I had lunch with a group of Reliability and Maintenance Managers recently, and the conversation at the table turned to AI. One said that it would not be too long until someone develops an AI-powered, fully integrated Maintenance Work Management platform, or "Asset Management AI," making life a lot easier.
He went on to describe how the "Asset Management AI" would use real-time data analytics to dynamically optimize planning, scheduling, and execution of maintenance tasks. It would not only predict equipment failures but also autonomously generate optimized maintenance schedules, prioritize tasks based on their operational impact, and recommend resource allocation in real-time. Furthermore, he described how the AI would continuously learn from equipment performance, historical maintenance data, and operational conditions to adjust schedules and priorities in real time.
"The problem is," the Reliability Manager confessed, "the AI would have to be synced up to our CMMS and get its data from it, and so it would be doomed from the start." The Reliability Engineer then motions to everyone at the table and says, "We're supposed to be doing all that stuff now, so we're equally doomed." They agree and laugh, commiserating about the way their CMMS had been set up.
In the future, a lot will need to be done to enable the success of an AI system coupled with an organization's CMMS data. What about now, and enabling the humans who are doing all these great AI functions themselves? Aren't we sometimes hampering our people by turning a blind eye to CMMS shortcomings?
We often see many ways in which a CMMS needs to be reconfigured, as well as Work Management Standards and Processes that are lacking. Typical anomalies are:
- The need to validate and organize the asset registry into functional systems and asset types, as well as to close holes in data attributes and tagging.
- Well-designed hierarchy is often a problem (equipment items with no parent, items in the incorrect position, etc.).
- Need for risk management and criticality analysis.
- Failure Code Chaos and a Lack of Failure Reporting and Corrective Action Process.
- Poor utilization of the CMMS Job Plan Library and a lack of detailed Job Plans for repetitive corrective maintenance, rebuilds, and refurbishments.
- Poorly designed Work Order Priority systems and Work Order Types.
- Workflows and Standards: Clear workflows and process standards for work request initiation, approval, scheduling, execution, and closure. These are important now as well as in the future to guide an "Asset Management AI's" automation and optimization logic.
Obviously, proper CMMS configuration is vital for an Asset Management AI to function, but it is also vital to us now to make the right choices and course corrections in our maintenance strategies.
A well-defined asset hierarchy is critical to the plant's ability to drill down into costs and identify where an improvement effort should be focused. A well-configured CMMS enables us to focus on which equipment has the most failures, which ones incur the most costs, and which ones generate the most corrective work orders, but only if we are collecting and harvesting the right data. It allows us to focus on what matters most.
It also allows reliability staff to identify common issues across specific equipment types and classes, enabling what may be an improvement targeted for a specific area to be spread out across the site.
To maximize the value of your staff and set them up for success, it's worthwhile to conduct a CMMS check-up, a gap analysis against ISO 14224, and other relevant assessments to close the gaps that matter most and work towards removing all the hurdles they currently face.
A properly configured CMMS is crucial for Reliability Engineers and reliability personnel, as it provides a structured, data-driven foundation for making informed decisions about maintenance strategies, reducing failures, extending Mean Time Between Failures (MTBF), and enhancing overall reliability and asset management.
Benefits for Decision-Making and Reliability:
Informed Maintenance Strategies: A properly configured CMMS provides high-quality, organized data that enables Reliability Engineers to develop proactive maintenance plans (e.g., preventive, predictive, or condition-based maintenance) tailored to specific assets and their failure modes.
Failure Reduction: By analyzing failure data and trends, engineers can identify recurring issues, implement corrective actions, and reduce unplanned downtime.
Extended MTBF: Accurate data on asset performance and maintenance history helps optimize maintenance schedules and interventions, extending the time between failures (MTBF).
Improved Reliability: By leveraging insights from criticality analysis, failure modes, and historical data, engineers can prioritize high-impact improvements, such as equipment upgrades or process modifications, to enhance overall system reliability.
Enhanced Asset Management: CMMS enables lifecycle tracking, cost analysis, and performance monitoring, supporting strategic decisions about asset replacements, refurbishments, or investments.
In summary, a well-configured CMMS ensures that Reliability Engineers have access to accurate, consistent, and actionable data. This empowers them to make evidence-based decisions, optimize maintenance strategies, reduce failures, extend MTBF, and improve asset reliability and management, ultimately driving operational efficiency and cost savings.
We, like the Asset Management AI of the future, can only be as good as our data allows us to be.
