Skip to content
Randhir VermaPMP®
All articles

Smart Utilities5 min read

Sequence Before Software: DERMS in the Field and India's 2032 Roadmap

A feeder-level DERMS enters the field in the US, Georgia Power ties one to firm tariffs, and the CEA places DERMS last. What the ordering means for DISCOMs.

Randhir Kumar Verma, PMP®
Published 26 Oct 2026

Two announcements from the past six weeks sit at opposite ends of the distribution-automation conversation. In the United States, a vendor is putting a feeder-level DERMS into the field, while a utility pairs one with firm, dispatchable tariffs. In India, the Central Electricity Authority has published a roadmap that places DERMS firmly in its last phase. Read together, they show that the hard part of ADMS and DERMS is the sequence of capabilities underneath them, not the software label.

A DERMS that dispatches at the feeder

On September 8, 2026, T&D World reported that FlexEnergi has announced field deployment of its Ora Edge DERMS, which it says utility partners have adopted. The platform manages flexible load at feeder and substation level for congestion relief, load shaping, load building and dispatchable capacity, across residential, small business, agricultural and large commercial customers.

Two design choices stand out. First, dispatch can be triggered by wholesale price signals, weather events or direct commands from a utility's SCADA or ADMS. That makes the ADMS an upstream command source for customer-side resources, not a separate silo. Second, the commercial model is Capacity-as-a-Service: utilities buy flexible capacity in defined megawatt blocks, and the vendor is paid on delivered performance while handling programme design, customer recruitment, settlements and payments. The article contrasts this with traditional demand response, which is built around periodic system-wide peak events, whereas this approach aims at continuous, locational optimisation.

Firm capacity as a tariff

A second pattern comes from Georgia Power. A DER weekly digest from Utility Community reports that, under a Georgia PSC-approved stipulation tied to its most recent integrated resource plan, the utility is deploying a DERMS for visibility, forecasting, control and optimisation of behind-the-meter resources, with spending projected for 2025 through 2028.

The utility's own DER Customer Program FAQ describes the commercial structure. Under the Resiliency Asset Service tariff, Georgia Power designs, installs, owns, operates and maintains a dispatchable resource behind a commercial or industrial customer's meter, and the customer pays a fixed monthly charge. Under the Demand Response Credit tariff, customers with at least 1,000 kW of annual peak load receive a bill credit in exchange for the utility's right to use the resource during grid reliability events, for up to 200 hours a year, with 30 minutes' notice. The page describes a controls gateway for remote monitoring and dispatch; the DERMS link comes from the digest rather than the FAQ.

The digest's own caveat applies: because the resiliency framing puts customer outage value first, realised grid capacity depends on enrolment depth and how often the resources are called.

India sequences DERMS last

The CEA's Unified Framework for Digitalisation of the distribution sector, summarised by Power Peak Digest on October 19, 2026, takes the opposite approach to ordering. Its guiding rule is that a function should not be automated until it is observable, data-quality assured, controllable and operationally owned.

The roadmap follows that rule. The 2026 to 2029 foundation phase covers readiness assessment, master data and GIS, connectivity, AMI, HES and MDMS, feeder and transformer metering, SCADA or RT-DAS where needed, and cyber and governance baselines. DMS and ADMS pilots arrive in 2029 to 2032. ADMS scale-up, state estimation, FLISR, Volt/VAR optimisation and DERMS are scheduled for 2032 to 2035. Agentic autonomous control is classed as a research and sandbox technology.

The framework also separates deterministic real-time control from analytics and AI. Protection and safety functions stay deterministic, and AI starts as decision support, with human override, fail-safe modes and auditability required for any later autonomy. The same summary notes that India has sanctioned 20.33 crore smart meters and installed 3.9 crore under RDSS, and that rooftop solar grew 123% year on year in 2025, adding 7.1 GW. The pressure from distributed generation is therefore arriving before the tooling to manage it.

What it means for utilities

The US examples show what a DERMS is for: a dispatchable, locational resource with a contract and a settlement model behind it. The Indian framework shows what must exist first. For DISCOMs, a few implications follow.

  • Feeder visibility is the gating item. Feeder and transformer metering, GIS quality and SCADA coverage decide whether any later ADMS can be trusted.
  • Commercial design matters as much as software. Both US examples define who owns the asset, who is paid and for what delivered performance.
  • Rooftop solar will not wait for 2032. With DER growth running ahead of the roadmap, DISCOMs may need interim, rule-based controls on high-penetration feeders, kept deterministic and auditable.
  • Keep control and analytics separate. The CEA's split between deterministic control and AI decision support is a sound design rule for any ADMS procurement.
  • Plan shared services carefully. The framework allows smaller DISCOMs to share non-real-time functions while keeping critical OT control in-house.

Key takeaways

  • A DERMS is being sold as a locational, performance-paid capacity service, triggered by price, weather or ADMS and SCADA commands.
  • Georgia Power shows the utility-owned alternative: the asset is owned and dispatched by the utility under paired tariffs, with a 200-hour annual call limit.
  • The CEA places ADMS pilots in 2029 to 2032 and DERMS in 2032 to 2035, after metering, data and SCADA foundations.
  • Rooftop solar growth of 123% in 2025 means DER management needs may arrive well ahead of that schedule.
  • Realised capacity, not announced capacity, is the number to track: enrolment, call frequency and feeder data quality decide it.

Sources

ShareLinkedInWhatsAppX

Enjoyed this? Get the next one by email

Occasional notes on smart metering, AMI and AI in utilities. No spam; unsubscribe in one click.

By subscribing you agree to the Privacy Policy.

Comments

No comments yet. Be the first to share a thought.

Leave a comment

Your email is never shown. Comments appear after they have been reviewed.