PMGT1865 Chap.11 Resource Levelling, Smoothing and Utilisation
Resource Levelling, Smoothing and Utilisation
A time schedule assumes resources are unlimited and can be allocated whenever needed. That assumption is harmless in long range planning and untenable in medium to short range scheduling, where the efficiency of committed resources matters most, because availability is almost always limited in quantity, in skill level or by cost.
Two failures follow from ignoring it: the resource is simply not there when the schedule calls for it, or the resource is there and idle because activities were scheduled concurrently beyond what it can handle. Resource levelling reshuffles work inside the float already available, flattening the busy days and filling the quiet ones. Two heuristic approaches exist.
Unlimited resource levelling minimises the resource input while holding the duration from the critical path method, and limited resource levelling minimises the duration while holding the resource level constant. This chapter works the seven step trial and error method by hand, measures the result with the resource utilisation factor, and is explicit about the three permanent consequences of levelling a schedule.
What this chapter covers
- 01
Why time schedules over promise, and the two failures the assumption produces
- 02
The four conditions efficient resource use requires, and what each one rules out
- 03
Levelling as a rearrangement bounded by the float already in the schedule
- 04
Unlimited against limited levelling, and which quantity each holds constant
- 05
How the two approaches map onto the words levelling and smoothing
- 06
Why heuristics displaced linear programming, and what a heuristic does not guarantee
- 07
The seven step trial and error method, and the step that does the work
- 08
The resource utilisation factor, and why it is computed before a move rather than after
- 09
Three permanent consequences: float falls, work shifts later, activities acquire fixed dates
- 10
What to do when the critical activities themselves are the problem
Levelling a small project by hand
- +1Convert the schedule to a scaled bar chart at earliest starts and allocate the crews. Clearance runs days 1 to 3, drainage 4 to 9, irrigation 4 to 6, lighting 4 and 5, paving 10 to 12 and handover on day 13.
- +1Compute the daily sums. Days 1 to 3 need 3. Days 4 and 5 need 2 plus 4 plus 3, which is 9. Day 6 needs 6. Days 7 to 9 need 2. Days 10 to 12 need 3 and day 13 needs 2.
- +1Plot and compare against availability. The peak is 9 on days 4 and 5 against a ceiling of 6, so the schedule is not deliverable as drawn and the excess is confined to two days.
- +2Measure where you are. Total labour is 9 plus 12 plus 12 plus 6 plus 9 plus 2, which is 50 person days, and the utilisation factor is 50 divided by 9 times 13, or 42.7 per cent.
- +2Choose the move. Only irrigation and lighting have float. Lighting is three workers with four days of float, so its latest start is day 8, and sliding it the full four days into days 8 and 9 places it in the deepest trough. Irrigation has the larger crew but its shorter float window sits inside drainage, so moving it would relocate the peak rather than remove it.
- +1Recompute and re-measure. Days 4 to 6 become 6, day 7 stays at 2, days 8 and 9 become 5, and the new peak is exactly 6. The same 50 person days over the same 13 days against a peak of 6 gives 64.1 per cent.
Key terms
- Resource scheduling
- Allocating work efficiently to resources that have already been committed, on the assumption that those resources are always limited in some way. It makes a schedule more representative of the actual production process.
- Resource levelling
- Reshuffling work inside available float so that busy days flatten and quiet days fill. The two heuristic families differ in whether the duration or the resource level is held constant.
- Resource smoothing
- The case in which a rearrangement stays inside the available float and the project finish date does not move. It is the outcome the week's second learning objective describes.
- Resource utilisation factor
- The planned person days divided by the resource peak multiplied by the project duration, expressed as a percentage. It answers what fraction of the crew you must carry is working on an average day.
- Heuristic
- A bundle of workable rules that walk somebody, by successive approximation, towards an answer that holds. It costs little to compute and does not guarantee an optimum, which is why it displaced linear programming here.
Resource Levelling, Smoothing and Utilisation FAQ
Why is the utilisation factor computed before the move rather than only after it?
Because the method uses it to evaluate alternatives rather than to report a result. A histogram can look better after a rearrangement and be worse: spreading a spike across two extra days at a slightly higher baseline raises the average without lowering the peak, and the peak is what determines the crew you have to carry.
The factor forces both numbers into a single figure, which is why step six sits between plotting the histogram and moving anything.
What happens when float runs out before the demand is under the ceiling?
Three responses are available and a good answer names them. Extend the duration, which is limited resource levelling and moves the finish date. Acquire more of the resource by hire or internal request, which carries a lead time. Or change the work by splitting an activity or varying its crew day by day.
In the worked example, a ceiling of five would be unreachable within float, because irrigation and drainage always overlap somewhere and together they need six.
Why does a levelled schedule absorb delay so badly?
Because the float it used to hold has been spent buying the lower peak. Non critical activities that were slid along their float have become critical, work has shifted towards the end of the project where coordination is hardest, and every activity now has a fixed start and finish it must hold or the demand goes out of balance with the committed crew.
Where the risk of time overrun is high, the prudent response is to build precautionary buffers in rather than to level to the last worker.
Can anything be done when nearly every activity is critical?
Yes, but the options change what a resource does rather than when an activity runs. A crew size can be varied across the days of a single activity, since a crew of three over six days need not be exactly three every day. Or you can split an activity and vary the distribution across the parts.
Neither is free, because both introduce the interruptions that continuous, uninterrupted work was supposed to avoid, and idle time costs money.
Assessment move
A full answer here has seven visible elements, so practise producing all of them: a daily demand table, a histogram with the availability drawn on as a line, the utilisation factor before, the move you chose with the float that allowed it, the recomputed demand, the factor after, and one sentence on whether the duration survived. An eight mark question is usually worth about one for each.
Drill against the four recurring errors: moving a critical activity, moving an activity past its own successor, reporting a new peak without recomputing the factor so there is no evidence the move helped, and calling a schedule level when one day still sits above the line. Then get into the habit of finishing with the sentence that matters most to a reader, which is how much slack is left and where.
Before levelling the worked project had seven days of float across two activities; afterwards it has three, all on one. That single comparison tells a reader exactly how much bad luck the schedule can now absorb.
Working through Resource Levelling, Smoothing and Utilisation in PMGT1865? Sia is AskSia’s AI Project Management tutor — ask any PMGT1865 Resource Levelling, Smoothing and Utilisation question and get a clear, step-by-step explanation grounded in how PMGT1865 is taught and assessed. Read this chapter free, then take your hardest questions to Sia.
