Facility teams that manage a restaurant group, a central kitchen, a hotel back-of-house or a food plant usually already have a written cleaning routine. What is often missing is a clear answer to one question: when is a kitchen exhaust duct clean enough to call clean? Without acceptance criteria, different crews judge the same duct differently, contractors deliver inconsistent results, and the evidence trail a fire-insurance surveyor asks for becomes hard to assemble. The following framework turns that judgment into documented, repeatable criteria your team can inspect and sign off.
Start with the standard your site is actually measured against
Most kitchen exhaust cleaning programs in North America are inspected against NFPA 96 (Standard for Ventilation Control and Cleaning of Commercial Cooking Operations), which defines required cleaning frequency by cooking-load category, sets the residue-thickness expectation used on duct interiors, and requires that fans, plenums and housings be cleaned to bare metal where grease deposits exist. Local fire codes and your insurer may add stricter wording, so write the criteria in two layers: the minimum your jurisdiction and insurer impose, and the tighter internal standard you hold crews to. Record which layer each criterion comes from so a future audit does not turn into an argument about whose rule applies.
Define four measurable criteria, not one
"Clean" is not a measurable target. Split it into four items a supervisor can verify with a flashlight, a straight edge, a photograph and a checklist.
- Coverage. Which components were reached — hood filters, the plenum, the horizontal duct run, vertical riser, fans, and the exterior terminal cap. Any segment that could not be accessed must be listed as an exception, not quietly skipped.
- Deposit thickness. Set a numeric pass line for the duct interior consistent with the standard you follow, and inspect at agreed sample points along each run rather than only at the access door nearest the hood.
- Bare-metal finish on high-load parts. Fan blades, housings and plenum interiors where grease films exist should read as bare metal, with no soft film remaining that can reignite.
- Reassembly and function. Access panels resealed, dampers free, fans running without vibration, filters seated correctly, and the work area left dry and usable for service.
Write each criterion as a pass/fail statement with the inspection method named. A criterion without a stated method is an opinion, and opinions do not survive a claim investigation.
Set the inspection evidence before the work starts
Agree in advance what has to be captured, because evidence collected after the fact is worth much less. A practical minimum is: dated before-and-after photographs at each sample point, the measurement reading at those points, the crew and equipment used, the cleaning agents or methods applied, and a signed checklist per section. Store the file by site, date and duct section. When you buy cleaning equipment, ask whether it supports that evidence workflow — a machine that leaves surfaces dry and free of chemical residue makes the follow-up inspection and the food-safety paperwork easier at the same time.
Choose a cleaning method that can meet the criteria on your ducts
Manual scraping works but is slow, drives crews into awkward positions, and leaves residue on shoulders and weld seams. High-temperature steam changes the mechanics: saturated steam softens and lifts polymerised grease from vertical and overhead surfaces, and the low water volume means the duct does not flood or need long drying time. For in-house programs, a fixed unit plumbed to a water supply is the practical starting point. The Fixed Steam Cleaning Machine PowerDe HW02 runs on 220 V with a 6 kW heater, 160 °C steam at 0.8 MPa, an 80 L SUS304 tank, and switches between dry and wet steam, which is the profile facility teams look for when scrubbing hoods, filters and equipment surfaces between deep cleans. Confirm the model's duty cycle, hose reach and accessory set with the manufacturer for duct work, and never enter a duct that is still hot or energised.
Pair the machine with the right chemistry policy. Many criteria now cap or prohibit caustic strippers in food areas because of residue and rinse-water control. If your standard allows steam-only cleaning, say so explicitly in the specification; that single line removes an entire class of disputes with contractors about what "cleaned" meant.
Set frequency from cooking load, then defend it with records
Frequency is the criterion most often argued about. Base it on the cooking-load category of each kitchen — charcoal and solid-fuel cooking, high-volume wok and frying, moderate-volume frying and grilling, low-volume baking and roasting — and inspect against those categories when the mix changes. A venue that adds a charcoal grill changes its cleaning class even if square footage is unchanged. Then require that each completed clean close out the loop: measurement recorded, photographs filed, exceptions resolved, next date scheduled. If your team handles the routine cleaning while a contractor handles the duct, define who owns which criterion and who signs the final release. For a multi-site portfolio, publish one written kitchen cleaning process review that fixes the sample points, the pass lines and the form; site-by-site variation is where acceptance criteria quietly collapse.
Turn the criteria into a two-page sign-off sheet
Keep the deliverable short enough that it actually gets used: one page of criteria with pass/fail boxes and sample-point locations, one page of exceptions and the corrective action with a due date. Have the same person who inspected sign the sheet, and file it with the photographs. Done this way, acceptance criteria stop being a document nobody reads and become the routine that protects your kitchen, your equipment investment and your insurability at the same time.
