5 Traps to Avoid When Selecting Asset Performance Management Software
The market is saturated with software that promises to improve asset performance and reduce or eliminate downtime. Whether it brands itself as asset performance management, predictive maintenance, reliability centered maintenance or condition-based maintenance software, all seem indistinguishable from each other.
But they are not. Behind the claims are deal-breaking differences that separate true operational intelligence (APM) solutions from those that come short of expectations.
The key is knowing when claims are misleading or downright untrue. Here are five “red flag” statements you may have heard.
The Full Value of APM Software Is in Alerting You To Potential Asset Failures – FALSE While alerting you to risks is certainly part of what APM software should do, it is just that: one part. True APM yields value on so many more levels. APM software should also facilitate the downstream diagnosis of root causes and prioritize issues according to how they impact your financial, safety and environmental goals. Finally, to ensure speedy and efficient issue resolution, the software should support corrective actions and provide a collaborative platform where operations and asset managers work together.
You Have To Start Building Math Models From Scratch To Detect Real-time Asset Anomalies – FALSE Critical to the timeliness and successful APM deployment is an automated, no-code implementation environment. A successful implementation builds custom models starting from model templates. Automation of data cleaning, feature selection, and model deployment supplement the templates to ensure high-quality custom models. For those who want to dive deeper, your APM solution should have no-code tools to simplify further custom-model building. This is typically less than 20% of any deployment, after all you are buying an APM software not consulting and software development.
It’s Hard to Prioritize Issues Based on Financial Impact – FALSE
Your operations exist to achieve strategic business goals, including the return on investments. Therefore, true APM software must inherently quantify the risk of different options against one another and recommend actions based on their economic, environmental, or health and safety impact. Prioritization based on business impact should be an automated part of the overall workflow and process following an alert. The only way to make the right decisions for asset maintenance is to quantify how these alerts will directly affect your operations and business goals.
You Will Need a Team of Data Scientists To Implement Your APM Solution – FALSE The math required to accurately detect, diagnose, and resolve asset issues in a complex environment is extraordinarily challenging. However, a mature APM software will have libraries of proven calcs and models available to get you up and running quickly. So, don’t squander the time and talents of your data scientists by reinventing the wheel. Look for APM software that can automate the deployment of your detection environment, freeing up data scientists to work on unique problems.
It Will Take Several Months, or Even Years, To Create and Implement a Solution for Your Complex Operations – FALSE
1. Your APM solution connects remotely to your asset and process data sources.
2. Your data is used to build out an asset hierarchy and data structure.
3. Performance calculations supplement your data with calculated performance metrics. 4. ML models are created easily, and a predicted value is established as a baseline for operations.
That’s it – your APM solution is live!
What you need from an APM solution is very straightforward. Your solution should be able to easily connect to existing data sources, detect both emerging performance and reliability issues, diagnose potential root causes, and provide a collaborative environment to resolve issues to improve performance and stop downtime from draining your bottom line.
Find out how we offer all this and more to our customers at booth F6 at Asset Performance 2023 and www.prometheusgroup.com.