Recently an online discussion and commentary motivated me to discuss some issues on reliability and the use, misuse and misunderstanding of words to describe what goes on in maintenance and reliability departments. In industry we talk of reliability, uptime, minimizing downtime and utilization as the main metrics of asset performance. But of these, what is important? 

Reliability is used very loosely and without proper context. A key leader in the industry discussing improving reliability said, “The language is not correct. Many don’t care, because very few organizations practice real reliability (they just renamed it maintenance) and even fewer practice asset management.”

Several definitions of reliability occur: 

1. The Oxford English Dictionaries:
   The quality of being trustworthy or of performing consistently well.
   The degree to which the result of a measurement, calculation, or specification can be depended on to be accurate.
2. Merriam Webster:
   The quality or state of being reliable
   The extent to which an experiment, test, or measuring procedure yields the same results on repeated trials
3. Dictionary.com:
   the ability to be relied on or depended on, as for accuracy, honesty, or achievement.

I am not convinced that any of those definition help sufficiently. Could such differing definitions be why don’t we collectively understand reliability? Do words make a difference?

Let’s explore the nuances of meaning, starting with reliability. A breaker that trips once a week and takes two minutes to reset is not reliable. A compressor that goes down once a year and takes half the plant down is much more reliable but now half the plant is down for weeks. Which is more important and is reliability the main metric? 
Some have pointed out that a risk or criticality review will assist in reliability management. True -- but that just points out that reliability alone is an incomplete metric; it needs more. 

How must we frame reliability? Clearly, the word reliability alone is insufficient and some measure of availability must be part of the assessment to describe what we need in manufacturing. Measuring availability proposes to give us what we need. One website defines equipment availability or asset availability as a metric that measures the probability that a system is not failed or undergoing a repair action when it needs to be used. It is defined in an equation as:

Availability = Uptime ÷ (Uptime + Downtime)

Sometimes this is called utilization. Generally, this is a lot closer to what we are really trying to do, but still falls short because it implies the equipment is required at all times. Availability as a standalone metric has its critics, since a piece of equipment can be available but not reliable. Additionally, as an example, especially in some batch or discrete manufacturing situations, the equipment may not be needed all the time, therefore servicing may occur without directly affecting required uptime. 

Many years ago, I was consigned to develop a dashboard with the key metrics for managing activities of the maintenance department. At the time, we posted no explicit metrics for reliability or “raw” availability; instead we preferred “availability to plan” as the appropriate measurement. It is my opinion that availability to plan is the most useful and correct metric to measure asset performance. My equation is:

Availability to Plan = Required Uptime ÷ (Required Uptime + Downtime)

Availability to plan implies far more than availability. If I plan to use a piece of equipment for 12 months, then that availability to plan means it must be available the for the whole duration of the plan – which incorporates reliability. Additionally, now we can define how equipment meets its required duties, we need to consider proper definition of utilization or overall equipment effectiveness (OEE). Those metrics must incorporate an assessment of the planning function, since poor planning may not change availability to plan but it can reduce OEE. The word puzzle deepens. 

Want to discuss how your organization defines and measures reliability? Stop by the AspenTech booth at IMC this week.