By AutodeskLearning Objectives
- Describe what Emagin HARVI does and why autonomous control matters for water plants
- Explain how reinforcement learning optimizes pumping and treatment operations
- Identify the operating-cost and payback benefits reported for the technology
What Is Emagin HARVI?
Emagin HARVI is an artificial intelligence assistant for water and wastewater control rooms. Built by Emagin — now part of Autodesk Water — HARVI learns from a plant's SCADA operating history and then autonomously optimizes how that plant runs, tuning pump schedules and treatment operations to cut cost and energy while meeting operational targets. Instead of relying on fixed rules or manual operator judgment, HARVI continuously figures out better ways to operate based on what the plant's own data shows works.
The approach has reported roughly 20 percent operating-cost savings with a fast payback at water and wastewater facilities. That combination — meaningful savings with a short time to value — is what makes an autonomous control assistant attractive to utilities under pressure from rising energy prices and tight budgets.
💡Key Concept
Reinforcement Learning for Plant Operations: Reinforcement learning is a branch of machine learning where a system learns by trial and feedback, improving its decisions to maximize a defined goal. Applied to a water plant, the goal might be to minimize energy cost while keeping pressures, flows, and treatment within safe limits. By learning from historical and live operating data, the system discovers control strategies — such as smarter pump schedules — that fixed rules and manual operation would miss.
What Emagin HARVI Does
- Autonomous pump optimization — schedules pumping to reduce energy cost while meeting demand and pressure requirements
- Treatment optimization — tunes treatment operations for efficiency within operational and compliance limits
- Learns from SCADA history — builds its strategies from the plant's own operating data
- Control-room assistance — supports operators with recommendations and autonomous optimization
- Cost and energy reduction — targets operating-cost savings reported around 20 percent with fast payback
How AI Is Applied
HARVI's AI is genuine reinforcement learning, and that is the heart of the product. The system studies a plant's SCADA history to understand how its equipment and processes behave, then learns control strategies that improve a chosen objective — typically minimizing energy and operating cost while respecting the plant's operational constraints. Because pumping and treatment involve many interacting variables and shifting demand, this is exactly the kind of sequential decision problem reinforcement learning is well suited to.
The result is an assistant that can optimize operations autonomously rather than merely reporting on them. Operators keep oversight and the plant's safety limits are respected, but HARVI takes on the continuous optimization work — searching for better pump schedules and treatment settings than static rules or manual tuning would find. The reported roughly 20 percent operating-cost savings reflects the value of running this optimization continuously against real plant behavior.
Who Uses Emagin HARVI
Emagin HARVI is used by water and wastewater utilities that operate pumping and treatment plants and want to cut energy and operating costs. It serves control-room operators and operations managers seeking autonomous optimization, and utility leaders looking for efficiency gains with a fast return. As part of Autodesk Water, it fits within a broader water-infrastructure software portfolio.
Pricing
Emagin HARVI is enterprise software sold on a quote basis, now under Autodesk Water. Pricing depends on plant size, the operations being optimized, and integration scope. Utilities contact Autodesk Water directly for a tailored quote.
Company Details
| Detail | Info |
|---|---|
| Company | Emagin |
| Parent | Autodesk (Autodesk Water) |
| Category | AI control-room assistant for water and wastewater operations |
| AI Method | Reinforcement learning on SCADA operating data |
| Reported Benefit | Around 20 percent operating-cost savings with fast payback |
| Website | autodesk.com |
Strengths
- Genuine reinforcement learning — real AI that learns control strategies, not fixed rules
- Autonomous optimization — continuously tunes pumping and treatment rather than just reporting
- Meaningful savings — reported roughly 20 percent operating-cost reduction with fast payback
- Learns from real data — builds strategies from the plant's own SCADA history
- Autodesk backing — part of the broader Autodesk Water infrastructure portfolio
Limitations and Considerations
- Data dependence — effectiveness relies on good SCADA history and reliable live data
- Trust and oversight — autonomous control requires operators to gain confidence and keep safety limits enforced
- Plant-specific tuning — value depends on the specifics of each plant's equipment and processes
- Integration effort — connecting to SCADA and operational systems takes setup
Key Takeaways
- Emagin HARVI is an AI control-room assistant, now part of Autodesk Water, that uses reinforcement learning to optimize water-plant operations
- It learns from SCADA history to autonomously tune pump schedules and treatment, reporting around 20 percent operating-cost savings with fast payback
- The AI is genuine reinforcement learning — it discovers control strategies fixed rules and manual operation would miss
- Best for water and wastewater utilities seeking autonomous, cost-cutting optimization of pumping and treatment
