SAP or SBEM: choosing the calculation for a mixed-use scheme
A mixed-use block needs both calculations. Here is where the dwelling boundary falls, how to weight a combined figure and how to handle plant serving both parts.
On a single mixed-use building the SAP or SBEM question is answered by splitting the building, not by choosing one method for the whole of it. SAP is the national calculation methodology for dwellings and produces a dwelling emission rate for each unit. SBEM, or an approved dynamic simulation model, is the methodology for buildings other than dwellings and produces a building emission rate. A block with flats above a commercial ground floor needs both, and the energy statement has to report a combined figure that does not double count the shared elements or the shared plant.
The awkward part is not the modelling. It is the reporting, because local plan policies asking for a percentage carbon reduction were generally written with one methodology in mind.
What each methodology models
SAP assesses a dwelling as a whole: fabric, ventilation, heating, hot water, lighting and any generation, against a notional dwelling of the same size and shape. The outputs that matter for planning are the dwelling emission rate, the target emission rate, and the fabric energy efficiency figures. Each flat is modelled individually, then the results are aggregated for the residential part of the building.
SBEM assesses non-domestic space by activity. Each zone is assigned an activity type with its own occupancy, lighting, small power and ventilation assumptions, and the model compares the actual building against a notional building with the same geometry. Where the building has complex servicing, unusual occupancy or a need to demonstrate overheating performance for planning purposes, a dynamic simulation model is used instead of SBEM.
| Aspect | SAP | SBEM or dynamic model |
|---|---|---|
| Applies to | Dwellings | Buildings other than dwellings |
| Unit of assessment | Each dwelling | Zones by activity type |
| Headline output | Dwelling emission rate | Building emission rate |
| Comparison basis | Notional dwelling | Notional building |
| Approved Document | Part L volume 1 | Part L volume 2 |
Where the boundary between the two falls
The boundary follows the building regulations definition of a dwelling, not the tenure and not the planning use class. A self-contained flat is a dwelling. A room in a house in multiple occupation may or may not be, depending on the arrangement. Student accommodation with shared facilities is usually assessed as non-domestic, and purpose-built student accommodation with self-contained studios can fall either side depending on the design.
Get this decision made and recorded early, because it determines which model produces the numbers everything else depends on. Where the answer is unclear, put the question to the building control body and to the case officer in the same email, and record the answer in the Energy Statement.
Three elements sit on the boundary and cause most of the confusion: the common parts of the residential element, the plant space serving both parts, and the thermal envelope where a dwelling sits above a commercial space. Residential common parts are generally assessed in the non-domestic model, which surprises people the first time.
Reporting a combined figure without double counting
The scheme-level figure most policies ask for is a percentage improvement over a baseline, and to produce one from two models you have to weight the parts.
The method that stands up is weighting by regulated emissions rather than by floor area. Sum the notional or baseline regulated emissions for the residential part and for the non-domestic part, sum the actual regulated emissions for each, and calculate the percentage improvement on the totals. Weighting by floor area gives a different answer, and on a scheme where a small commercial floor plate has high regulated emissions the difference can be material.
Two double counting traps. Photovoltaic generation on a shared roof cannot be claimed in full in both models, so allocate it, state the allocation and justify it. And a heat pump serving both parts appears in both models, so make sure the efficiency and the load split are consistent between them.
State the method in one short paragraph. An officer who can see how the blended figure was produced will accept it. One who cannot will ask, and the answer will arrive three weeks later.
Communal plant serving both parts
Shared plant is where the two models most often disagree with each other, and where a reviewer looks first.
A communal heat pump or a heat network connection serving flats and a commercial unit has one set of performance characteristics, one distribution system and one metering arrangement. Both models need the same seasonal efficiency, the same distribution losses and the same carbon factors. Where the residential model assumes a higher system efficiency than the non-domestic model for the same machine, the figures are wrong somewhere.
Distribution losses deserve attention. A communal system with long horizontal runs and poorly insulated risers can lose enough heat to move the residential result substantially, and the loss is modelled differently in the two methodologies. Where the scheme relies on communal plant to meet a policy target, model the losses explicitly rather than accepting a default. The wider design implications of a shared low temperature system are set out in energy masterplanning across a phased site.
How officers read a blended percentage
Case officers are not energy assessors, and what they look for is a clear number against the policy requirement, with a method they can follow and a consistency check against the rest of the submission.
What causes questions: a blended figure with no stated weighting, a figure that changed between the design and access statement and the energy statement, generation claimed at a capacity the roof plan cannot accommodate, and a percentage improvement quoted against an unstated baseline.
What helps: a single table showing residential and non-domestic parts separately and combined, with baseline and proposed regulated emissions in kilograms of carbon dioxide per year and the percentage improvement for each. One table, three rows, and the officer has what they need.
Where the local policy requires on-site renewable capacity as well as a carbon reduction, report both, and show the roof plan area available. Where the Sustainability Statement covers the wider policy compliance, keep the numbers identical between the two documents.
Presenting two calculations in one statement
- State which parts of the building are dwellings and which are not, with the basis for that split.
- State the methodology used for each part, including the software and its version.
- Present the residential results as a range and an aggregate across the dwellings, not as a single sample flat.
- Present the non-domestic results by zone activity where the mix is varied.
- Present the combined table with the weighting method stated.
- Set out the renewable generation, its capacity, its location and its allocation between the two models.
- State the energy hierarchy steps, showing demand reduction achieved before supply and generation are counted.
Point three matters. A statement reporting one representative flat invites the question of which flat, and the top floor corner unit and the middle floor internal unit are not comparable. Reporting the range is more work and far more credible.
Consistency with the eventual Part L submission
The planning energy statement is not the building regulations submission, and the two are produced at different design stages with different levels of information. That is normal. What is not acceptable is a Part L submission that contradicts the planning commitment.
The practical control is to keep the same models and update them, rather than rebuilding at technical design stage with a different assessor and different assumptions. Where a change is unavoidable, and specifications do change during procurement, identify whether the planning condition tolerates it. A condition requiring a stated percentage carbon reduction is a real obligation, and falling short of it needs a variation rather than a note in the file.
One honest point. The gap between modelled and measured performance is well documented across the industry, and neither methodology predicts what a building will actually use. Say so where the statement makes claims about outcomes, and prefer to describe the compliance position rather than the expected energy bill.
This is general guidance, not legal advice. Check the position with your local planning authority, your building control body or a qualified professional before relying on it. PlanScope provides AI-assisted draft reports for planning applications. Reports should be reviewed before submission. We do not guarantee planning approval.
Frequently asked questions
Can one methodology be used for a whole mixed-use building?
No. SAP applies to dwellings under Part L volume 1 and SBEM or an approved dynamic simulation model applies to buildings other than dwellings under volume 2. A block with flats above commercial space needs both, and the energy statement has to report a combined figure that does not double count shared elements or shared plant.
Where does the boundary between the two fall?
At the building regulations definition of a dwelling, not the tenure or the planning use class. Self-contained flats are dwellings. Student accommodation with shared facilities is usually assessed as non-domestic. Residential common parts are generally assessed in the non-domestic model, which catches people out. Confirm the split with building control early.
How should a combined carbon reduction percentage be calculated?
By weighting on regulated emissions rather than on floor area. Sum the baseline regulated emissions for each part and the actual regulated emissions for each, then calculate the improvement on the totals. Floor area weighting gives a different answer, and the difference is material where a small commercial floor plate has high regulated emissions.
How is shared plant handled in two models?
With identical assumptions in both. A communal heat pump or heat network connection has one seasonal efficiency, one set of distribution losses and one carbon factor. Where the residential model assumes a better efficiency than the non-domestic model for the same machine, something is wrong. Model distribution losses explicitly rather than accepting defaults.
Does the planning energy statement bind the Part L submission?
The two are separate, produced at different design stages with different information, so differences are normal. A Part L submission that contradicts a planning commitment is not. Keep and update the same models rather than rebuilding at technical design, and where a specification change breaches a condition, apply to vary it.
Sources and further reading
This article is general guidance about the planning system in England and, where stated, the rest of the UK. It is not legal or professional advice. Check your local planning authority’s validation requirements and take professional advice for your specific site.