Maximo

Blog: How Generative AI is Transforming Enterprise Asset Management  (Author: Kiran Bhogle)

Disclaimer: This post is based on general industry trends and a hypothetical scenario. The companies, individuals, and events are purely fictional, and any resemblance to real-life entities is purely coincidental. The views expressed here are my own and not necessarily those of my current employer.

Understanding the Challenges

I recently visited a mid-sized manufacturing company that specializes in producing high-performance Industrial components for industrial applications. During the visit, I had the opportunity to meet with Ashish, the company's Head of Operations, who shared the ongoing asset management challenges they’ve been facing.

As we sat down to chat, Ashish shared the ongoing challenges that had been hindering the company’s asset management strategy:

• Unpredictable Downtime: Their Conveyor motors frequently break down unexpectedly, which is leading to unplanned downtime and delays in production.
• High Maintenance Costs: Routine maintenance is often reactive rather than preventive, resulting in high costs for repairs and parts replacements.
• Overwhelming Data: Ashish mentioned that while they collected large amounts of sensor data from assets, it is becoming overwhelming, and the team lacked the tools to derive actionable insights from it.
• Fragmented Asset Life Cycle: The management of the assets’ lifecycle, from procurement through maintenance to decommissioning, is disjointed.

Curiosity: Why Gen AI?

During the conversion Ashish brought up the topic of Generative AI, he said “I’ve been hearing so much about Gen AI recently”. He had met multiple vendors, and each one had given him high-level pitch about Generative AI, but they couldn’t go beyond basic understanding and most of the use cases are siloed and lacking unified vision.   

Ashish was eager to understand how Generative AI could help him address these challenges across the full asset lifecycle and whether it could be a worthwhile investment for his team. He wanted a deeper understanding of how Gen AI could help his team improve their asset management processes.

I started by explaining what Generative AI is: an advanced type of AI that learns from vast amounts of data to generate predictions, insights, and even automate solutions. Unlike traditional machine learning, Gen AI can analyze and synthesize data from various sources, predict potential failures, and recommend proactive actions.

As he was looking from asset lifecycle point of view, we decided to go back to drawing board and understand what the challenges are anticipated or faced throughout the asset lifecycle and how those can be addressed

·       Acquisition and Planning

How do I select the right model of asset in this case for Motor as an asset?

What would be an assets total lifecycle cost?

·       Installation and Deployment (Commissioning)

How do I ensure the installation of asset is perfect and align to Manufacturer guidelines?

How to get optimal operations parameters and settings?

·       Operation and Monitoring

How to avoid unpredictable downtime?

How do I get Data-Driven Insights and Reporting for an asset

How to I detect an asset anomaly in advance?

·       Maintenance and Repair

How to reduce repair cost and unplanned downtime

Is there way to predict when asset needs maintenance or going to fail and schedule maintenance

·       End-of-Life and Decommissioning

Can we analyze asset usage, data generated and recommend when to replace the asset?

What would be recommendations for replacement assets?

After listing down key challenges in Asset Lifecyle journey, we continued our discussion to see how Gen AI can help to address these challenges. I said let’s first understand what’s Gen AI is and what are its key characteristics for Gen AI enabled solution approach

According to Gartner “Generative AI refers to AI techniques that learn a representation of artifacts from data, and use it to generate brand-new, unique artifacts that resemble but don’t repeat the original data. These artifacts can serve benign or nefarious purposes.”

In the context of Asset Management, Gen AI Solution involves deploying of advanced artificial intelligence technologies that leverage large amounts of data to optimize the management of assets throughout their lifecycle. This enables the use of machine learning algorithms, predictive analytics, and data driven decision making to enhance the efficiency and reliability of asset management practices.

The Role of Gen AI in Addressing These Challenges

Now let’s understand what the key characteristics of Gen AI are will be of most useful for Asset Management Scenario

·       Through out the Asset Life Cycle lot of data gets generated or maintained such as maintenance logs, sensor data, technical and operational guides etc., Gen AI will help to understand context of data to get more insights e.g. Gen AI will help to understand Maintenance logs to get better visibility to recurring issues OR Generate SOPs & Best practice from Manuals and documentation.

·       Gen AI can identify patterns and potential failures using historical and real-time data to generate Predictive Insights

·       Enables human-like conversations for users such as Chat-GPT to understand the asset better e.g. Imagine Ashish interacting with Maintenance Buddy to query asset health or troubleshoot steps or receive step-by-step instructions for field operations.

·       As we, humans learn over the time with new experiences and data, Gen AI Model also learns from incidents, user feedback to adapt learning to provide better recommendations.

·       Traditional AI stops at predicting when failure might occur but fails to provide why it might happen, but Gen AI bring explain ability, the reasoning behind its predictions or recommendations in a user-friendly manner.

·       Unlike Traditional AI, Gen AI has flexibility to process multiple types of data, including text, images, and sensor data

All these and many other characteristics make Gen AI enable solution approach interesting. We then discussed how Gen AI could address challenges across the asset lifecycle with some examples & use cases

·       Your Team will have vast amount of historical data on performance, maintenance logs, incidents history from over the years, Imagine Gen AI analyze this historical performance data and your environmental conditions to recommend best suitable motor model to procure.

·       Gen AI can provide best industry practices with context to your environmental, recommend motor parameters configurations to avoid installation errors and optimize performance from the start.

·       Specially in operation phase with Gen AI, you could set up a solution to analyze real-time sensor data from Motor (like temperature, vibration, current/voltage readings, energy use) and detect abnormal behavior—such as unusual vibration or heating issues, for recommending preventative actions to avoid unplanned downtime.

·       Solution can process vast amount of data to get data driven insights and present in simpler report or you can build conversational chat bot to derive more out of data.

·       Based on motor’s historical data, operational patterns, and sensor readings Gen AI can predict when a motor will need maintenance or when bearings will be worn out or issues because of vibration etc.,

·       Gen AI can help to calculate Expected-Useful-Life, how to decommission asset safely or recommendations for handling sensitive information about asset post decommissioning or find best suitable replacement model

Conclusion: Empowering Asset Management with Gen AI

In summary Gen AI can be great tool to unlock actionable insights, improve decision making and enhance asset efficiency. However, it is not one size fit for all solution; it depends on specific challenges or use case you are looking for and tailoring solution approach accordingly.

As Ashish and I concluded our conversation, I suggested starting with a pilot project to implement Gen AI on a few critical Motors, using real-time monitoring and predictive maintenance features is right way forward. With this pilot, we could demonstrate tangible improvements and then scale the solution across the entire fleet of Assets.

Stay tuned as we continue to explore the transformative potential of Generative AI in revolutionizing asset management and its use cases!

Empowering for Today and Tomorrow!!