AVEVA™ PI System™: From trusting operational data to expanding its value

Written by Harpreet Gulati, Head of Information Management (PI) business, AVEVA
Believe it or not, my first job was as a developer for process simulation software. I had to handle rigorous mathematics and heavy coding. As I shifted to different roles and started to work more closely with customers, I quickly learned that no matter how sophisticated its coding, any process simulation model would only be as good as the data that users feed it.

This is true for any type of modeling, from rigorous first-principles to sophisticated AI. You need reliable data. Feed your model with garbage data—you will have garbage outcomes!

It’s more crucial than ever to keep that principle in mind now that plant operations rely on multiple models and algorithms built upon plant data to meet key performance indicators (KPIs). Those KPIs determine the course of action for operations each day, whether it’s planning production for the next month or year or identifying and responding to abnormal process behavior. The question is:

How do you know you can trust your data?

If your company has adopted AVEVA PI System, your data is fully visible and you have foundational capabilities to create a reliable operational data management environment. In your case, the question is:

How can you expand the value of your operational data?

The answer will be different depending on the stage of your digital transformation and the maturity of your AVEVA PI System adoption. But, there are several common approaches everyone can benefit from:

• Understand the purpose of your data – what actions will it inform?
• Identify what relationships among data sets will deepen your understanding.
• Catalog the gaps in your data and the assumptions you have to make.
• Create analytics for KPI generation and event detection.
• Make data and insights accessible—evaluate options for dashboards, cloud connection, sharing, etc.

Asset-heavy industries, such as chemicals, oil and gas, power, and mining, must improve their use of data to face their operational and business challenges. Enhanced data infrastructure, models, and analytics let these industries:

• Centralize disparate data to provide insight across the value chain.
• Use AI-based analytics to move from reactive to predictive and prescriptive operations and maintenance.
• Manage complex data sets with reliable models to quickly de-bottleneck and optimize operations with real-time data.
• Increase asset utilization, implement product tracing, and improve productivity.