Most kitchen cleaning plans are written for a room. They say the line must be degreased weekly, the fryers daily, the floor every night, and they are written by someone standing in the middle of the kitchen with a clipboard. The fryer line is where those plans fail first, because a fryer is not part of a room in the way a worktop is. It is a hot machine that throws oil at its own surroundings for ten hours a day, and everything about cleaning it follows from that single fact. A plan that works is written one machine at a time, starting from what that machine collects, where it collects it, and who is standing there when it stops collecting.
The method below produces a short card per machine rather than one page for the kitchen. It also produces something more useful than a schedule: a set of decisions somebody can be held to, with a name against each one.
Write the plan one machine at a time
Two fryers of the same model, bought on the same day, do not get equally dirty. The one beside the wok range wears a coat that the one against the wall never acquires, because airborne oil travels and settles on the coldest surface nearest its source. A frequency written for "the fryers" is therefore an average, and averages are the reason a plan is quietly ignored: the crew knows the third machine from the left needs more than the sheet allows, and the sheet says nothing about it.
So number the machines on your own floor plan, in the order they stand, using the names your staff already use. The unit of the plan is one machine, and each card answers four questions in plain language: what this machine collects, which of its surfaces may be touched and by whom, what is done to each surface and when, and what has to be true before anyone cooks on it again. Nothing on the card applies to another machine unless the walk-around says it does.
This sounds like more paperwork than a one-page schedule. In practice it is less, because the card is written once from an inspection and then only edited when an inspection finding repeats.
Put the boundary at the cooking surface
Every argument about heat, water and equipment damage in a kitchen comes from plans that cross a boundary without noticing it. Draw the boundary at the cooking surface and there are three zones, not one.
- Inside the cooking surface. The fryer vat, the griddle top, the oven cavity, the inside of a kettle or bratt pan. These are food-contact surfaces. Their cleaning is governed by the maker's instructions and by your own food-safety routine, including the rinse step and whatever verification your system uses. No cleaning tool changes that routine, and nothing in this article should be read as a claim that heat alone makes a surface safe for food.
- The outer skin. Front panels, control knobs and their skirts, handles, the top rim, splash shields, the stand, the legs, the drain valve and its spout, the grease trough and cup, the backsplash. This is the part that carries the oil the machine throws, and it is the part a cleaning plan is actually about. Here the state of the soil decides the method.
- The service internals. Burner box, burner ports and air openings, gas train and its connections, spark electrode and flame supervision device, thermocouple, element terminals and seals, the inside face of a door seal, the jacket and tilt mechanism of a bratt pan, the flue. These depend on clean, dry interfaces and on parts being in adjustment. They belong to the engineer or to the maker's service visit, and they are the reason a plan needs a boundary before it needs a method.
Write an owner against each zone: the cook on the station, the cleaning team, the gas or electrical engineer, the contractor, the manufacturer's service. A plan that says "steam clean the fryer" without a boundary is a plan that will one day fill a burner box with warm water and discover it three days later as a fault.
There is a fourth zone that kitchens forget to assign: the oil itself. Used oil, the filter cart and its media, the route from the drain valve to the drain and from the drain to the interceptor. It is a handling task with its own container and its own record, and it is not a cleaning task by any definition.
Walk one machine and write down what it collects
Do this at the end of a service, when the deposits are honest, and do it with the person who cooks on that station. Ask them where they stop wiping. Their stopping point is where the card begins.
- Deep-fat fryer. Oil mist settles on the front panel, around the knob skirts, along the top rim and on the face that looks at the wall, which is the only face nobody ever sees. Oil drips at the drain valve and runs down the spout, so the valve area and its drip tray carry the darkest film. Boil-over crusts at the rim and around the basket rack. Oil tracks down the outside of the vat to the stand legs and the strip of floor beneath the feet, where it collects dust and turns into a mat. The filter cart lid and stand, and the drain channel with its surround, are usually worse than anything on the machine itself.
- Griddle. The top is scraped as a matter of course, so the failures are at the edges: the joint between the top and the splash back, the corners, the trough and the cup, and the panel below the front lip where scrapings fall.
- Range. Boil-over runs down the front and into the gutters and drip pans. Film builds on knobs and on the control panel where hands touch. Splatter dries on the backsplash and on any shelf above the cooking line.
- Oven. Handle, the outside of the door glass, the panel, and the outer face of the door seal. The cavity is a separate task with its own chemistry and its own time.
- Kettle, bratt pan, bain-marie. Rim, hinge, crank or tilt handle, panel and surround. The well of a bain-marie carries scale and standing water, which is a different deposit from oil and needs a different answer.
Photograph the two worst places on that machine from a fixed point: the same tile joint, the same floor mark, the same light. Those photographs become the reference for every later decision, and they are the only part of the card that will still mean the same thing in six months.
Then note the two things that decide whether any method can be used at that position: how far the nearest water point and the nearest socket are, and whether the machine can be pulled out at all without disturbing a gas connection or a flue.
Sort those surfaces into three lists
Take the walk-around notes and split them. This is the step that makes the plan hold up when an experienced person leaves.
- List A: clean it. Panels, handles, rim, splash shields, stand legs, troughs and cups, the drain valve area, the floor strip under the feet. Method chosen by the state of the deposit, nothing else.
- List B: clean it with restrictions written down. Control panel edges and any gap around a knob; the seam between the vat and the splash back; the outer face of a door seal; nameplates, service labels, printed instructions and any decorative or brushed stainless; the area around an unsealed cable entry. These tolerate heat and water only in short, controlled passes, with masking, less moisture and immediate drying. A label that lifts is a small disaster: it is the label the engineer and the warranty both ask for.
- List C: not a cleaning task. Burner box and ports, gas train and connections, electrode and supervision device, element terminals and seals, the oven cavity and the inside of its seal, the jacket and tilt mechanism, filter media and used oil, duct above the first reachable section, the trap and everything beyond it, the interceptor. Give every line an owner and a date, not a cleaner.
Anything left off List C will still get done, badly, by whoever has a wand and twenty minutes. That is how electrical faults appear in kitchens that are cleaned more often than they used to be.
Match the method to the state of the deposit
Kitchen oil behaves according to how long it has been heated, so the useful distinction is between states, not quantities. Your team can tell them apart with a torch, a cloth and a plastic scraper, and once they can, most arguments about equipment stop.
- Fresh film. Lifts under a warm damp cloth, no colour change, no resistance. Warm water and detergent deal with it. Anyone demonstrating a machine on fresh film is showing you the easiest surface in the building.
- Polymerised coat. Dull, dark, sometimes tacky, with darker edges; it smears under a cloth instead of lifting. This is the state that consumes labour. It responds to heat held on the surface and then removed, repeated in short passes while the coat is still soft. Scrubbing it before heating it mostly moves it around.
- Carbonised crust. Hard, black, gritty boil-over that has been heated past softening. No practical amount of heat removes it. Scrape most of it into the bin first, then treat what remains as a film. A plan that never gives anyone time to scrape will conclude, wrongly, that no machine works.
Check state by raking a torch across the surface rather than relying on the ceiling lights, running the agreed white cloth along an edge and looking at the cloth, drawing a plastic scraper edge across a corner, and pressing a fingernail into a seam.
Heat and low water volume genuinely help in three specific places, and it is worth naming them so the plan does not oversell the method. There are no buckets to carry beside a hot line and no hose to drag across a wet floor. Surfaces near controls dry quickly, which is what makes the outer skin of an electrical machine worth cleaning at all. And there is no detergent residue to rinse off surfaces that stand close to food. In commercial kitchen steam cleaning, as in any other method, the state of the deposit decides the pass, the dwell and the number of repeats; the machine only decides how much heat is available and how much water comes with it.
It is equally important to write down what the method does not do. It does not remove crust. It does not reach behind a sealed panel or into a burner box. It does not clean an oil line, a filter cart or a pump. It does not reduce how often the interceptor is emptied. It does not make a food-contact surface safe, and it does not replace the rinse step or your own verification. A plan that states its limits is believed by the crew; a plan that does not is quietly reinterpreted.
Set the rhythm from what you find, in three tiers
Tie each tier to a window that actually exists in your building, and count the whole task inside it.
- Every service. The surfaces hands touch, the front panel, the knob faces, the handle, the griddle top, the grease cup emptied, the drain valve wiped as it is closed. Warm detergent and a cloth. No machine, no argument, no waiting.
- One planned block each week. The outer skin: panels, knob skirts, rim, valve area, splash shields, stand legs, troughs. Heat, dwell, short repeated passes, then drying. In most kitchens the only reliable long window is a quiet morning before delivery, not the end of a Friday service, when the crew is the tired remainder of the day.
- Planned downtime, quarterly or twice a year. Crust removal, the rear faces where the machine can be moved safely, the drain channel and its surround, everything on List B that needs masking and time, and the items on List C that belong to the engineer. This tier is where the machine's real condition is decided.
Cooling, isolation, masking, moving, warm-up, refilling, drying and the checks all belong inside the window. Drying is part of the task, not the end of it; a panel that looks clean behind a knob and is still wet an hour later has simply moved the water somewhere.
Then let the log set the frequency. If a surface is back in the same condition before its next scheduled visit, the interval is wrong or the method is too weak, and one of those two is the thing to change. If a surface has not changed since the last inspection, stop doing it on a calendar and inspect it instead. Kitchens that do this end up doing less work than before, on the surfaces that matter.
Name the people and the hand-back
Three names against every task: who cleans, who checks, who returns the machine to service. In most kitchens the third is the chef on duty, and the plan should say so out loud rather than leaving it to be assumed at eleven at night.
- Before work starts. Surfaces cool enough to work beside, gas isolated with the key in one named pocket, electrical isolation where panels and knobs are being cleaned, food covered or moved, the walkway protected, the drain channel clear before anything warm and greasy is sent down it.
- Before the machine is handed back. No water standing under the feet or in the drain channel, knobs and control areas dry, the drain valve seated with its drip tray in place, labels legible and still stuck down, nothing left damp inside a seam, floor dry where people walk with trays.
- On the first use afterwards. The person who will cook on it watches the first ignition and the first boil. Uneven flame, a delayed light or a smell that should not be there is a finding about the cleaning, and it belongs in the record, not in a conversation the next morning.
The record is one line per machine per event: date, machine, what was done, what is still there, who signed. The column that matters is "what is still there", because that column is what moves a job from one tier to the next. For central kitchen degreasing written this way, the log is also the document you show an auditor and the one you send to a supplier, since both of them need findings rather than assurances.
Read a specification the way a plan needs it
A machine's data table answers a few questions well and several others not at all. Take the published figures for the PowerDe HW02 fixed steam cleaning machine as an example: 220 V working voltage, 6 kW heating power, a steam temperature of 160 degrees Celsius, a vapour pressure of 0.8 MPa, an 80 L tank in SUS304, a tap-water inlet with a half-inch internal-thread fitting, a choice of dry or wet steam, and a footprint of 1,200 x 800 x 1,200 mm including handles.
- The temperature and the pressure are one state described twice, not two separate strengths to compare. What they tell a planner is how fast the machine softens a polymerised coat, which is not the same as how fast a person finishes a fryer. The time is spent in dwell, wiping and drying, and only a little of it is spent with a wand moving.
- The tank size and the inlet decide the working cycle and the length of the refill walk. A plumbed point changes the plan completely; a machine that must be filled by hand from a sink two rooms away sets its own rhythm, and that rhythm belongs in the card.
- The electrical figures mean a circuit checked by your own electrician at the point where the machine will stand, not at the socket nearest the catalogue photograph.
- The footprint and the door widths decide whether the machine reaches the position at all, and whether the work converges on one fixed point or the machine has to move between machines.
- Dry steam and wet steam are a moisture choice, not a power choice: less moisture near controls and electrical work, more for the heaviest coat, and whatever is added has to be dried and has to go somewhere.
- The protection functions listed on the page, covering leakage, short circuit, high temperature, high voltage and water shortage, protect the machine. They do not protect your circuit, your drainage, your labels or your operator, and none of them shorten a task.
Four figures the page does not publish decide whether any plan fits: warm-up time, hose and wand length, working time per tank, and whether the stated temperature and pressure are measured at the boiler or at the nozzle. Ask for those in writing. A hose length you have guessed is a plan built on a guess about reach, and reach is what a fixed machine is bought for.
One caution before any of this goes into a kitchen document: that product page describes vehicle cleaning at fuel stations and car parks, including interior work and engine compartments. Ask in writing whether the machine is suitable for the surfaces and the deposits on your line, and what the maker's conditions are. The commercial steam cleaning equipment range is offered in fixed and mobile forms, and which form suits a kitchen is settled by the floor plan and the rota, not by preference.
Prove it on one fryer before you write the rest
Pick the dirtiest machine that can realistically be released, and write the pass standard down before anyone starts: the photographs, the cloth, the torch angle, the seams to be checked. Use two of your own people, one of whom works that station every day, and treat their objections as data rather than resistance. Time the whole thing end to end, including cooling, isolation, masking, warm-up, refills, drying and the checks. Clean one half of the machine with the new method and the other half as it is done today, so the comparison is on the same deposit rather than on two different machines.
Then read the result honestly. If the work does not fit the window, the answer is scope or schedule, not a faster machine. If the deposit is too thick for anything to move, the answer is a mechanical first step that nobody has ever been given time to do. If the power, water or drainage is not there, the answer is in the building, and it is far cheaper to find that out on paper than after an installation. If it works, extend one machine at a time and set the frequency by what inspection finds.
Send that record to suppliers rather than a summary of how it went. Two people, forty-eight minutes, one refill, film still present behind the third knob, photographs attached, can be answered with something useful. It did not work gets you a brochure. For the written confirmations this article keeps asking for, including the four figures a data table does not carry, send your machine list, your measured window and your utility details through the contact page and ask for the answer in writing.
A plan for fryers and cooking equipment is not a schedule. It is a boundary at the cooking surface, a list of what each machine collects, three lists of what may be touched, a method chosen by the state of the deposit, and three names against every task. Write that once, per machine, and the calendar becomes a detail.
