Using Rolling Averages for Results Monitoring

Share: FacebooktwitterredditlinkedinmailFacebooktwitterredditlinkedinmail

Production downtime is undesirable for any plant. Every hour the production is off line, output decreases. At a paper bag manufacturing plant, a key performance indicator the plant manager looks at is the plant uptime. The result is tracked weekly.

When the result is presented on a dashboard, a rolling average is used. This is how they compute the rolling average. They take the uptime for the current week and average the number with the uptimes for the previous two weeks. They call it the rolling average uptime. The rationale for looking at the rolling average is that scheduled maintenance is known and the unanticipated downtime should be infrequent. Hence, the rolling average would present a better overall production picture.

Averages are usually useful for estimation purposes. For example, estimating the production capacity for the month. It is important to note that averages smooth out variations and make it difficult to detect change. In this case, it is more informative to look at the actual uptime because the rolling averages might present a business-as-usual picture when there might be underlying changes that should be noted.

The purpose of results monitoring is to identify anomalies early so that you can do proper diagnosis and take proactive action. Smoothing out the actual results by averaging them camouflages the real situation.

Leave a Reply

Your email address will not be published. Required fields are marked *

To maximize business results, call Connie at 604-790-1220 or email us today!