Foreword

This document has been prepared by the OpenEAC Alliance, a collaborative group of energy engineering and measurement and verification professionals focused on creating transparent and standardized methods to measure the environmental impacts of demand side interventions. These methodologies are designed to align with best practices and reflect input from industry experts. The OpenEAC Alliance aims to enhance accountability and consistency in documenting energy and carbon savings from distributed energy resources (DERs).

About the OpenEAC Alliance

The OpenEAC Alliance is an organization of professionals dedicated to standardizing the measurement and verification of DERs. By focusing on clarity and rigor, the Alliance aims to overcome challenges that have hindered the proper valuation of DERs for their climate benefits. Central to this effort is the Energy Attribute Certificate (EAC), a standard that provides a detailed, auditable record of energy project impacts, including attributes such as type, timing, and location of energy production or reduction.

The OpenEAC Alliance collaborates to develop open-source methodologies, ensuring that DER projects can be accurately measured, verified, and valued in a way that promotes trust and drives market adoption.

Introduction

This document introduces a standardized approach for quantifying the change in a customer’s electricity bill that results from a demand-side intervention (for example, energy efficiency, electrification, solar, storage, or load shifting). The approach prices an hourly electricity load shape against the customer’s published utility tariff, separating the bill into its energy, demand, and fixed components. By pricing both a counterfactual (baseline) load shape and a reporting (observed) load shape against the same tariff, the difference yields the bill savings attributable to the intervention.

The intent is to reflect how a real utility bill is constructed — including tiered energy blocks, time-of-use periods, and demand charges — closely enough to produce a credible savings figure, while being explicit about the assumptions where exact replication of a bill is not possible. Because tariffs are nonlinear (tiers and demand charges), savings are always computed by differencing two complete bills, never by applying a marginal price to the difference in consumption.

Scope

This document outlines the methodology, assumptions, and data sources used to translate an hourly electricity load shape and a utility tariff into a bill, and to compute bill savings as the difference between two priced load shapes.

This methodology supports:

• Energy charges, including flat (single-rate), tiered (cumulative usage blocks), and time-of-use (TOU) energy rates.

• Tier thresholds expressed in energy (kWh) and in demand-scaled energy (kWh per kW of billing demand), including variants whose allotment resets each day.

• Demand charges, including flat monthly demand, time-of-use demand, and coincident demand, with seasonal month applicability, ratchet provisions, and proration of off-length billing periods.

• Fixed monthly charges.

• Billing-period determination from metered bill data, with a calendar-month fallback when bill periods are unavailable.

This methodology does not currently support:

• Tier thresholds that require quantities that cannot be derived from electricity consumption alone — specifically thresholds expressed per unit of connected horsepower (kWh/hp) or per unit of apparent power (kWh/kVA).

• Rate structures that mix tier-threshold units within a single time-of-use period, as the accumulation window for such a period is not well defined.

• Sell-back or net-metering credits within the energy rate structure.

• Pass-through and rider charges layered on top of the base tariff (fuel-cost adjustments, renewable/transition/grid riders, efficiency-program surcharges, franchise fees, and taxes). See “Assumptions and Accuracy.”

Where a tariff falls outside the supported scope, the calculation degrades rather than fails: energy charges fall back to a documented fixed volumetric rate, and the demand component is omitted. Any such substitution is recorded so that the resulting savings figure can be interpreted accordingly.

Normative References

The following documents are essential for the application of this specification. For references with no date, the latest edition applies:

• OpenEI Utility Rate Database (URDB): https://openei.org/wiki/Utility_Rate_Database

• OpenEI URDB Web Services (rate retrieval API): https://openei.org/services/doc/rest/util_rates/

Terms and Definitions

• Tariff (rate plan): The published schedule of charges a utility applies to a customer class, comprising energy, demand, and fixed components.

• Energy charge: A charge proportional to electricity consumed, expressed in dollars per kilowatt-hour ($/kWh).

• Demand charge: A charge proportional to peak power drawn during a period, expressed in dollars per kilowatt ($/kW).

• Fixed charge: A recurring charge that does not vary with usage, such as a monthly meter or service charge.

• Billing period: The interval over which a single bill is computed; typically a monthly cycle defined by meter read dates.

• Billing demand: The peak power, in kW, used to assess the demand charge and to scale demand-based energy tiers for a billing period.

• Tier (usage block): A segment of consumption priced at a distinct rate. A tier’s upper bound is its threshold; consumption accumulates across a billing period (or, for daily tiers, across a day) and is priced at successive tiers’ rates as thresholds are crossed.

• Tier threshold unit: The unit in which a tier’s threshold is expressed. “kWh” denotes an absolute energy block; “kWh per kW” denotes an allotment that scales with billing demand; the “daily” variants reset the allotment each day. The threshold unit describes the size of the block, not the price, which is always $/kWh. Where the tariff data omits the threshold unit, an absolute energy (kWh) threshold is assumed.

• Time-of-use (TOU) period: A labeling of hours (by month, day type, and hour of day) into periods — such as peak, partial-peak, and off-peak — each with its own rates.

• Coincident demand: A demand charge assessed on the customer’s load at the time of the system or utility peak, rather than the customer’s own peak.

• Ratchet: A provision that holds billing demand at no less than a set percentage of a prior peak for a number of months, so a single high peak raises demand charges in subsequent periods.

• Proration: Scaling a period-based charge (demand, fixed, or minimum) by the ratio of the billing period’s length to a standard 30-day month, applied when the period falls outside the standard window. Energy (kWh) is never prorated.

• Counterfactual load shape: The hourly electricity that would have been consumed absent the intervention.

• Reporting load shape: The hourly electricity actually consumed during the reporting period.

Data Requirements

For a bill to be computed, the following are required:

• An hourly electricity load shape covering the period to be priced. Consumption is recorded per interval; all tier accounting is performed in kilowatt-hours, with explicit conversion from the load shape’s native interval units.

• A mapped utility tariff for the site, identified in the Utility Rate Database.

• Billing periods for the site. Where metered bill read dates are available they define the billing periods; otherwise calendar months are used as a fallback, and this substitution is recorded.

Methodology

1 Determine Billing Periods

1.1 Establish the set of billing periods covering the load shape. Where metered bill read dates are available, each distinct read interval defines one billing period.

1.2 Where no bill read dates are available, fall back to calendar months spanning the load shape. This fallback is recorded, as misalignment between a calendar month and the true billing cycle is a source of difference from the actual bill.

2 Calculate Energy Charges

2.1 Energy charges are volumetric: every kilowatt-hour consumed is priced, regardless of the billing period’s length, so the energy component is never pro-rated.

2.2 Assign each hour to its time-of-use period.

2.2.1 Using the tariff’s weekday and weekend period schedules, map each hour of the load shape (by month, day type, and hour of day) to the energy period that governs it.

2.3 Determine whether the energy rate is tiered.

2.3.1 A rate is tiered if any period defines more than one usage block or sets a usage threshold. Flat single-rate energy is priced directly: each hour’s consumption multiplied by the applicable period’s rate (energy rate plus any adjustment).

2.4 Accumulate consumption within the applicable allotment window.

2.4.1 For tiered rates, consumption is accumulated cumulatively, in kilowatt-hours, so that thresholds are assessed against running usage, not a single interval:

2.4.1.1 For tiers measured in absolute energy or scaled by demand, consumption accumulates over the billing period.

2.4.1.2 For tiers whose allotment resets daily, consumption accumulates over each calendar day within the billing period. The daily variant resets the accumulation only; where the threshold is also demand-scaled, the scaling demand remains the billing period’s peak.

2.4.2 Accumulation is tracked separately per time-of-use period, so that only the hours governed by a period draw down that period’s tiers.

2.4.3 An interval with missing consumption contributes nothing to the running total and is left unpriced; it does not reset or otherwise perturb the accumulation.

2.5 Scale demand-based tier thresholds.

2.5.1 A tier threshold expressed per kilowatt of demand is converted to an absolute energy allotment by multiplying it by the billing demand (the period’s peak demand) for the billing period. Thresholds already expressed in energy are used directly. The same peak demand value is used here and in the demand-charge calculation, so the two components cannot disagree about a period’s demand.

2.6 Price each hour at its applicable tier.

2.6.1 As cumulative consumption crosses a threshold, subsequent consumption is priced at the next tier’s rate. Each interval is charged entirely at the rate of the tier its running total falls within after the interval is added (energy rate plus any adjustment); intervals are not split across a threshold.

2.7 Handle overflow past a capped final tier.

2.7.1 A well-formed tariff leaves its final tier uncapped. Where the tariff data nonetheless caps the final tier and cumulative consumption exceeds it, the excess is priced at that final tier’s rate, and the condition is recorded.

3 Calculate Demand Charges

3.1 Determine billing demand per period.

3.1.1 For each billing period, establish the peak demand, in kW, from the load shape. This is the same peak used to scale demand-based energy tiers.

3.2 Resolve the billing period’s season. A demand charge is assessed once per billing period, so the season is fixed for the whole period — taken from the month in which the period begins — and applied uniformly to every hour in it; hours are not re-seasoned individually by their own calendar month. The period’s season selects which months’ demand rates and time-of-use period mapping apply, and, where a demand charge applies only to certain months, whether it applies at all. Fixing the season per period ensures that a billing cycle straddling a seasonal boundary is priced under a single season, rather than having its peak charged once under each season it touches — which would double-count the demand charge.

3.3 Apply ratchet provisions.

3.3.1 Where the tariff specifies a ratchet, the billable demand for a period is held to no less than the ratchet percentage of the relevant prior peak.

3.4 Apply the applicable demand structure.

3.4.1 Flat demand charges price the period’s billing demand at the flat rate; time-of-use demand charges price the peak within each demand period, with each hour assigned to its demand period using the billing period’s season; coincident demand charges price the customer’s load at the system peak. The resulting charge is booked on the first interval that reaches the relevant peak; when several intervals tie at the peak, the whole charge is assigned to that earliest interval rather than split across them, so a single interval carries the period’s demand charge.

3.5 Prorate off-length billing periods.

3.5.1 A demand charge is defined against a standard 30-day month. When a metered billing period falls outside the standard window (approximately 27–33 days) — typically an estimated or catch-up read that spans extra days, for example when a monthly read is missed and trued up at the next read — the demand charge is scaled by the ratio of the period’s length to 30 days. Periods inside the standard window are treated as a normal month and are not scaled. This proration is applied to the demand charge only; energy (kWh) is billed on actual consumption and is never prorated. (Fixed monthly and minimum charges, where modeled, would prorate the same way.)

4 Calculate Fixed Charges: Apply any fixed monthly or service charges for each billing period.

5 Compute Bill and Savings: The total bill for a load shape is the sum of its energy, demand, and fixed charges across all billing periods.

5.1 Price the counterfactual (baseline) load shape and the reporting (observed) load shape against the same tariff and the same billing periods, using the full methodology above for each. Pricing complete bills on each side ensures that nonlinear effects — a tier crossed in one scenario but not the other, or a shaved peak — are reflected in the savings.

5.2 Bill savings is the counterfactual bill less the reporting bill. Component-level savings (energy, demand) may be reported separately by differencing the corresponding components.

Assumptions

The following assumptions shape how closely the computed bill — and the savings derived from it — matches an actual utility bill, listed roughly from largest to smallest expected effect:

1 Demand is derived from interval energy. Peak demand is inferred from the highest-energy interval in a period. When the load shape is hourly, this understates the utility’s typical fifteen-minute demand peak, which lowers both demand charges and any demand-scaled energy allotment. This is the most significant systematic source of difference.

2 Counterfactual load shapes are smoother than metered data. A counterfactual produced by a regression model lacks the short-duration peak excursions present in observed data, so its inferred peak demand — and therefore its demand charges and demand-scaled tier allotments — is biased low. Demand-related savings are correspondingly understated. The error direction is conservative, and the assumption is accepted on that basis.

3 Tariff-defined charges only. The bill reproduced here comprises the tariff’s energy, demand, and fixed charges. Pass-through and rider items that ride on top of the base tariff — fuel-cost adjustments, renewable/transition riders, efficiency-program surcharges, and franchise or tax percentages — are outside the priced tariff and are not included, so the computed bill is systematically lower than a real bill total by the sum of those riders. The shortfall grows with consumption, so it is largest on high-usage periods. Savings (a difference of two bills on the same tariff) are far less affected than the absolute bill.

4 Demand proration of off-length periods. A billing period outside the standard window has its demand charge scaled by days/30, approximating a utility’s treatment of an over- or under-length read. This differs from a utility that instead bills a separate demand charge for each calendar month within a multi-month read; the day-weighted scaling and the per-month treatment converge for a roughly two-month period but can differ when monthly peaks within the period diverge.

5 Seasonal assignment of a straddling billing period. A billing period that spans a seasonal boundary is assigned a single season — that of the month in which it begins — when selecting demand rates and time-of-use period mappings. A utility that instead keys the bill’s season off its read (end) date, or day-weights the seasonal rates across the period, would assess a different demand rate for such a period; the resulting difference is bounded and confined to boundary-crossing cycles.

6 Billing-cycle alignment. When billing periods fall back to calendar months but the true cycle differs, the peak and the cumulative tier totals are computed over a shifted window.

7 Tier allotments scale with actual peak demand. Demand-scaled energy tiers use the period’s actual peak demand rather than a ratchet-adjusted billing demand. This differs from the bill only for tariffs that ratchet the energy-tier allotment, and only in months following a high peak.

8 Whole-interval tier assignment. Each interval is priced entirely at the tier its cumulative total falls within; an interval that spans a threshold is not split. The resulting error is bounded by at most one interval per threshold crossing.

Appendix

Appendix 1: Supported Tier-Threshold Units

Threshold unit	Meaning	Supported
kWh	Absolute energy block (e.g., first 1,000 kWh)	Yes
kWh per kW	Energy allotment scaled by billing demand	Yes
kWh daily	Absolute energy block resetting each day	Yes
kWh per kW, daily	Demand-scaled allotment resetting each day (scaling demand remains the billing period’s peak)	Yes
kWh per horsepower	Allotment scaled by connected horsepower	No — requires data not derivable from energy
kWh per kVA	Allotment scaled by apparent power	No — requires data not derivable from energy
(unspecified)	Threshold unit omitted by the tariff data	Treated as kWh

All tiers within a single time-of-use period must share one threshold unit; periods mixing units are out of scope.

Appendix 2: Worked Illustrations

Appendix 2.1: Flat energy and demand

Consider an industrial tariff with a flat energy charge of $0.08874/kWh (plus a $0.00029/kWh adjustment), a flat demand charge of $9.67/kW, and a fixed monthly charge. For a standard (≈30-day) billing period:

• The energy charge is the period’s total consumption multiplied by $0.08903/kWh (rate plus adjustment), since the energy rate is a single flat block.

• The demand charge is the period’s peak demand, in kW, multiplied by $9.67/kW, booked on the single interval that reached that peak.

• The fixed charge is added once.

The period’s bill is the sum of the three. Repeating this across the baseline and reporting load shapes and differencing the totals yields the bill savings.

Appendix 2.2: Demand-scaled energy tiers

Consider a tariff whose energy rate has two tiers with thresholds expressed per kilowatt of billing demand: the first 200 kWh per kW at $0.10/kWh, and all remaining energy at $0.15/kWh. For a billing period whose peak demand is 100 kW:

• The tier 1 allotment is 200 kWh/kW × 100 kW = 20,000 kWh.

• Consumption is accumulated hour by hour across the billing period; the first 20,000 kWh of cumulative consumption is priced at $0.10/kWh.

• The hour in which the running total crosses 20,000 kWh, and every hour after it, is priced at $0.15/kWh (whole-interval assignment).

Because the allotment scales with the period’s peak, an intervention that reduces peak demand shrinks the cheap-tier allotment even as it reduces the demand charge — an interaction that is captured only because both the baseline and reporting bills are computed in full before differencing.

Appendix 2.3: Proration of an off-length (catch-up) billing period

A demand-metered account is normally read monthly. When a read is missed and the next actual read covers, say, 59 days, the period falls outside the standard 27–33 day window:

• Energy is priced on all kilowatt-hours actually consumed across the 59 days — not prorated.

• The demand charge is computed from the period’s single peak demand and then scaled by 59 ÷ 30 ≈ 1.97, approximately the two months of demand the read represents.

• The fixed/minimum charges, where modeled, scale by the same 59 ÷ 30 factor.

This mirrors common utility tariff rules for service rendered over a non-standard number of days, where the demand and fixed charges are prorated to the 30-day basis while consumption charges are not.