The traditional structure of the electricity system has a clear division of labor. Transmission system operators (TSOs) or independent system operators (ISOs) balance generation and load at the bulk power level. Distribution utilities deliver that power to end customers over radial distribution feeders. The distribution utility's job, in this model, is largely passive: maintain the hardware, respond to outages, and accept whatever load the customers choose to place on the system.
That model was built for a grid where power flowed one direction and customers were passive loads. Neither assumption holds any longer. Behind-the-meter solar, battery storage, managed EV charging, and demand-responsive commercial loads mean that distribution networks now contain substantial generation, storage, and controllable demand capacity that, if coordinated, could provide meaningful services both to the local distribution system and to the bulk grid above it.
The concept of the Distribution System Operator (DSO) is the emerging organizational and technical framework for making that coordination possible. But the DSO model is not a single well-defined thing. Understanding what it means in practice requires separating the conceptual vision from the near-term operational realities.
What the DSO Concept Actually Proposes
At its most developed, the DSO model proposes that distribution-level operators should function similarly to how ISOs function at the transmission level: actively managing a network with multiple competing participants, procuring flexibility services through markets or structured procurement processes, and providing neutral access to distribution infrastructure for DER operators and aggregators.
This full vision, sometimes called a "Distribution TSO" model after the European regulatory structure where it has received the most serious policy attention, involves the distribution operator running localized flexibility markets where DER aggregators bid in curtailment and dispatch services to relieve distribution constraints. Under this model, the distribution operator would issue congestion signals at the feeder level, and aggregators would respond with bids to reduce or shift load in congested areas.
The United States context is somewhat different. In most of the US, distribution utilities are vertically integrated or regulated monopolies. The ISO/RTO structure handles bulk power markets, but distribution remains largely under utility control without an independent market operator. The path to a full DSO model in the US requires regulatory changes that vary significantly by state and are, in most places, still in early discussion stages.
We're not saying the full DSO market model is imminent in most US jurisdictions. The operational improvements that are achievable now, within the current regulatory structure, are worth examining on their own terms.
What Distribution Operators Are Actually Doing Now
Several regulatory proceedings and utility programs represent practical steps toward DSO-like capabilities even within current structures.
California's distribution resource planning process (under CPUC proceedings) has pushed utilities to identify distribution deferral opportunities, where DER deployment and management can defer or avoid traditional infrastructure upgrades. This requires the utility to develop much more detailed models of distribution system constraints and DER capacity by location than they historically maintained.
New York's Reforming the Energy Vision (REV) initiative went further, explicitly promoting the concept of a Distributed System Platform Provider (DSPP) as an evolution of the distribution utility role toward DSO functions, including transactive energy mechanisms at the distribution level. Implementation has been uneven across utilities, but the regulatory framework created space for pilot programs and procurements that wouldn't have happened otherwise.
At the operational level, utilities running active DER management programs are already performing DSO-like functions without the formal designation: issuing curtailment orders to aggregated solar and storage, coordinating with demand response program managers to reduce load on congested feeders, and using real-time visibility into behind-the-meter assets to inform distribution operations decisions. The infrastructure for more formal DSO operations is being built incrementally.
The Data Infrastructure Gap
The fundamental challenge for any distribution operator moving toward DSO functions is data infrastructure. Traditional distribution operations required substation-level monitoring, trouble call systems, and field crew dispatch. Active DER management requires near-real-time visibility into individual assets at the feeder level, including behind-the-meter assets that the utility may not own or directly control.
Advanced metering infrastructure (AMI) is the baseline. Most large US utilities now have AMI deployed across their residential customer base, providing 15-minute interval data with 24-hour latency in most implementations. That's sufficient for billing and aggregate load analysis, but the 24-hour latency is not useful for real-time distribution operations. Some AMI implementations support near-real-time meter reads, but network backhaul infrastructure and data management systems are often not architected for that operational use case.
Beyond AMI, a distribution operator managing significant DER capacity needs visibility into individual asset states: battery state of charge, solar inverter generation, EV charger status, and automated demand response enrollment status. These data streams come from a variety of vendor systems, each with their own APIs, data formats, and telemetry latencies. Building and maintaining integrations across those systems is a significant engineering undertaking that most distribution utilities have been reluctant to invest in ahead of clear regulatory mandates.
This is where the software infrastructure for DSO operations is being actively built now. The specific requirement is not just data collection but the ability to run optimization models over that data in 15-minute intervals, produce dispatch signals to controllable resources, and maintain a real-time picture of distribution system state. That combination of requirements defines the core technical function of a DSO platform.
Feeder-Level Locational Value
One of the most operationally important DSO concepts is locational marginal value at the distribution level. At the bulk power level, ISOs calculate locational marginal prices (LMPs) that reflect the value of energy at each transmission node, including congestion costs when transmission constraints bind. At the distribution level, similar locational logic applies but has historically been ignored because power flows were assumed to be largely predictable and unidirectional.
With significant DER penetration, distribution-level congestion is real. A feeder serving a dense residential area with high rooftop solar penetration may be export-constrained in the middle of a sunny day: generation exceeds local load and the feeder's export capacity is limited. The value of controllable load or battery storage on that feeder during that period is very high locally, because it can absorb excess generation that would otherwise require curtailment, but that value is not reflected in any current market price signal received by those resources.
DSO models, even in their current partial implementations, are beginning to address this through locational procurement: utilities issuing solicitations for demand flexibility or storage capacity in specific feeder areas where they project constraint risk, rather than procuring system-wide capacity that may not provide value in the locations where it's needed. This locational specificity changes what an aggregator needs to know when building a portfolio: not just how much capacity they can assemble, but whether their assets are in the right locations to provide value to a distribution operator facing specific feeder constraints.
What This Means for Dispatch Software
For operators building DER portfolios with an eye toward DSO markets, the implication is that dispatch optimization needs to become location-aware at the feeder level, not just portfolio-level. The value of dispatching a battery is not the same at every location. A battery on a congested feeder at a specific time has higher locational value than one on an unconstrained feeder, and that value is what a mature DSO market would price.
Voltsynth's dispatch model incorporates feeder-level constraints when that data is available from the utility or grid operator. When feeder topology and loading data are provided, the optimizer will weight dispatch decisions to prioritize resources in locations where the distribution value is highest, not just minimize aggregate cost or maximize aggregate revenue. This alignment between portfolio dispatch and distribution system needs is what makes DER aggregation genuinely useful to the distribution operator, rather than just a resource that happens to be connected to their network.
The DSO model is being built in practice, incrementally, through a combination of regulatory change, utility operational investments, and the technical infrastructure that DER operators and aggregators are building to participate. The full vision of a competitive distribution market with locational pricing and neutral platform access remains years away in most US jurisdictions. The operational foundations, better data infrastructure, locational procurement programs, and real-time DER visibility, are being put in place now.