Work design, ergonomics and human-system performance

Design the task, information and environment together

Work design includes what people do, which information they use, the physical setting, tools, coordination and opportunities to recover or adapt. People differ in reach, strength, sensory capabilities, experience and other relevant characteristics. Consider actual users and maintenance/support tasks when assessing the design rather than imagining a single average operator.

Physical and cognitive demands can interact. A poorly placed container may require repeated twisting while a confusing label requires repeated checking. A redesign that moves the container can reduce handling without resolving the label ambiguity. Training can help a person understand the task, but it does not remove an obstructed access path or create missing information.

Read the account and its omissions

The original time account contains 40 minutes of routine activity, 8 of travel and 12 unallocated in a 60-minute window. Adding 8 minutes of checks leaves 4 unallocated. The new total occupies 56/60≈93.3% of the window. This is an arithmetic description under the account definitions, not a safe workload threshold or evidence that the duties can be sustained.

We have not shown whether tasks overlap, arrive in peaks, require uninterrupted attention or permit recovery. Averages can hide a ten-minute interval with incompatible demands. Underload can also affect attention; simply maximising or minimising busyness is an inadequate design rule. Inspect normal, unusual and transition conditions and involve the people who perform the work.

Turn a task description into a feasible timeline

Task analysis makes the sequence and its conditions visible. Identify the goal, the information needed at each step, what must precede it, when it is released, when it is due, and which person or equipment it occupies. HSE’s task-analysis guidance distinguishes sequence, spatial relationships and timing; mapping simultaneous demands is a useful first inspection. The following exclusive-attention model is deliberately simpler than a full human-factors assessment.

In an original ten-minute window, one person records an intake for 2 minutes, starting at minute 0 and due at 2. Label matching then needs 4 uninterrupted minutes, can start only after intake, and is due at 10. A pickup verification is released at minute 3, needs 3 uninterrupted minutes, and is due at 6. Each task needs that person’s full attention; there is no second worker or automation in this model. Times include the task’s information access, with no additional changeover assumed.

The nominal plan places intake at 0–2, matching at 2–6 and pickup at 3–6. Nine minutes of work fit arithmetically in ten, but the person is double-booked during 3–6. The problem is the temporal intersection of two exclusive demands, not evidence that a person lacks commitment or that 90% utilisation is intrinsically unsuitable.

A feasible order is intake 0–2, idle 2–3, pickup 3–6, then matching 6–10. Every release, precedence and deadline is respected. The idle minute at 2–3 cannot be used for the first part of matching, with the remainder after pickup: matching cannot be paused. A persistent order/status record must preserve what is needed when matching starts. This reordered plan demonstrates feasibility for the supplied tasks; it does not establish adequate recovery, variation tolerance or acceptability over a whole shift. Matching has zero deadline slack: any extra duration would require reconsideration.

Explain what each redesign changes

Compare two retrieval proposals which both make 5 trips of 4 m per batch. Layout A places the container at a compact fixed position and leaves order IDs and completion status on separate screens. The person must retain up to three unfinished order/status associations while handling the container. Layout B provides an adjustable nearer position and a single order card showing its ID, expected quantity and completed/not-completed state. The card carries the task state through interruptions instead of requiring all of it to remain in memory. The ID and written state also distinguish orders without relying on colour alone.

A simple external-load model helps isolate one physical mechanism. Apply the force-and-perpendicular-distance relationship from the statics preparation: moment magnitude M=F×d. Here F is a downward force of 60 N, supplied directly rather than calculated from a person’s mass; d is its horizontal distance from the same chosen reference point. A uses d=0.50 m and B uses 0.20 m. This schematic is not an anatomical model. The force a muscle must produce depends on internal geometry and other loads, which are absent here.

The proposals also have costs and constraints. In this constructed room A occupies 0.6 m² and B 1.0 m² of a 1.2 m² available footprint, so both fit that particular space account. An adjustable stand needs stable adjustment, access and maintenance assessment; a persistent card needs a reliable update rule or it can display stale status. Neither removing walking nor adding a display is automatically a successful whole-task redesign.

Keep affected users visible in the comparison

The table supplies fictional observations from 20 scripted checks per layout for each of two use conditions. These are teaching data, not a trial conducted with participants. Assistance means another person was requested to complete a retrieval; a mismatch is an incorrect order/status match. Both can occur on the same check, so their counts must not be added as if they were disjoint failures. Each row retains its own denominator.

B has fewer supplied assistance and mismatch counts in both conditions at the same counted travel distance. Its nearer position provides a plausible physical mechanism; its external task-state record provides a plausible information mechanism. But both features changed together. These small constructed counts cannot isolate their effects, establish a population error probability or demonstrate freedom from pain or fatigue. A real formative comparison would record task completion, errors, assistance, relevant effort/discomfort reports and timing with the intended users and conditions, including setup and unusual cases.

If the question is which feature changes mismatches, compare card arrangements under matched layout conditions rather than attributing everything to the nearer bin. If the question is whole-task suitability, retain all relevant outcomes and access needs. A single average time or an overall sum across people would hide the barrier that the row-level record is intended to reveal.

Use evidence to redesign the system

For the proposed checks, observe when information arrives, what must be remembered, how interruptions are handled, what movements recur and what errors or near misses reveal. Explain observation and recording arrangements appropriately and protect participants’ information. Compare task conditions and outcomes without using a small formative trial to label individuals as inherently slow or careless.

Possible responses include relocating materials, clarifying status feedback, removing duplicate entry, distributing peaks, adjusting staffing or changing automation support. Each can introduce new demands: automatic batching may require supervision, and a dashboard may add alarms. Reassess the whole task and its failure modes after the change. Faster average completion is useful evidence only alongside quality, accessibility and sustainable work.

Compare travel, external load geometry and whole-task outcomes

Suppose one original task arrangement requires 12 retrieval trips per batch and a proposed arrangement requires 5 for the same defined batch. That is 7 fewer trips, or 7/12≈58.3% fewer. If each assumed trip is 4 metres, counted travel drops from 48 to 20 metres per batch.

This is a movement-account comparison. It does not measure lifting force, posture, repetition, fatigue, accessibility or mental workload. The new arrangement could reduce walking yet place a heavy container at an awkward reach. Observe the changed task before calling it an overall ergonomic improvement.

Both A and B retain the 5-trip, 20 m proposal, so walking alone cannot choose between them. In the supplied geometry, A has moment 60×0.50=30 N·m and B has 60×0.20=12 N·m. The reduction is 18/30=60% for this external turning effect, while force remains 60 N. This is positive evidence about one modelled demand, not a 60% reduction in injury risk or in internal muscle force.

Under the supplied footprint limit both fit, and the illustrative task records favour B on the listed access and mismatch outcomes. B is therefore the stronger candidate for further whole-task evaluation under these assumptions. Preserve the zero-slack timeline issue separately: even a better bin and card cannot make two full-attention tasks simultaneous, and a second worker’s help consumes that worker’s capacity and introduces a handoff.

Try the reasoning

1. Does 93.3% scheduled time establish an acceptable workload?

   Show the answer and reasoning

   No. It is a total-time ratio. Timing, variability, physical/cognitive demands, recovery and other task conditions need assessment.

2. Why involve workers in task analysis?

   Show the answer and reasoning

   They can identify real sequences, adaptations, information gaps and unusual conditions that a nominal process description misses.

3. Can less walking coexist with a worse task?

   Show the answer and reasoning

   Yes. A layout might reduce distance while worsening force, posture, accessibility or attention demands.

4. Keep the ten-minute task model, but make label matching due at minute 8. Can the reordered single-person plan still meet every deadline?

   Show the answer and reasoning

   No. Pickup must occupy 3–6 to meet its release at 3 and three-minute duration by 6. After intake, only 2–3 is free before pickup and only 6–8 after it: neither uninterrupted interval fits the four-minute match. A compatible extra resource, a changed release/deadline or a substantiated task redesign is needed. Treating the match as pausable changes the model and requires evidence; the same nine-minute total does not resolve the conflict.

5. If the available footprint is reduced to 0.8 m², should B be selected solely because its external moment and illustrative outcomes are better?

   Show the answer and reasoning

   No. B requires 1.0 m² and violates the new space constraint, while A’s 0.6 m² fits that account. Reconfigure space or redesign B and recheck the task, rather than declaring an infeasible option acceptable. A fitting the room does not resolve its attention, access and error concerns; feasibility and whole-task suitability are separate tests.

6. For another supplied load, F=45 N and reach decreases from 0.35 to 0.25 m. What changes in the external moment, and what additional task evidence is needed?

   Show the answer and reasoning

   The moments are 45×0.35=15.75 and 45×0.25=11.25 N·m. The fall is 4.50 N·m, or 4.50/15.75≈28.6%, with force unchanged. Duration, repetition, posture, grip, other loads, user access and cognitive demands still matter. This calculation alone establishes neither a recommended reach nor acceptable workload.

What to carry forward

Compare task sequences and demands with explicit users, constraints and outcomes: a redesign must be feasible in time as well as useful physically and cognitively.