16-Civ-B8 Management of Construction · December 2016

Question 3 of 6: Project Control

Nivaar worked solution (AI-drafted; not reviewed by a licensed engineer)

Question 3: Project Control (20 marks)

Question text not reproduced: the examination questions are © Engineers and Geoscientists BC. Open the official past paper (linked at the top of this page) to read the question, then follow the worked solution below.

(a) The S-curve, the expense-versus-payment profiles, and project financing

Cumulative expenditure on a construction project plots as a shallow S. Early on, only mobilisation, survey, site preparation and submittals are running, so the curve is flat; through the middle of the job the major trades are all working at once and the curve is steep; near the end only deficiency work, commissioning and demobilisation remain, so it flattens again. The same shape appears whether the ordinate is the owner’s budget, the contractor’s cost, or physical percent complete, which is what makes the S-curve the standard baseline against which progress is judged.

The payment profile is not the same curve. Cost is incurred continuously, day by day as labour is paid and material is delivered, but payment arrives in discrete monthly steps, after a progress claim has been submitted, certified by the consultant, and paid within the contractual period — typically 30 days under CCDC 2, and now within the statutory deadlines of the prompt-payment legislation being adopted across Canada. On top of that lag sits the statutory holdback (10 per cent in most provinces, retained until the lien period expires after substantial performance). The result is the picture in Figure 3.1: a smooth expense curve and a stepped payment curve that lies below it for essentially the whole job, closing only when the holdback is released. The vertical gap between them is money the contractor has spent and not yet been paid — the working capital it must finance, and the interest on that borrowing is a real project cost that must be recovered in the bid.

Several measures reduce that interest charge, and a competent contractor uses all of them:

• Negotiate a mobilisation payment or advance. An up-front payment against the mobilisation and bonding costs lifts the whole payment curve at day zero, which is where the deficit is proportionally worst.
• Front-end load the schedule of values (within honest limits). Assigning a fair but generous share of the price to early activities pulls cash forward. Loading it dishonestly is unbalanced bidding and will be rejected by the payment certifier, so the legitimate version of this is simply to make sure early work — excavation, foundations, site services — is priced at its true value rather than being subsidised from later trades.
• Bill promptly, completely and accurately. The single largest controllable lag is the contractor’s own claim preparation. A claim submitted on the first of the month with correct quantities and complete backup is certified on time; one submitted late or with disputed items sits for a further payment cycle.
• Match supplier and subcontractor terms to the owner’s payment cycle. Paid-when-paid clauses, negotiated 30- or 60-day supplier terms, and equipment rental rather than purchase all shift the outflow to the right and shrink the gap directly.
• Reduce the holdback exposure. Substantial performance should be certified as soon as it is genuinely achieved, and staged or partial releases negotiated for early-completing portions; where the contract allows, a holdback bond or letter of credit can be substituted for cash retention. Deficiency lists should be cleared quickly, because the last 1 per cent of work commonly holds 10 per cent of the money.
• Level the resource and expenditure profile. Smoothing peaks in the S-curve (and using float on non-critical work to defer expenditure rather than accelerate it) lowers the maximum cash exposure, which is what the line of credit must be sized for.
• Use the cheapest available financing. A negotiated operating line secured against receivables is far cheaper than trade credit taken by paying late, and much cheaper than the reputational cost of a lien filed by an unpaid subcontractor.

(b) Performance indices for schedule and cost control

Schedule and cost cannot be controlled independently, because a job can be on budget only by being behind, or on time only by spending more. Earned-value analysis solves this by measuring all three quantities in the same units — dollars of budgeted work — so that progress and expenditure can be compared directly. The three primitives are the planned value \(PV\) (budgeted cost of work scheduled), the earned value \(EV\) (budgeted cost of work actually performed) and the actual cost \(AC\) of that performed work.

Given. To make the indices concrete, take a job at its data date with \(PV = \$500{,}000\), \(EV = \$450{,}000\), \(AC = \$480{,}000\) and a budget at completion \(BAC = \$1{,}200{,}000\).

Find. The schedule and cost variances and indices, and the forecast cost at completion.

1. Variances measure the gap in dollars. The schedule variance compares work done with work planned, and the cost variance compares work done with money spent: $$SV = EV - PV = 450{,}000 - 500{,}000 = -\$50{,}000, \qquad CV = EV - AC = 450{,}000 - 480{,}000 = -\$30{,}000 .$$ Both are negative, so the job is behind schedule and over cost. Variances are useful for reporting an absolute exposure but they cannot be compared between projects of different size.
2. Indices normalise the same information into a ratio. Dividing rather than subtracting gives dimensionless numbers that can be trended and benchmarked: $$SPI = \frac{EV}{PV} = \frac{450{,}000}{500{,}000} = 0.90, \qquad CPI = \frac{EV}{AC} = \frac{450{,}000}{480{,}000} = 0.9375 .$$ $$\boxed{SPI = 0.90 \;(\text{10 percent behind}), \qquad CPI = 0.94 \;(\text{about 6 percent over cost})}$$ A value of 1.0 is on plan; below 1.0 is unfavourable in both cases.
3. Extrapolate to completion. If the cost performance achieved so far persists, the estimate at completion is the budget inflated by the reciprocal of the cost index: $$EAC = \frac{BAC}{CPI} = \frac{1{,}200{,}000}{0.9375} = \$1{,}280{,}000,$$ a forecast overrun of \(\$80{,}000\). The same logic applied to \(SPI\) gives a first estimate of the completion date, though schedule forecasting is better done by re-running the critical path than by extrapolating an index.

Beyond these four, three further indices are in routine use. The to-complete performance index, \(TCPI = (BAC-EV)/(BAC-AC)\), states the cost efficiency the remaining work must achieve to land on budget, and is the honest test of whether a recovery plan is credible — a \(TCPI\) far above the \(CPI\) actually being achieved means the plan is wishful. The critical ratio, \(CR = SPI \times CPI\), combines both dimensions into a single health indicator for portfolio reporting. And on the schedule side, earned schedule converts \(EV\) back into time units to correct the well-known defect that \(SPI\) drifts to 1.0 at completion no matter how late the project finishes, because at the end \(EV = PV = BAC\) by definition.