The New London Plan could change How We Design MEP Systems

The Draft London Plan, published in July 2026, introduces a much more measurable approach to building performance.
And for MEP engineers, this could be one of the most important changes to understand early in the design process.
Rather than simply demonstrating a percentage improvement against a baseline, the draft introduces absolute performance targets for energy and space-heating demand.

Looking closely at the numbers for Energy Use Intensity (EUI) baseline targets:

🏨 Hotels: Baseline target of 120kWh/m²/yr (with an aspirational goal of 80).

🏢 Offices: Baseline target of 90kWh/m²/yr (with an aspirational goal of 70).

🎓 Student Accommodation: Baseline target of 65kWh/m²/yr (with an aspirational goal of 50).

🏫 Residential / Schools / Industrial: Baseline target of 40kWh/m²/yr (with an aspirational goal of 35).

🔥 Space Heating Demand (SHD): Capped at 20kWh/m²/yr (with an aspirational target of 15).

🏢 Energy Use Intensity:
Different building typologies are given specific benchmarks, including offices, hotels, student accommodation and residential/school/industrial buildings.
🔌 Peak Load Profiles:
Projects are expected to consider peak electrical and water demand much earlier — directly affecting infrastructure capacity and utility requirements.
📊 Post-occupancy verification:
The focus is increasingly moving from what we predict a building will consume to what it actually consumes in operation.
🌱 Whole-life carbon:
Embodied carbon is also becoming a measurable part of the development process rather than something considered separately from building services.

For us, the key takeaway is simple:
MEP design can no longer be treated as something that happens after the architectural strategy is established. If energy intensity, peak demand and operational performance are becoming measurable planning outcomes, then plant selection, system efficiency, controls, zoning, electrical infrastructure and building fabric need to be considered together from the earliest stages.

This also changes the conversation around value engineering. The cheapest system to install isn't necessarily the system that delivers the lowest whole-life cost — or allows the building to achieve its required performance.

The question we should increasingly be asking is:
Are we designing buildings to meet the metrics on paper, or designing them to perform against those metrics in the real world?