• Written by Muhammed Abdulla NC, CEO

A maintenance manager stares at a dashboard that nobody opens, filled with work orders and asset history, wondering if AI can reduce reactive work crowding out planned work. The answer depends on what is in the data. For AI to make a difference, the data must include a complete record of work orders, including both planned and reactive work, as well as a detailed asset history. This information is crucial because it allows AI to identify patterns and trends in the data, enabling it to provide insights that can inform maintenance decisions.

What has to be in the data for this to be possible

The data must be comprehensive and up-to-date, covering all aspects of maintenance work, including routine inspections, repairs, and replacements. A shift supervisor checks the maintenance log to ensure that all work orders are properly documented and recorded. This includes details such as the type of work performed, the assets involved, and the resources used. Without this information, AI cannot provide a accurate picture of the maintenance workload and identify areas where reactive work is crowding out planned work.

A good example of this is when a category manager compares two suppliers, one of which provides detailed records of maintenance work, while the other does not. The category manager can use this information to make informed decisions about which supplier to use, based on their ability to provide reliable and efficient maintenance services. In this case, the data is used to support a business decision, rather than simply being collected for its own sake.

The mechanism, stated plainly

The mechanism by which AI reduces reactive work crowding out planned work is straightforward. AI analyzes the data to identify patterns and trends, and then provides insights and recommendations to maintenance managers. For instance, AI may identify that a particular asset is prone to breakdowns, and recommend that a planned maintenance schedule be put in place to prevent these breakdowns. This allows maintenance managers to prioritize planned work and reduce the likelihood of reactive work crowding out planned work.

Most people get wrong the idea that AI can simply be applied to any dataset and magically produce results. This is wrong, because AI requires high-quality data to produce meaningful insights. Without proper data, AI is simply a tool that cannot provide useful recommendations.

What it will not fix

AI will not fix underlying issues with maintenance processes or asset reliability. If the maintenance process is flawed, or if assets are not properly designed or maintained, AI will not be able to overcome these fundamental problems. For example, if a particular asset is prone to breakdowns due to a design flaw, AI may be able to identify the problem, but it will not be able to fix the underlying design issue.

In some cases, AI may even mask underlying problems, by providing temporary solutions that do not address the root cause of the issue. This is why it is essential to have a deep understanding of the maintenance processes and asset reliability, and to use AI as a tool to support, rather than replace, human judgment.

How you would know it worked: planned versus reactive ratio

To determine whether AI has been successful in reducing reactive work crowding out planned work, maintenance managers can track the planned versus reactive ratio. This ratio compares the amount of planned work to the amount of reactive work, and provides a clear indication of whether the maintenance workload is becoming more proactive or reactive. By tracking this ratio over time, maintenance managers can see whether AI is having a positive impact on the maintenance workload.

For more information on how to implement AI in a production environment, please visit our production page. If you have any questions or would like to discuss how AI can support your maintenance operations, please contact us.

What this does not do, or when it is the wrong choice

This approach does not replace the need for human judgment and expertise in maintenance decision-making. AI is a tool that can provide insights and recommendations, but it is not a replacement for experienced maintenance managers who understand the complexities of the maintenance workload. In some cases, AI may even be the wrong choice, such as when the maintenance workload is highly variable or unpredictable, or when the assets involved are highly complex or critical.

In these cases, human judgment and expertise are essential, and AI may not be able to provide the necessary insights or recommendations to support maintenance decision-making. By understanding the limitations of AI, maintenance managers can use it effectively to support their operations, and avoid relying too heavily on a tool that is not suitable for every situation.