Managing maintenance is no different than any other business. You must make good decisions that add value. This means you need input and lots of it. Making decisions based on gut feelings just doesn't cut it these days. Once repair and maintenance (R&M) dollars have been allocated, your organization will expect value to be delivered in the form of reliable assets. Key performance indicators (KPIs) can provide the input you need to help meet this lofty objective.

Ground Rules

KPIs can be very specific to an individual's activity or broad enough to measure an entire plant's performance. They can be leading or lagging, simple counts or complex calculations. However, they should follow a few simple ground rules to ensure quality input.

  • Metrics should be easily connected to the business value that strong performance will deliver.
  • Metrics should be aligned with the current state of the organization. The percentage of proactive work will be very misleading if you are only capturing 50 percent or your work on work orders.
  • Metrics should be automated. Manual counts are riddled with errors and often get neglected.
  • Metrics should follow a process or practice. Activities should not be added to a process solely to make the KPI measurable.
  • Metrics should have reasonable, defined and understood targets. Be careful of the law of diminishing returns.
  • Metrics should be monitored, and responsible parties held accountable for their results. Building them into performance objectives and incentives has proven very successful.

Alignment to Business Process

A well-defined maintenance business process must be executed in order to deliver the desired results. In other words, the engine needs to be running on all cylinders to provide the necessary horsepower to win the race. The figure below illustrates ArcelorMittal's maintenance business process. Each component contains specific activities and practices that are essential to delivering business value to stakeholders. Each component's activities are monitored with a set of standardized KPIs to provide managers with the input needed to make informed decisions and enact corrections and improvements as appropriate.


The plan component of the business process sets the stage for all other activities. It involves understanding the business objectives, identifying physical assets that support those objectives, analyzing the consequences and risks those assets represent, and establishing performance requirements needed to reach the business requirements. While these activities primarily lead to the development of targets for metrics used to monitor downstream activities, there are some examples of KPIs used to monitor this component, such as the percentage of assets with asset prioritization analysis completed and the maintenance budget as a percentage of the replacement asset value.


The improve component is all about work identification. This involves the development of a technically valid maintenance program for the assets identified in the plan phase. It is accomplished by using methodologies such as reliability-centered maintenance, maintenance task analysis and root cause analysis to identify the tasks that will most effectively mitigate the consequences of failure. Some examples of KPIs used to monitor progress include the number of failure modes identified, the percentage of maintenance action plans implemented, and the number of condition inspections implemented.


The control component is about effectively executing the maintenance program developed and implemented in the improve component. Maintenance tasks are planned, scheduled and executed, and follow-up activity is performed to improve maintenance plans. This is called the "sustained maintenance loop." Metrics monitor many aspects of this phase to understand the quality of maintenance planning, the effectiveness of scheduling and the efficiency of execution. Some examples include the percentage of backlog ready to work, planned vs. actual man-hours, the percentage of scheduled outage work orders completed, and the percentage of preventive maintenance compliance.


The assess component is where performance is compared to targets to understand where further improvement is needed. Lagging KPIs are primarily reviewed in this phase. These KPIs vary based on the type of unit. For example, mean time between failures is valuable for continuous processes where each line stop results in significant yield loss. Batch processes tend to look more at delay rates or production metrics. Some examples include working ratio, heats per day, budget vs. actual repair and maintenance spending, and prime yield.

Calculation and Reporting

KPIs can be automatically calculated and made available in real time on dashboards within reliability software. The input to the calculations come from the reliability system, the computerized maintenance management system (CMMS), operating data and financial reporting.

A standardized list of KPIs should be deployed at the corporate level, plant level and the individual business level within each plant. They can then be accessed for real-time data at any time.

Dashboards can be created to meet the needs of specific roles in the organization such as plant managers, maintenance managers or planners. KPI data can also be compiled in monthly reports for review meetings and distribution to upper management.

These key performance indicators can provide valuable input for making informed decisions. Good decisions will ensure business value that moves the company toward meeting its objectives. Just be sure your KPIs are defined, understood and communicated across your organization.