In defining our future first we must understand our past. The late management guru Peter Drucker, when asked how he made such accurate predictions, said, “I don’t forecast; I look out the window and identify what’s visible but not yet seen.”
As reliability and maintenance professionals, to prepare for the future and to implement a new strategy, we must understand how our needs will change. If we are going to implement a new strategy or improve an existing one, we must first understand what will change for all involved stakeholders.
The reliability group at my facility recently performed an extensive review of our processes to determine areas of possible improvement and learned that, from operations to maintenance, process improvement has to be a total team effort. Here’s how we identified areas of improvement and how we deployed solutions that moved us from a reactive mindset to one that was proactive among all teams.
Identifying the issues
We began the project by determining the overall mean time between failure (MTBF) at the plant. Using plant failure data and completing a Crow/AMSAA Reliability Growth model, we were able to determine that MTBF was averaging approximately 14 days. Few on the team thought this data was surprising or unnerving: They were so numb to all the failures that they hadn’t realized the circle of despair that had become the new norm. However, the reality of the MTBF data also presented us with a challenge to make a difference in how the plant was operated and maintained.
After admitting we had a problem, the next step was to determine how bad the problem really was. With the work of a few great individuals we pulled the failure data for the past three years and charted the dates on a simple wall calendar purchased from the local office products store (Figure 1). We color-coded the corresponding events and dates for each year and then posted the process flow diagrams (PFDs) of the plant on the wall, adding a colored dot to where each failure had occurred (Figure 2). The reason for this was to identify where the actual failures took place and the corresponding work orders that were written to make the repairs.
Figure 1. Charting equipment breakdowns through time can help make failure patterns more visible.
Figure 2. Using process flow diagrams can help identify the spatial relationships between failure clusters.
Upon review of the data, we uncovered a surprise: The locations where failures had taken place did not align with the locations where the majority of plant personnel felt our biggest problem areas existed.
We also discovered that just focusing on maintenance and reliability improvements would not get the job done. We needed to focus our attention also on the operations team. Your operators are your first line of defense when it comes to understanding the health of your assets, and the daily walk-around inspections they do can be a blessing or a curse.
After walking down the inspection rounds with the operators, it became very clear that we had become numb to the noise of what our equipment was telling us. What I found were some basic items that in the grand scheme of things made a big impact on how we operate: broken gauges, busted conduit with exposed wiring, constant level oilers with oxidized oil, steam piping with missing insulation – and that was just on the first walkdown (Figure 3).