Making enterprise asset performance management work for your plant

Combine known rules with machine learning for better maintenance decision-making.

By Kim Custeau, Schneider Electric

New Call-to-action

It’s no secret that organizations today are focused on driving efficiencies throughout every area, and are at the point where their focus is on maintenance efficiency and effectiveness. As a consequence, in order to reduce unplanned downtime and maximize labor efficiency, the new generation of plant personnel are empowered to act before equipment failure occurs. From an asset management perspective, organizations are leveraging industrial data and analytics while integrating all the various elements of a maintenance system. Keeping plant equipment running is critical, but how do plants accomplish that when capital and operational expenditures are extremely limited?

The Enterprise Asset Performance Management (EAPM) approach is about empowering the enterprise, transitioning from a corporate view of assets (i.e., managing the lifecycle through improved maintenance visibility and standardized practices) to a holistic, integrated, and operations-centric view. These solutions enable customers to exceed safety, reliability, and performance goals through data collection and analysis coupled with actions and optimization for proactive maintenance execution.

This vision involves being able to build connections from the sensor or smart device assets all the way up to the ERP systems, making valuable information from the plant floor more accessible and delivering context to plant teams on the device of their choice in order to learn and make better decisions over time. The idea is to have a broad portfolio for users to collect information on assets, analyze it, determine the next course of action, and then use that action to further refine the next set of actions.

In other words, this is a continuous improvement program, bolstered where appropriate by automated workflows and predictive modeling, one that starts with an understanding of the Maintenance Maturity Pyramid. The pyramid represents a different kind of approach to take for the different kinds of assets that exist, one that follows the integrated EAPM model.

At the bottom of the pyramid, plant personnel often say, "we have to move away from reactive maintenance modes." A more accurate statement is, "for this specific set of assets, reactive is the right way to perform maintenance, because failures do not otherwise affect my process, and letting these assets run to fail costs less than the effort to replace them ahead of schedule."

The next level is where preventive maintenance (PM) programs come into play, and workflow automation begins. On the majority of your assets you can set up PMs, you can identify a time frame, and then use the OEM manual to set up and schedule periodic work. You're starting to create automated ways to remind you that for these pieces of equipment, you have this specific regular maintenance action to take.

Moving up the pyramid, condition-based maintenance (CBM) represents the initial stage of a more proactive maintenance approach. The primary benefit of condition-based approaches is that they can automate the maintenance process through the monitoring of user-defined rules that initiate necessary maintenance activities. Condition-based approaches and technologies are commonly employed to monitor industrial asset performance when the asset condition is known and definable using rule-based or algorithmic logic.

In essence, instead of executing a PM in reaction to elapsed time,  plant personnel can create a maintenance strategy that is based on one or more conditions, and can have each of those conditions (or combinations of conditions) trigger events. This does not require any user intervention – the system automatically creates a work order that is then assigned to a team member to execute.

To take it to the next level of the pyramid, organizations need to tie in predictive maintenance (PdM) approaches to model the performance of critical assets. The combination of known rules (CBM) plus advanced pattern recognition and machine learning (PdM) results in a robust industrial asset analytics platform where you can look at the model and identify either a performance issue or an impending failure days, weeks or months before traditional practices.

With predictive maintenance, personnel know and understand the actual and expected performance for an asset’s current operational state. Access to contextual data then enables you to go back and look at previous maintenance and production data for that asset, combine that information, and then make an informed choice of what action needs to be taken.

Laying the foundation to achieve these kinds of process automation benefits involves getting a baseline solution in place to manage assets: both a work management process to handle routine maintenance (including some PM) and an asset structure that makes sense. At that point teams need to start identifying critical assets, and understand exactly how they are operating; part of this effort includes identifying and monitoring the conditions in which these assets are operating.

The result is maximum economic return for all assets, either through early warning notification of equipment issues, through making the workforce more efficient, or through improved access to relevant and contextual information that can be served up on a mobile device. The benefits are real, with the value rooted in system connectivity which enables continuous improvement and better choices. It's simply a matter of asking, what will an hour of downtime cost me, and what if I can prevent that?

Show Comments
Hide Comments

Join the discussion

We welcome your thoughtful comments.
All comments will display your user name.

Want to participate in the discussion?

Register for free

Log in for complete access.

Comments

No one has commented on this page yet.

RSS feed for comments on this page | RSS feed for all comments