Heat Rotation Schedules for Outdoor Workers

A heat rotation schedule is a within-shift rotation plan whose interval is set by measured heat stress rather than by habit. As conditions get hotter, people leave exposed positions sooner, spend more of each hour under shade, and every relief handoff downstream is re-timed to match.

Most rotation policies were not written for that. They were written so nobody works a till for eight straight hours, and the interval landed at 45 or 60 minutes because it fits a break pattern. That is a quality of working life argument, and a good one. Heat is a different argument. It has a measurement, a dose response, and in a growing number of jurisdictions a statute behind it, and it will not accept a number chosen because it divides neatly into a shift.

What follows is the measurement, what the rules do and do not tell you, and the arithmetic that catches most teams out: shortening the interval does not reduce how long anybody stands in the sun.

Why WBGT and not the thermometer

Wet bulb globe temperature combines the four things that decide whether a working body can shed heat: air temperature, humidity, radiant load from sun and hot surfaces, and air movement. Outdoors in sunlight the natural wet bulb reading carries most of the weight in the index, with the black globe and dry bulb readings making up the rest. That weighting is the whole point. Muscles generate heat continuously and evaporation is the main way to lose it, so on a humid day the cooling mechanism is throttled whatever the thermometer says.

The practical consequence for a scheduler is that air temperature forecasts mislead you on exactly the days that matter. A dry 30 degree afternoon with a breeze across an open concourse and a still, humid 30 degree afternoon on the same concourse are two different jobs, and only one of them needs the interval halved.

The second consequence is that the reading is local. Unshaded asphalt at midday and a planted plaza two hundred metres away can differ enough to justify different intervals, so a single reading at the gate does not govern a large site. Measurement points should map onto position clusters, not onto the org chart.

What the rules require, and where they stop

The regulatory direction of travel is clear, the pace is not, and the detail differs enough that a policy written for one country is unsafe in another. Four reference points show the spread.

• Japan: the amended Ordinance on Industrial Safety and Health took effect on 1 June 2025. It applies to work expected to run continuously for an hour or more, or for more than four hours a day, in an environment at a WBGT of 28 degrees Celsius or higher, or an air temperature of 31 degrees Celsius or higher. It requires a reporting system so workers with symptoms, and workers who spot symptoms in others, can raise them, plus written procedures for removing someone from work, cooling them, and getting them to medical care. Japan's labour ministry notes that around 70 percent of workplace heatstroke fatalities occur in outdoor work.
• Singapore: the Ministry of Manpower publishes a framework for outdoor work that sets out hourly WBGT monitoring during working hours, graduated acclimatisation over at least seven days for workers new to Singapore or returning from more than a week away, and a minimum of 10 minutes of shaded rest per hour for heavy physical work once WBGT reaches 32 degrees Celsius, rising to 15 minutes at 33 degrees and above. Hydration is specified too: at least hourly, with a recommended intake of 300ml an hour or more depending on work intensity. Worth noting how this one binds. The framework is not a numbered rule in a statute the way Japan's is. It is enforced through the general duty every employer owes under the Workplace Safety and Health Act, which means an inspector reaches it by way of that duty rather than by citing a figure.
• California: the outdoor heat illness prevention standard in Title 8 section 3395 requires water, shade, and cool-down rests from 80 degrees Fahrenheit, high-heat procedures from 95 degrees, and close observation of new employees during their first 14 days. A separate indoor standard, section 3396, has applied since 23 July 2024 and drops its measurement trigger from 87 degrees to 82 for workers whose clothing restricts heat removal.
• United States federal and the United Kingdom: OSHA published a proposed heat injury and illness prevention rule in August 2024, the informal hearing closed in July 2025, and post-hearing comments closed in October 2025, but no final rule has been issued. The United Kingdom sets no maximum working temperature at all. HSE publishes a minimum, normally 16 degrees Celsius or 13 where the work involves rigorous physical effort, and expects the risk assessment to carry the rest.

Even the strictest of these writes duties, thresholds, and rest minimums. Almost none of them writes your rotation for you. Singapore's shaded rest minimum comes closest to a scheduling instruction, and it still says nothing about which position anyone occupies for the other 50 minutes. Jurisdiction by jurisdiction detail, including which rules are statute and which are official guidance, sits in the scheduling laws reference, and none of this is legal advice.

Shorter intervals do not reduce exposure

This is worth working through with real numbers, because it changes what you spend money on. Take a cluster of six positions covered by six people across a six hour operating block. Four positions are in full sun. Two are sheltered, say a shaded till and a stockroom.

1. At a 60 minute interval, each person occupies six positions across the block, four of them exposed. Total exposure per person: four hours.
2. Drop the interval to 20 minutes. Each person now occupies 18 positions across the block, 12 of them exposed. Total exposure per person: still four hours.
3. What changed is the longest unbroken stretch in the sun, which falls from 60 minutes to 20. That is worth doing, because strain builds within a bout and recovers between bouts. The daily dose is untouched.
4. To move the daily dose you have to change the ratio. Add a sheltered position and a seventh person, and each person is exposed for four sevenths of the block, about three hours and 26 minutes. Roughly 34 minutes less per person, bought with one extra head.
5. Mandated shaded rest does the same arithmetic from the other direction. Ten minutes of rest per hour is one sixth of every person's hour that cannot be spent at an exposed position, and the exposed positions still need filling.

So the honest version of a heat rotation policy has two levers, not one. The interval controls continuous exposure and is nearly free. The ratio of sheltered to exposed positions controls total exposure and costs headcount, which puts it in the same conversation as every other coverage headcount decision you make before the season starts.

Interleave the sheltered positions, do not batch them

Order matters as much as ratio. If the two sheltered positions sit next to each other in the rotation ring, someone can draw all four exposed slots back to back and take the whole four hours in one unbroken run, whatever the interval says. Alternating exposed and sheltered positions around the ring, as far as the ratio allows, caps that run at two slots. Same people, same positions, same interval, materially different exposure profile.

The walking is exposure too

Every rotation costs a walk, and on a hot site most of that walk is unshaded. At a 60 minute interval each person makes five moves across a six hour block. At 20 minutes they make 17. At three minutes a leg that is 15 minutes on the path against 51. Tripling the rotation rate triples the walking, triples the number of handoffs that can go wrong, and hands the difference to the person at the end of the chain. This is why walk time belongs in the plan as scheduled minutes rather than as slack, and why the fastest interval is rarely the safest one.

Acclimatisation is a roster constraint, not a training slide

A body adapts to heat over days of repeated exposure, and loses the adaptation nearly as fast. Both halves of that sentence have scheduling consequences, and the second half is the one that gets missed.

Singapore's framework ramps new and returning workers over at least seven days. California requires close observation of new employees through their first 14. Different mechanisms, same underlying fact: the first days are the dangerous ones.

The trap is who counts as new. Someone back from two weeks of leave in August is unacclimatised again. So is someone transferred from an indoor cost centre to an outdoor one mid-season, and so is a whole crew after a fortnight of rain. None of them are new hires and none of them appear on an onboarding list.

Handled properly, a ramp is a per-person cap on exposed assignments for a defined number of days, which is a hard constraint on who may be placed where. The plan has to know, per person, when they last worked outdoors and where they are in their ramp, and rules-driven assignment has to hold that as a constraint rather than a note. It is the same class of rule as a certification with an expiry date, which is why it belongs in the same layer as the rest of intrashift scheduling.

Costume and PPE: the worker carries their own weather

Heavy character costume, full-body suits, chemical PPE, and anything impermeable all do the same thing: they block evaporation and trap hot, saturated air against the skin. The worker is no longer experiencing the site's WBGT. They are experiencing their own, and the meter twenty metres away understates it.

Standards handle this by adding a clothing adjustment to the measured value before comparing it against a threshold. California's indoor standard does something similar, more bluntly, by lowering the measurement trigger for workers whose clothing restricts heat removal. Either way the scheduling implication is the same: a costumed or fully suited position needs its own interval, and intervals of 15 to 30 minutes are common where the site standard is 45 to 60.

The other implication is time that does not look like work. Getting out of a costume, cooling, rehydrating, and getting back in can take longer than the rest itself. Unscheduled, that time comes out of the rest period, which is precisely the minutes the policy existed to protect. Treat it the way a well-built position rotation treats walking: a known duration between two assignments, not a rounding error.

Running it on the day

A heat rotation plan is not a document. It is a thing that changes at 13:20 when the reading crosses a band, and that is where most implementations fail.

The mechanics are the ordinary relief chain. A relief walks to a position, takes over, and tells the outgoing person where to go next, and the displacement cascades one position at a time. Attractions and hospitality teams call each handoff a bump. In union contexts that word also means seniority displacement during layoffs, an unrelated matter of contract rights, so a heat policy written for a represented workforce should say which sense it means. The operational sense, and how a chain re-times itself, is covered under bump scheduling and staggered relief.

When the interval changes mid-shift, everything printed before that moment is wrong. That matters more than it sounds, because a large share of frontline staff at big venues carry no phone on shift, sometimes as a matter of policy. Break-room boards and printed slips go stale instantly, so the instruction travels through the bump itself and through leads on radio. The plan has to reach a few people accurately and fast rather than everyone slowly.

Three things are worth logging as the day runs, because none of them can be reconstructed later:

• The readings, with times and measurement points, not just the daily high.
• Who occupied which exposed position, and for how long, including the walks.
• When the interval changed band, and who was told.

If a claim or an inspection arrives months later, the roster proves who was on site. Only the rotation record proves what they were exposed to.

A starting policy that survives contact with a hot day

• Measure locally and often. Readings at the position clusters that actually run hot, at least hourly during the hot part of the day.
• Define two or three bands, not a curve. People have to be able to act on it over a radio.
• Write an interval and a shaded rest per band. The interval controls the bout, the rest controls the dose.
• Interleave sheltered positions in the ring. Free, and it caps the worst run.
• Budget the walking and the doffing. Both are minutes, both come out of somewhere.
• Cap exposed assignments during acclimatisation ramps. Include returners and transfers, not only new hires.
• Review after each heat event. Which chains ran late, which positions ran hottest, which band was wrong.

The shift pattern decides who is on site and for how long. Heat rotation decides where those people stand once they are there, and it changes while they are standing there. Planning that layer explicitly, with intervals, position rules, walk time, and same-day re-solves in one place, is what intrashift scheduling is for. Whether or not the law where you operate has caught up, the arithmetic has not moved: the interval protects the bout, the ratio protects the day, and only one of them is free.