DER Value Stack Sequencing Efficiency (DVSSE)

GMI-EU-038Energy & UtilitiesResearch-Forward

Definition

Measures the percentage of theoretically available DER value captured by the AI's sequential dispatch decisions, compared to the global optimum achievable with perfect foresight.

The Measurement Gap

Traditional DERMS performance reporting tracks revenue earned from each dispatch event without providing a benchmark for what could have been earned. A program capturing $50,000 monthly in grid service revenue may be capturing 90% or 55% of theoretically available value with the same fleet. Without DVSSE, those scenarios are indistinguishable to program managers and DERMS vendors alike.

The specific failure point is multi-service value stack interaction. DER batteries can provide energy arbitrage, frequency regulation, capacity commitment, and demand charge reduction simultaneously within an operating day. These services have different dispatch profiles, revenue rates, and timing requirements. The optimal full-day dispatch positions the battery for the highest-value service at each moment. A greedy optimizer may execute frequency regulation in the morning (depleting the battery) when the afternoon demand charge reduction would have been worth significantly more per kWh dispatched. The optimizer was locally rational at each interval and globally suboptimal across the day.

DVSSE enables the specific decision of whether to invest in improved DERMS optimization algorithms (longer lookahead horizons, multi-day planning, better price forecasting) by quantifying the dollar value of the performance gap. If DVSSE shows the optimizer is capturing 65% of available value and the fleet generates $2M annually, the theoretical improvement opportunity is $600K, a number that either justifies or does not justify algorithm investment.

Without DVSSE, DERMS vendors claim optimization performance without a verifiable benchmark. Utilities accept reported revenue as the performance measure without knowing how much better performance was achievable.

AI-driven retroactive optimization engines make the perfect-foresight benchmark computationally feasible as a daily analytical process separate from the operational DERMS.

Formula

DVSSE = Revenue_actual / Revenue_perfect_foresight

Variables:

  • Revenue_actual: actual revenue from all DERMS dispatch decisions across all value streams for the operating period (energy arbitrage, frequency regulation, capacity payments, demand charge reduction ($/period)) Source: DERMS settlement records, ISO/RTO market settlement, utility tariff calculation
  • Revenue_perfect_foresight: revenue achievable from ex-post optimal dispatch computed retroactively using realized prices and actual grid conditions across all value streams for the same period, subject to the same physical battery constraints (SoC limits, ramp rates, degradation limits, capacity commitments) ($) (Source: retroactive optimization model run against actual market and grid data; NOT available from operational systems) requires a separate analytical optimization engine
  • Battery constraints applied to perfect-foresight calculation must match actual fleet constraints: SoC floor and ceiling, maximum discharge rate, minimum reserve for committed capacity obligations

Calculation Frequency: Daily for continuous monitoring; weekly aggregation for reporting
Output: Ratio, 0 to 1.0. DVSSE of 1.0 = the optimizer dispatched identically to perfect foresight. DVSSE above 0.75 is reasonable for real-time optimization. DVSSE below 0.60 indicates significant sequencing improvement is available. Track DVSSE by value stream: the optimizer may sequence well for arbitrage but poorly for ancillary services. The perfect-foresight benchmark is an achievable upper bound, not a realistic target: some gap is inherent in any real-time system.

Proposed formula, pending validation pilots.

At a Glance

Category
Energy Economics and DER Fleet Management

Decision Supported

Is the DERMS leaving significant value on the table through greedy interval-by-interval optimization, and how much would a better sequencing algorithm be worth?

← DER Dispatch Tolerance Index (DDTI)Net Cycling Economic Value (NCEV) →

Assess Your Organization

The AI-GPS Assessment evaluates your AI governance maturity and recommends KPIs like this one for your implementation roadmap.

Take the Free AI-GPS Assessment