on reducing or eliminating the risk. The top score is 1000. The lowest score is 1.

The team identified the most critical work on the project and how we would mitigate the risks to the work. In my experience, conducting an FMEA can be quite tiring, as we are looking for everything that can go wrong and then deciding how to counteract those identified risks. It feels very negative, and I wanted to figure out how to make it a more positive experience.

After approximately six hours of identifying risks and countermeasures to those risks, I came up with a positive approach. We would review the FMEA at every monthly team meeting and see if we had reduced any of the identified risks. We would celebrate every risk we reduced. The more risks we reduced, the more we would celebrate.

As the project progressed, we saw reduction in many of the risks and improvement in overall changeover times. Team members started competing to see how many risks they could reduce and how much they could impact the RPN.

After nine months, we had minimized the top risks in the project and the overall changeover time had comfortably settled below 15 minutes. Keeping critical risks in front of everyone focused us on the most important work to achieve our objectives. Years later, changeover time was at 11 minutes on average and stayed there.