One metric to rule them all !!

I get asked pretty regularly about metrics. Being the author of a book on the subject it doesn't surprise me...but the most common question is generally something like...

"Which metric do you recommend to give an overview of reliability improvement?"

This is almost always followed by the questioners own preference for either Availability, or OEE, MTBF or something similar.

The only answer that I can give, from my own experience, is that no such thing exists.



Where does reliability have an impact?

Before we think about what to measure, we always need to think about what we want to achieve. All metrics and measures here are at the KPI level, meaning that they are the most important indicators of an area of performance.

There are, of course, thousands of metrics. And there is just no way I am going to try to list them all here.

Reliability improvement will impact on four areas only, and if you want to challenge that then feel free to - I have been trying to challenge it for many years myself.

(Before you try... reputation would be a product of one of the other four, so it is already there)

Increasing Revenue

Often the number 1 goal of any client I have ever had. However, it does depend a lot on the business cycle, demand levels and the price they are selling at. (profits)

Metrics

• Maint. Cost / Produced Unit (Ton, Liter, Barrel etc) (More production = less cost per unit. Don't be fooled into thinking this is a cunning way to cut costs only. If that is how it is being used it will fail as a measure)
• Availability (More uptime = more opportunity to produce. Therefore more revenue potential)
• MTBF (Because you can have great availability and lousy MTBF. The two are not joined at the hip and it is best to track both of them

Insert OEE as you see fit but personally I prefer to run OEE as a separate improvement initiative altogether. Normally coupled with defect elimination and / or RCM.

I really like OEE but it needs to be very clear that it isn't actually Overall Equipment Effectiveness, but primary function effectiveness. Sounds semantic but it is very relevant.

Decreasing Direct Costs

The second most asked for impact of reliability. Again it depends very much on the present economic climate, but this one is normally a good starting point.

Metrics

• Maint. Cost / Produced Unit (Again. This time for the cost side of the equation, not just the production side of it.)
• Maintenance Productivity Factor (One of the metrics I was involved in creating)
• Cost comparisons (Actual versus budget etcetera)

Knowledge Increase

This used to be pretty hard, until I actually worked out what Knowledge increase meant for companies. It used to be filled with garbage like number of technicians trained in (say) RCA.

But that was the old way of seeing knowledge. When we used to think of it as what employees held in their minds only. Today we realize that corporate knowledge is a collection of data + information.

For many reasons that we wont go into here, the goal is to move away from information (the stuff in experienced peoples heads) and towards data. So good metrics here really revolve around the Asset Data Scorecard concepts.

Risk Reduction

Another one that used to be harder to measure. Again until it all clicked during the mid 1990's.  Assuming, and it is sometimes a big assumption, that an asset or plant is already being managed with the minimum safe level of maintenance (via RCM or whatever) then a practical measure here is Schedule Compliance. (Particularly of safety related tasks like vessel inspections, function tests and so on)

Another great risk forecasting tool includes charts of priority versus age of corrective work orders, and reports from online and manual condition monitoring data.

In summary...

Not surprisingly there is no silver bullet, and no metric to rule them all. Instead what we find are a range of metrics depending on what you are looking for and what you are trying to achieve.

There is also a different slant here on what a metric actually is. We used to think that it was a number that represented a trend, but it could just as easily be the age vs priority graph of corrective work orders out of the backlog. 

Good luck...