A late week costs the contractor its site as well as the damages, and the right answer is rarely to recover every week.
Accelerate for as long as the cost of recovering one more week is lower than what that week of lateness costs you: the liquidated damages rate plus your own time-related site costs for the week. With damages at 15,000 a week and site costs of 18,500, every week recovered is worth 33,500. On an illustrative project six weeks behind, buying back the four cheapest weeks for 101,500 and paying for the other two brings the cost down to 168,500, against 201,000 for doing nothing and 182,500 for recovering all six weeks.
Worked in full in Construction Cost Control by Julian R. Sterling, with every figure reproduced in a free workbook.See the book on Amazon →
A contractor running late through its own fault has no extension of time to claim, so every week past the completion date carries the damages rate. The tempting comparison sets each acceleration measure against that rate alone. It understates what a late week costs, because the contractor also keeps its site running: management, welfare, hoists, temporary services, the time-related part of the preliminaries. Finishing a week earlier saves both.
The other half of the decision is that acceleration measures are not equally priced. Saturday working on finishing trades is cheap per week recovered; a night shift is not. Sort them by cost per week recovered, and the answer becomes a marginal one: take measures in order until the next one costs more than the week it buys.
The case is illustrative, and it assumes every measure shortens the critical path. A measure that accelerates work with float recovers nothing, however much it costs.
| Input | Value |
|---|---|
| Liquidated damages, a week | 15,000 |
| Contractor's time-related site costs in the final weeks, a week | 18,500 |
| Cost of a late week to the contractor | 33,500 |
| Forecast culpable delay, weeks | 6 |
The site cost is the rate for the resources on site at the end of the job, not the average preliminaries rate: by the finishing period the tower crane and much of the scaffold have gone. The same distinction drives how prolongation costs are calculated.
Value of a week recovered = damages rate + time-related site cost a week
Cost per week of a measure = its cost ÷ weeks it recovers on the critical path
Accelerate with every measure whose cost per week is below the value of a week, cheapest first, up to the weeks of delay
Excel, measures sorted by cost per week in D2:D7: =IF(D2<LD+Site,"yes","no"), and =SUMIF(E2:E7,"yes",C2:C7) for the cost
| Measure | Weeks recovered | Cost | Cost per week | Below 33,500? |
|---|---|---|---|---|
| Saturday working, finishing trades | 1.5 | 27,000 | 18,000 | yes |
| Weekday overtime, time and a half | 1.0 | 26,500 | 26,500 | yes |
| Second drylining gang | 1.5 | 48,000 | 32,000 | yes |
| Resequenced M&E with extra supervision | 1.0 | 35,000 | 35,000 | no |
| Night shift for screeds | 1.0 | 46,000 | 46,000 | no |
| Prefabricated risers | 0.5 | 30,500 | 61,000 | no |
The overtime line already allows for the fact that overtime is not full value: an hour paid at 1.5 times the rate, worked at 85 per cent of normal productivity, costs 1.76 times a normal productive hour. Pricing overtime at the pay premium alone makes it look cheaper than it is.
| Strategy | Weeks recovered | Acceleration cost | Cost of weeks still late | Total |
|---|---|---|---|---|
| Do nothing: pay damages, keep the site | 0.0 | 0 | 201,000 | 201,000 |
| Accelerate where cheaper than a late week | 4.0 | 101,500 | 67,000 | 168,500 |
| Recover all six weeks | 6.0 | 182,500 | 0 | 182,500 |
The first three measures recover 4.0 weeks for 101,500 and avoid 134,000, a saving of 32,500. The resequencing would recover a week for 35,000 to save 33,500, so it loses 1,500, and the night shift loses 12,500. Recovering every week costs 14,000 more than stopping at the right point. Finishing on time is not the objective; the lowest total cost is.
| Damages a week | Value of a week | Weeks accelerated | Acceleration cost | Total cost | Saved against doing nothing |
|---|---|---|---|---|---|
| 5,000 | 23,500 | 1.5 | 27,000 | 132,750 | 8,250 |
| 15,000 | 33,500 | 4.0 | 101,500 | 168,500 | 32,500 |
| 25,000 | 43,500 | 5.0 | 136,500 | 180,000 | 81,000 |
| 35,000 | 53,500 | 6.0 | 182,500 | 182,500 | 138,500 |
A low damages rate buys the client very little acceleration: at 5,000 a week the contractor recovers only the Saturday working. That is the client's side of the same arithmetic, and the reason a damages rate set well below the client's own weekly cost of delay, as in whether liquidated damages cover what a late week costs the client, leaves the client paying for most of its own delay.
Comparing each measure with the damages rate alone. At 15,000 a week no measure on this list qualifies, the cheapest costing 18,000 a week, so the contractor finishes six weeks late and spends 201,000, 32,500 more than it needed to. The site costs are as real as the damages; they are just less visible because they are already in the cost report.
If the delay is not the contractor's fault, the question changes: the contractor is entitled to an extension of time, and acceleration should be instructed and paid for by the client. Accelerating to beat damages that the contract would not have imposed spends the contractor's money on the client's risk. Settle entitlement before choosing the measures.
Damages, extensions of time and their effect on the final account are worked through the book's Pemberton Yard case, and the free workbooks for this case carry it from the contract sum to the final account.
Only for weeks that cost less to recover than they cost to lose, and a lost week costs the damages plus the contractor's own site costs. With 15,000 of damages and 18,500 of time-related site costs, any measure below 33,500 per critical-path week pays. In the illustration that is three measures recovering 4.0 weeks for 101,500.
Usually not. Acceleration gets dearer per week as the cheap measures run out. In the illustration recovering all six weeks costs 182,500, while stopping after the four weeks that cost less than 33,500 each and paying for the last two costs 168,500, 14,000 less.
Overtime is paid at a premium and worked at lower productivity, so its real cost per productive hour is higher than the pay rate suggests. At 1.5 times pay and 85 per cent productivity, an overtime hour costs 1.76 times a normal productive hour. Price acceleration measures on weeks actually recovered, not on hours paid.
This article is one calculation from Construction Cost Control. The book takes the same case from first principles to the decision, chapter by chapter, and every figure it prints is a live formula in the free companion workbooks.
Get the book on Amazon →Free companion files
Also on Amazon UK · Amazon Germany · Amazon France · Amazon Canada
Reading guide: business operations: cost, contracts, pricing and supply chain → · All 1340 articles →
If this book helped, or didn’t, a few lines on Amazon are worth more than they look: they are what the next reader goes on. Write a review. The workbook stays free either way.