Why the Metric of Man-Weeks Is Insufficient for Managing Maintenance Backlogs

Why the Metric of Man-Weeks Is Insufficient for Managing Maintenance Backlogs

With rising cost pressures and restrictions on increasing labor, many organizations find themselves facing a substantial maintenance backlog. Ensuring that you understand your maintenance backlog is vital to the proper operation of your site – and the most-used metric for managing it is the number of man-weeks the backlog represents.

What’s the problem with a large maintenance backlog?

The obvious issue is the uncertainty that important repairs will be completed in a timely manner. It’s not uncommon for repairs to be prioritized based on how individuals advocate for them instead of the repair’s true importance. A consequence of this behavior is an organization being surprised by an item that has become a higher priority over time and subsequently having to deal with the resulting operational challenges.

What is a backlog?

The term backlog has several meanings, and each type needs to be handled differently. The figure below shows the various backlogs in the maintenance work process.

Types of Maintenance Backlogs

Figure 1: Types of Maintenance Backlogs

When people refer to a backlog, they are normally talking about the Execution Backlog. This consists of work that has been planned and is available to be placed on the maintenance schedule once materials or services are ready. As hours have been allocated to the activities, you can measure the Execution Backlog in terms of man-weeks. This metric is calculated by adding together the sum of the work within the Execution Backlog and dividing by the available workforce hours. A classic rule of thumb for the Execution Corrective Backlog is to have 6-8 man-weeks of work.

The lower boundary of the rule of thumb is to allow enough planned work to be available to maximize the opportunities for optimizing maintenance scheduling and execution. The upper boundary is to give the organization confidence that important work requests will be completed in a timely manner.

It is also important to understand whether the Execution Corrective Backlog is growing or shrinking. This provides you full visibility into whether your maintenance labor force is staffed correctly. As the number of man-weeks grows, expansion of the labor force may be warranted on a temporary or permanent basis. As the number of man-weeks drops, consideration should be given to contracting the labor force.  Ideally this is accomplished via contracting or expanding the contractor workforce instead of company workforce.

This rule of thumb is often misused and applied to the Unplanned Backlog as well. Since the Unplanned Backlog hasn’t yet had hours put to the activities, assumptions must be made to translate the count into man-weeks. Artificially translating the count of unplanned work orders to man-weeks can be misleading and doesn’t address the major concern within the Unplanned Backlog, which is ensuring that delayed repairs don’t have unanticipated consequences at inopportune times.

If the rule of thumb for man-weeks isn’t appropriate for the Unplanned Backlog, what can you use instead?

To address the concern of delayed repairs within the Unplanned Backlog, a different measure is needed.  The best measurement is total cumulative risk of the Unplanned Backlog, which sums each work request from a risk standpoint and gives a total cumulative risk for the Unplanned Backlog. The cumulative risk can then be tracked to determine if the risk level is increasing or decreasing over time. Higher-risk work requests can be identified and monitored for the length of time in the Unplanned Backlog, and decisions can be made to expedite these work requests to Planned status as needed. This methodology is typically extended to include the entirety of the Approved Backlog, consisting of the Unplanned Backlog, the Planner Backlog, and the Execution Backlog.

The methodology is similar to using risk to prioritize maintenance work orders in the Execution Backlog, where risk is defined as a combination of consequence and probability. However, there are some key differences when measuring the total risk in the Approved Backlog compared to prioritizing maintenance work orders. When prioritizing maintenance work, various scenarios are considered and the one with the highest risk is used to determine priority and timing of the repair. When evaluating risk in the Approved Backlog, various scenarios are considered and the Total Risk is calculated.

Another key difference is in the determination of probability. When calculating the probability of a consequence in the prioritization of a maintenance work order, the timing of the consequence is considered (e.g., already occurring, likely within two weeks, likely within three months, etc.). When calculating the probability of a consequence in the Unplanned Backlog, the likelihood is determined by the nature of the scenario is (e.g., has occurred at the site, has occurred in the company, has occurred in the industry, etc.).

In addition to measuring Total Risk in the Unplanned Backlog, visibility into how long high-risk items remain in the backlog is necessary to understand potential issues and the likelihood of having to deal with their consequences.

What should the complete backlog measurement process look like?

The various maintenance backlogs are measured and managed in different ways:

  • The Unapproved Backlog is measured to determine how long work requests take to be reviewed and either approved or rejected.
  • The Unplanned Backlog is measured for total risk and how long items remain in the Approved Backlog to understand how likely it is that an unexpected consequence occurs.
  • The Technical/Engineering Backlog is measured to understand how efficient technical support is for maintenance work requests.
  • The Planner Backlog is measured to help understand how efficient the planning process is.
  • The Execution Backlog is measured to enable efficiency in scheduling and execution as well as gain insights on how the labor workforce should be managed.

The overall view of metrics used to manage backlogs is shown below.

Metrics for Managing Maintenance Backlogs

Figure 2: Metrics for Managing Maintenance Backlogs

Becht personnel, most with over 30 years of experience, have long-term maintenance and reliability careers within owner organizations and, as a result, approach maintenance demand with an owners’ perspective. If you need help getting started on managing your backlog, Becht can help.

Like what you just read? Join our email list for more expert insights and industry updates.

|

About The Author

Contact:
Charles Maier has a broad background focused on plant operations, maintenance, reliability and cost containment. His expertise is analyzing situations, identifying opportunities and implementing strategic solutions that deliver bottom line results. He has spent the last 20 years working for a major oil company with previous experience in the pulp and paper industry as well as serving as a nuclear officer in the United States Navy. He is a key member of the Becht team delivering site improvement programs focused on maintenance and reliability. While working for a major oil company, Charles led multiple site business improvement efforts in downstream and upstream sites. Key sustainable results include a reduction of contractor workforce by 10%, a reduction of pitstop durations by 30% and a reduction of total maintenance cost by 20%. He also implemented a site reliability program which captured over a $1 Billion dollars through the establishment of a reliability department and associated processes including RCA, RCM and bad actor programs. Charles was the lead author for the maintenance work process of a major oil company and has helped develop company-wide tools and programs to deliver improved maintenance and reliability performance. He has also facilitated multiple company technical networks in the area of maintenance. Mr. Maier received his Bachelor of Science in Marine Engineering and Naval Architecture from Webb Institute and his Masters of Science in Engineering Management from Southern Methodist University.

Authors Recent Posts

Why the Metric of Man-Weeks Is Insufficient for Managing Maintenance Backlogs
Let Becht Turn Your Problem
Into Peace of Mind