Energy claims need an actual duty profile

An efficiency claim becomes a procurement value only through a defined service duty and an applicable energy-price structure. A component's rated power is not its annual electricity consumption. A comparison of different outputs is not automatically a comparison of efficiency. An apparent saving can also arise because service has been reduced. The evaluator should first establish the useful service being purchased, then relate the energy used to that service under a stated operating envelope. Only then can the cost model give the energy difference a coherent financial meaning.

The service boundary should include relevant interactions. A component may use less electricity while requiring additional work elsewhere in the operating system. An integrated alternative may reduce local consumption while shifting demand to another utility or service. Those effects can be favourable or unfavourable; the analytical problem is whether the comparison records them. A meter at one boundary can support a narrow consumption proposition while leaving the wider service effect unresolved. State the intended boundary and identify material interactions before allowing a component result to stand for system value.

Annual consumption requires an operating-hours and load basis. A service that runs intermittently cannot be represented by multiplying its full-load rating by a calendar year without an operating convention. Standby, start-up, partial-load operation and degradation can also affect the aggregation. The required detail depends on what can change the decision. A constant annual amount may be a defensible screening input when its basis and range are clear. It becomes less adequate when load mix or operating windows distinguish the alternatives materially.

A benchmark test and a future duty forecast answer different questions. A controlled test can help establish consumption at the tested service level and condition. The forecast must also establish how much service will be required and under which conditions it will be delivered. Neither proposition automatically supports the other. An evaluator should retain the tested quantity, the extrapolation and the operating assumption as linked but distinct records. If the forecast changes, the test may remain valid while the annual consumption used in the comparison needs revision.

The energy tariff is another boundary. A uniform unit price can represent a deliberately simplified cash effect, but actual invoices may include fixed charges, demand charges or time-specific prices. Reduced annual consumption may not reduce a fixed charge. Lower use outside the demand peak may leave a demand-related payment unchanged. The workbench calculates annual electricity times a uniform supplied price with the stated energy escalation. It does not implement a tariff schedule. Where another tariff component is material, the buyer needs the relevant profile and a compatible verified extension.

Measurement should answer the disputed proposition. If the dispute concerns a single component under a defined duty, a bounded test may be informative. If the alternative changes several interacting services, a broader operating boundary can be necessary. A simulation may help organise a counterfactual where direct measurement is unavailable, but it needs calibration, assumptions and a limit on its interpretation. The buyer should explain why the chosen observation or estimation method can distinguish the alternatives, rather than select it solely because it produces a visually persuasive saving.

For an efficiency improvement, the savings counterfactual is particularly important. The evaluator needs an estimate of the service and energy that would have occurred without the proposed change under comparable conditions. Comparing an unusually demanding earlier period with a mild later period can attribute a duty change to the asset. Comparing unlike outputs can do the same. Adjustment can be appropriate when its mechanism is explained and retained. The workbench does not perform weather, production or service-level normalisation; any annual input derived through such a method needs its own evidence.

The fictional energy-price pair holds annual service assumptions and all bid fields constant while changing the common electricity-price basis. The more efficient alternative has a higher acquisition commitment and a lower supplied annual electricity quantity. Both alternatives have the same operating expense, service lives beyond the horizon, no downtime allowance and no residual credit. This controlled design asks when the financial weight assigned to the energy difference can offset the acquisition difference. It does not demonstrate the actual efficiency of an asset or supply a forecast tariff.

The low-price case gives the energy saving relatively little cash value. The high-price case gives the same annual consumption difference a larger value. These are fictional probes, with the supplied uniform price held constant inside each case. The pair contains no energy-price escalation and no change in the operating duty. It therefore isolates price rather than mix or inflation. A real evaluation should not describe the selected prices as market observations or policy forecasts; it would need current tariff evidence and an appropriate assumption about future treatment.

The lower acquisition alternative leads at the supplied low price, while the lower-consumption alternative leads at the higher price. The reversal is conditional on the common service and all unchanged input fields. It gives the committee a focused question: what energy-price basis and duty evidence should govern the intended decision? It does not grant authority to select the higher price because it rewards efficiency. A purchaser can value non-cost outcomes separately under a disclosed evaluation architecture, while retaining honest cash-cost assumptions.

Service output should remain visible beside the consumption amount. If an alternative uses less energy because it supplies less throughput, comfort or another useful service, its lower cost may be a consequence of reduced scope. The evaluator should decide whether that service level is admissible and comparable. If it is not, a consumption advantage cannot repair the deficiency through a favourable cost ranking. Where service levels intentionally differ, the decision question and valuation of the difference need to be explicit rather than silently normalised away.

Separate the acquisition of efficiency from the allocation of energy-price exposure. A buyer can purchase an efficient asset while retaining the future tariff risk. A supplied service can allocate some energy obligation to another party while adding a charge or qualification. The cost comparison should identify who pays and which commitment fixes or changes the price basis. A uniform model tariff is an assumption about economic incidence; it is not proof that an offer has hedged or transferred that exposure. Material contractual mechanisms require a corresponding analysis.

The operating team must be able to preserve the intended efficiency in use. Controls, settings, maintenance and operator behaviour can affect the service and energy outcome. A demonstrated saving under a prescribed configuration may not persist if the delivered operating arrangement changes. That is not an argument to ignore efficiency; it is a reason to include the implementation conditions in the handover. The lifecycle allowance should remain linked to the configuration and service assumptions whose performance will be monitored after the award.

Monitoring should distinguish a changed duty from a changed asset performance. If output increases, total energy can rise while energy per compatible unit of service improves. If operating hours fall, total consumption can fall without any equipment improvement. A useful record retains the service driver and consumption boundary so that the comparison can be updated coherently. The workbench's annual amount remains a summary input. It should not become the sole performance indicator if its aggregate cannot distinguish the mechanism the buyer intended to purchase.

An energy dispute can be routed through a duty reconstruction. In a hypothetical acceptance test, consumption falls while delivered output also falls. The result may reflect lower service rather than efficiency. Reconcile material, productive hours, load states and auxiliary demand before treating the difference as a saving. If comparable delivered service uses less purchased energy, the physical claim advances. If the same energy is purchased under a cheaper tariff, the financial result advances through price incidence instead. If standby consumption migrates to an omitted auxiliary system, the boundary remains incomplete. These dispositions identify the test or invoice evidence needed for each claim without conflating consumption performance with the applicable charge.

A tariff term must be tested for marginal relevance. In a hypothetical arrangement, reduced consumption lowers a variable bill component but does not change an unavoidable capacity commitment. Crediting the whole average bill rate to every unit saved would overstate the avoidable cash effect. Conversely, a duty change that reduces the committed peak can affect an obligation not represented by the simple energy-price field. Keep that mechanism in a separate charge account or use a method that implements it. The existing price pair and located boundary explain the linear variable-price case. They should not be stretched into a claim about every tariff component merely because the invoice reports an average unit cost.

The controlled tariff pair has a linear equality that can be located rather than left between two distant prices. Conventional and Efficient have the same annual operating allowance, supplied life, horizon, discount rate and terminal treatment. Neither requires an inside-horizon replacement. Efficient adds a base-date acquisition commitment while using less annual electricity. The common electricity price is the only varied scalar, and electricity escalation is zero. Those restrictions are the reason a single tariff equality is legitimate here. A changed duty, escalation or replacement calendar would create a different coefficient and require a new computation.

Let A denote the common annuity factor, ΔC the additional base-date acquisition of Efficient, and ΔE the annual electricity saved against Conventional. With common tariff p, the incremental lifecycle present cost of Efficient is ΔC minus p × ΔE × A. Equality therefore requires p* = ΔC / (ΔE × A). This is algebra for the supplied deterministic cash structure, not an estimate of the tariff a purchaser will face. Its usefulness is that a tariff evidence interval can now be compared with a specific boundary. A statement that electricity might be expensive no longer has to stand in for a decision test.

The analytic equality describes this fictional linear cost structure under real arithmetic. Its decimal boundary is converted to a finite Number input for the workbench, so the computed totals need not be exactly equal. The engine orders unequal computed costs by their actual difference and emits the tie warning only for exactly equal computed totals; it adds no near-tie tolerance. Equal computed totals receive literal code unit name order, which creates no substantive preference. The lower and higher tariff confirmations have opposite leaders, with every unvaried field preserved. A committee should inspect the complete raw difference and input precision before interpreting two printed totals that round to the same cent.

Consider three fictional evidence packets, each concerning the same common duty and price incidence. A first packet supports a tariff range from 0.030 to 0.040 USDk per MWh. Every point in that interval is below the equality, so the Conventional cost preference survives the stated tariff uncertainty. A second supports 0.060 to 0.080 USDk per MWh; the Efficient preference survives that interval. A third supports 0.040 to 0.060 USDk per MWh and crosses the equality. The third packet cannot support a unique cost preference from tariff evidence alone. These are authored interval assumptions, not observed contracts or market rates. Monotonicity of the restricted linear difference justifies the interval reading; it does not justify generalising it to another scope or duty.

The fictional Operations director owns the duty evidence, while the Commercial manager owns tariff applicability, charge perimeter and the supported interval. The committee can authorise desk clarification of a crossing interval without selecting a price favourable to a desired offer. If fixed charges, demand charges, taxes or a supplier-specific tariff make the common-price convention false, the owner stops this threshold reading and builds the appropriate common cost streams. If the required output cannot be delivered, the service gate stops the comparison regardless of the tariff interval. Locating a financial equality makes the next enquiry precise while preserving the separate requirement that the compared service be executable.