Restaurant Kitchen MEP Requirements — the invisible half of the build

Every restaurant build is really two builds. There is the visible one — finishes, joinery, the room guests will photograph — and there is the invisible one above the ceiling and under the slab: power, air, gas, water, drainage, suppression. Restaurant kitchen MEP is that invisible half, and it is the half that decides whether the visible one ever cooks at full capacity. Dining rooms are forgiving; services are not — an undersized supply is a kitchen that cannot serve its own menu, discovered under full load, after opening.

One framing note before the map, because it governs everything below. MEP design is licensed engineering: statutory design, engineering drawings and authority approvals belong to licensed professionals, through the official channels, and nothing here is legal or licensing advice or a substitute for those appointments. The owner’s job — and GGB’s, standing owner-side — is different and earlier: specification, coordination and programme control. Knowing what the services must achieve, writing the brief the licensed engineers design from, and making sure no discipline designs against another. This article is that owner’s map.

Designed from the cooking line, never retrofitted

The MEP glossary entry states the doctrine in one line: kitchen services are designed around the cooking line and the food-safety flow, never retrofitted around finished walls. It is worth spelling out why, because the whole sequence of a good build falls out of it.

The logic runs one way. The menu decides the equipment: what you cook determines what you cook on. Every appliance on that schedule carries a services demand — electrical load or gas, water in, drainage out, heat and grease-laden air up. The commercial kitchen layout then fixes where those demands land in space, shaped by the food-safety flow that keeps raw, cooked, wash and waste paths separated — the same zoning discipline HACCP inspections will later audit as records. Only then can the services be sized and routed: the ducts, cables, pipes and traps are the shadow cast by the equipment schedule onto the building.

Run it backwards — choose the unit, close the walls, then ask what the kitchen can cook — and the menu becomes a function of the ducting. What follows is the owner’s tour of the services, one by one.

Electrical load — the cooking line drives it

A restaurant is a light industrial load wearing a hospitality costume. Combi ovens, fryers, griddles, induction ranges, refrigeration compressors, extract fans, dishwashers, hot-water generation — the cooking line and its support plant dwarf the dining room’s lighting and music, and they draw hardest at exactly the hours the room is fullest. So the first electrical question on any project is never the light fittings; it is whether the shell can feed the line, with sensible headroom for the menu the venue will have after its first revision, not just its opening one.

That check happens against the base building, and it happens early or expensively:

Illustrative — the worked example from the MEP glossary entry, reused deliberately: the cooking line is specified at 90 kW of electrical load but the shell offers 60 kW, a 30 kW shortfall that surfaces as an upgrade application and weeks of delay. On an AED 800,000 fit-out where MEP is AED 200,000, that single early check governs 25% of the build budget.

The shortfall itself is common and survivable. What decides whether it is a line item or a crisis is when it is discovered — at due diligence, when it prices the deal, or after design freeze, when it prices the delay.

Ventilation and extract — sized from the menu

Extract is the service owners most often assume is generic, and it is the least generic of all. Hoods are classed conceptually by what they must capture: grease hoods over cooking that throws grease-laden vapour — fryers, grills, ranges — and lighter condensate or heat hoods over equipment that emits steam and heat alone. At the heavy end sit char-grills and solid-fuel cooking, which demand the most capture and often their own dedicated treatment before discharge. The extract volume, in other words, is sized from what the menu burns — not from the floor area of the room it happens in.

Extract also never travels alone. Every unit of air pulled out of the kitchen must be replaced with tempered makeup air, deliberately introduced, or the kitchen replaces it for you: the space goes into negative pressure, doors fight the fans, the dining room’s cooling is dragged into the hoods, and kitchen smells migrate to exactly the places you spent the design budget keeping them from. And the whole system needs somewhere lawful to go — an extract route and discharge point acceptable to the authorities and the landlord, which is a base-build reality no fan specification can conjure after the fact.

The failure pattern is always the same: extract designed to the menu, then quietly under it — a menu that adds the char-grill after the duct is sized, a hood value-engineered a size down. The fix is sequence, not heroics: the equipment schedule freezes, then the extract is engineered to it.

Gas — an energy decision before an engineering one

Gas illustrates the owner’s role at its clearest, because before any engineering happens there is a commercial decision: what fuel does this kitchen run on? Buildings differ — piped natural gas in some, bottled or bulk LPG arrangements in others, and a growing class of sites where all-electric is the practical answer. That choice reshapes the equipment schedule, the electrical load, the ventilation picture and the approvals path, so it has to be made before the schedule freezes, not discovered after it.

Where gas is used, it arrives wrapped in a safety system — detection, automatic shutoff, interlocks with the extract and suppression systems, ventilation of the spaces gas passes through — all of it designed, installed and approved through licensed channels and inspected accordingly. The owner’s contribution is not to design any of that; it is to make the fuel decision early, in writing, and to make sure the licensed engineers are briefed on the real cooking line rather than an assumed one.

Water and drainage — grease is a compliance reality

Water in is rarely the drama; water out is. A production kitchen needs hot water sized to its wash-up reality and floor drainage where the work actually happens — and above all it needs grease dealt with as the compliance matter it is. Grease interceptors are not an accessory: they are sized to the operation, they must be serviced on a schedule, and the servicing must be recorded, because municipal enforcement reads the records the same way food-safety inspectors read HACCP logs. A venue that cannot evidence its grease management is a venue with a finding waiting to be written.

Drainage also carries the least forgiving physics on the project: falls. Water runs downhill or not at all, and a drainage route that does not work on the drawings will not be argued into working on site. This is why drainage provision — like the extract route — belongs on the pre-lease checklist rather than the post-signature surprise list.

Fire suppression — the interlocked system

Serious cooking lines carry dedicated suppression over the hoods, engineered so that a discharge does not act alone: fuel and power to the line shut down with it, and the system’s relationship to the extract is part of the design. All of it is specified, installed and certified by licensed specialists through the civil-defence approval channels — emphatically not owner-side work.

What is owner-side is the coordination truth: suppression is designed against the final equipment layout. Move the fryer after the design is approved and the suppression design moves with it — another approval, another visit, another reason the layout freeze is a real gate rather than a formality.

Cold-chain power resilience — the quiet one

The service nobody photographs: what happens to your chillers and freezers when power fails, and how you would know. The cold chain holds your stock value and underwrites your food-safety records — temperatures held is precisely what HACCP logs exist to evidence — so the owner’s brief should ask the resilience questions explicitly: which circuits recover first when supply returns, is critical refrigeration separated from the cooking line’s supply, and does a night-time failure announce itself or wait to be found at opening? These have ordinary design answers — separation, monitoring, alarms — but only if they are asked at brief stage, while the answers are cheap.

The sequence — and what running it backwards costs

Put the map together and the coordination sequence writes itself:

1. Menu — what the venue cooks, decided first, because everything downstream is sized from it.
2. Equipment schedule — the menu converted into machines, each item carrying its services data: load, water, drain, gas, extract.
3. MEP brief — the schedule’s demands compiled, zone by zone, with the food-safety flow, into the requirements document the engineers design from.
4. Licensed engineers — statutory design and engineering drawings by the licensed professionals, coordinated as one package so no discipline designs against another.
5. Design freeze — the layout and services locked together before tender, so every bid prices the same building and every later change is a governed variation rather than a quiet redesign.

What governs the length of that sequence is not ambition but decision cadence and approval chains — how fast the menu and fuel decisions are made, how the authority and landlord cycles run, and the lead times on long-lead plant. What blows it up is running it backwards: walls closed before the schedule froze, services sized from a superseded revision, equipment arriving with demands the shell was never asked to meet. Backwards looks like coring finished walls, rerouting services around each other, re-approvals, variations priced under time pressure — an opening moved by paperwork rather than construction. Rework after tiling costs multiples; rework after opening costs customers.

The landlord and the shell — check capacity before you sign

The illustrative 90 kW cooking line meeting its 60 kW shell is not really a construction story — it is a leasing story, and it is avoidable at the due-diligence stage. Before signature, the services questions are blunt: What electrical load is genuinely available to this unit, in writing? Is there a lawful extract route to an acceptable discharge point? Is gas available, and on what arrangement — or does all-electric actually work here? What drainage and grease provision exists? A shell that fails a check can still be the right unit — leased knowingly, with the upgrade cost and obligation negotiated into the deal while you still have leverage. What kills projects is signing first and discovering second.

That services audit is one half of site and lease due diligence; the commercial half — the rent ceiling, the handover state, the clauses — is covered in the lease and fit-out traps. Together they are the two reads that belong before any signature, because the lease fixes both what you pay and what the building can physically feed.

The GGB read

We treat kitchen services as what they are: the production infrastructure of the business, wearing the least glamorous clothes on the project. Owner-side, that means the brief is written from the menu and the flow — the discipline that lives in HACCP kitchen design — the licensed engineers are coordinated as one package rather than four, and the freeze holds before tender inside fit-out management, so the services are sized, approved and priced before anyone argues about chair fabric. Choosing the people who do this well is its own discipline — the team-selection piece covers it. The kitchen your guests never see decides whether the one they do see works. Build the invisible half first.