04 Nov 5 Questions HR Should Ask Before Approving a Roster
Rosters are usually built to solve three problems at once: cover the hours, control the cost, and keep people reasonably happy. Fatigue rarely enters the conversation, because it does not show up in the rostering software as a number anyone is accountable for.
Under Australian work health and safety law, that is a gap with a duty behind it, and the duty points at the people who design the work rather than at the people doing it. For HR, that makes roster approval a safety decision wearing an operations hat.
Here are five questions worth asking before a roster goes out, each drawn from what the regulations actually direct employers to consider. Readers outside Australia will find the questions transfer even though the legal framing will not.
Where the duty comes from
Australia’s model WHS Regulations define a psychosocial hazard as one that arises from or relates to the design or management of work, a work environment, plant at a workplace, or workplace interactions or behaviours, and that may cause psychological harm, whether or not it may also cause physical harm.
Read that definition slowly and notice that three of its four limbs are things an employer decides. The design of work. The management of work. The work environment. None of those are worker behaviours.
Regulation 55C then requires a business to manage psychosocial risks, and regulation 55D sets out what must be taken into account when deciding the controls. That list is where the five questions come from, and it is more specific than most people expect.
1. How long, how often, and how bad?
The first matter named in regulation 55D is the duration, frequency and severity of the exposure.
Applied to a roster, that is three separate questions rather than one. A single long shift is duration. A pattern of six in a row is frequency. A night shift that ends with a two-hour commute is severity. A roster can be defensible on all three individually and indefensible in combination, which is why “nobody is over their hours” is not an answer to this question.
The practical version: look at the worst individual line on the roster, not the average across the team. Averages hide the person who always gets Sunday night.
2. What else is going on at the same time?
The regulation asks how the hazards may interact or combine, and this is the question almost nothing in a rostering system can answer.
Fatigue on its own is manageable. Fatigue in a customer-facing role where people are occasionally abusive is a different exposure. Fatigue during lone work is another. Fatigue in the middle of a restructure, when everyone is also anxious about their job, is another again. The combinations are where harm actually occurs, and they are invisible if each factor is assessed in its own meeting.
| Regulation 55D asks you to consider | The roster question it becomes |
| Duration, frequency and severity of exposure | What does the worst single line on this roster look like across a month? |
| How hazards may interact or combine | Who is tired and also dealing with difficult customers, or working alone? |
| The design of work, including job demands and tasks | Is the workload achievable in the hours, or does finishing require unpaid extra time? |
| The systems of work, including how work is managed, organised and supported | Can somebody decline an extra shift without it costing them future shifts? |
| The design, layout and environmental conditions of the workplace | Is there anywhere to take a genuine break, and does anyone use it? |
3. Is the work achievable in the hours?
Regulation 55D names the design of work, including job demands and tasks.
This is the question that separates a fatigue problem from a rostering problem. If the shift is eight hours and the work takes nine, the roster is not the cause and adjusting it will not help. The demand is the cause, and the symptom will appear as unpaid overtime, skipped breaks, or work quietly carried into the next shift.
HR usually sees this as a timekeeping anomaly before anyone sees it as a safety issue. Those payroll exceptions are often the earliest available fatigue signal in the business, and they are already sitting in a system you own.
4. Can someone say no?
The regulation names the systems of work, including how work is managed, organised and supported.
In practice the single most useful thing to test is whether a worker can decline an additional shift without consequence. Not formally. Actually. If declining tends to mean fewer offers next fortnight, then the business has built a system where tired people volunteer, and no fatigue policy survives that.
This matters disproportionately for casual and agency workers, whose income depends on being available, and who are therefore the least likely group to tell you they are too tired. Any fatigue control that relies on self-reporting will under-detect exactly where the risk is highest.
5. Is there anyway to actually stop?
The last of the matters above covers the design, layout and environmental conditions of the workplace, including facilities for the welfare of workers.
A thirty-minute break taken at a workstation, in view of the queue, is not recovery. Whether a break restores anything depends on the room it is taken in, and plenty of sites have a break entitlement and nowhere to spend it. That is a facility question with a safety consequence, and it tends to be owned by nobody.
What HR can usefully own here
Three things, none of which require a new system.
- Review the worst roster line each cycle, not the team average. One line, five minutes.
- Treat payroll exceptions as a safety input. Unpaid overtime and missed breaks are already measured. Send them somewhere they will be read as a fatigue signal.
- Ask the decline question once a year, honestly, of casual and agency staff. The answer tells you whether anything else you do about fatigue will work.
On the training side, be careful what you buy. In Australia, Basic and Advanced Fatigue Management are heavy-vehicle accreditation modules under the national heavy vehicle regulator, with their own units of competency and approved providers, and they are not what a general workforce needs. What most workplaces need is awareness: recognising fatigue, knowing what to do about it, and understanding that reporting it is expected rather than tolerated. Fatigue management training covers that awareness layer, and because fatigue so often arrives combined with workload pressure, stress management training covers the adjacent one.
Neither is a substitute for answering the five questions. The regulations direct the employer to consider the design and management of work, and no module changes a roster.
General information for employers and not legal advice. Regulations 55A to 55D as cited belong to Australia’s model WHS Regulations. Most states and territories took that scheme on with its numbering, Victoria runs a separate one, and readers outside Australia should check their own framework before relying on any of this.
No Comments