AVEVA and Databricks: We are solving real problems with AI—today, not tomorrow
Posted: March 17, 2026
As we enter 2026, industrial companies are done waiting for AI to mature. What they want now is simple: AI that actually works with their operational realities, delivers measurable improvements, and scales across plants, fleets, and global operations.
That’s exactly where the AVEVA and Databricks collaboration is breaking new ground.
For years, industrial data has been trapped in silos, locked behind legacy systems, or too complex to operationalize at scale. At the same time, corporate data science teams have lacked the deep operational context needed to deliver real outcomes on the factory floor, in the mine, or across energy networks.
The convergence of OT and enterprise AI is finally happening in a meaningful, production‑grade way and customers are feeling the impact today, not in future roadmaps.
Together, AVEVA and Databricks platforms are unifying industrial data, governance, and AI into a single, modern architecture that accelerates decisions, improves reliability, and unlocks new financial value. This isn’t about experimentation; it’s about execution.
The industrial data paradox
Yet despite this progress, a stubborn reality remains: the data foundations inside most industrial companies were never designed for modern AI.
Understanding why is essential to understanding how to fix it.
- Data gravity : First, asset-intensive companies are generating vastly more data than ever before but struggle to integrate, contextualize, and operationalize it at scale.
- The “hype” gap : While LLMs and AI have raised expectations, many organizations still face significant gaps in real-world industrial deployment.
Don’t overcomplicate AI
One year on, our collaboration with Databricks has progressed to hands-on execution with customers. Here's what we've learned: radical collaboration isn't about chasing moonshots or unproven technology.
Instead, there is massive value in improving the base environment for the customer— modernizing data platforms and improving governance, while upping scale and speed.
What radical collaboration looks like in practice
By connecting CONNECT (deep domain expertise and operational data) with the Databricks Data Intelligence Platform, we transform how decisions are made.
The value isn't coming from hypothetical industrial AI use cases. It's coming from existing dashboards and analytics that simply need to work better, faster, and more reliably.
Take a concrete example: one customer was refreshing operational dashboards twice a day. Through our joint solution, they now refresh 50 times a day—every 15 minutes instead of every 12 hours. That's modern data architecture delivering immediate operational advantage.
Customers consistently ask us how to combine their Lakehouse with leading operational data and AI to drive real industrial outcomes. Partnering with AVEVA allows us to deliver exactly that through Delta Sharing—helping customers like Agnico Eagle and EDP Renewables power the next generation of industrial intelligence.
The proof is in production : Customer impact
Here's what tightknit teamwork actually delivers:
SQM: Driving yield in specialty chemicals
SQM, a $4.5 billion specialty chemicals producer, faced a familiar challenge: operator decisions based on experience, not data, with siloed historian and lab data.
The solution: SQM adopted CONNECT to centralize sensor data , exposing it to Unity Catalog in Databricks for governed, enterprise-wide access.
The results:
- 1% increase in Nitrate yield
- 10-15% reduction in process variability
- 90% accuracy in predictive models
EDP Renewables: Near real-time profitability
As part of its digital transformation, EDP Renewables needed to unlock the value of operational data to make faster, smarter decisions.
Its fragmented, on-prem architecture with manual data transfers and updates only twice a day was slowing decisions and increasing risk.
The solution: A hybrid architecture that streams AVEVA™ PI System™ data directly into the Databricks Lakehouse.
The results:
- 50x more frequent data updates
- Faster anomaly responses through near real-time data
- Improved profitability by enabling accurate energy price adjustments
The way ahead
We aren't just talking about the future; we are building it. We're building on proven wins like SQM and EDP Renewables, with plans to scale these approaches across additional plants, business lines, and industries globally.
Our focus remains on a clear sequence: get your data house in order, then share it with enterprise-grade governance. This is radical collaboration: continuous improvement, OT/IT scaled collaboration, and an unwavering focus on customer outcomes.
Join the conversation
We’re excited because these results are visible and measurable today. I’d love to hear your thoughts on where industrial AI is headed next. Connect with me on LinkedIn
Ready to modernize your industrial data foundation?
Contact us to discuss how CONNECT and the Databricks Data Intelligence Platform can accelerate your operational performance.
Related blog posts
Stay in the know: Keep up to date on the latest happenings around the industry.
