Chapter 29 — Quiz

C. Percent complete is a cost and earned value input — it answers "how much of the money have we earned." A schedule needs remaining duration in days from the data date, which is a completely different question and frequently produces a different answer. An activity 70 percent complete after 21 days of a 30-day duration "should" have 9 days left; if the foreman says 16, you have learned something no percentage would have told you.

C. The data date is the instant the update describes; the issue date is when it goes out. On Northgate the data date is the last calendar day of the month and the issue date is the fifth working day of the next one. Confusing the two is how you end up with a report that appears to forecast the past.

C. Retained logic honors the relationship for the remaining work: B's 10 remaining days cannot begin until A finishes at day 20, so B finishes at day 30. Progress override would allow B's remaining 10 days to run from the data date, finishing at day 10 — a 20-day difference from one setting. The professional answer is neither: if the two answers are far apart, the logic no longer describes the job and the activities should be split by area.

C. A chain consumes its float before it consumes the completion date, so float erosion moves weeks or months earlier than any date variance. Northgate's curtain-wall path went 34 → 32 → 26 → 19 → 11 → 3 over six updates while the projected completion date never moved. By the time A or B fires, the cause is old and the cheap fix is gone.

B. Excusable answers "does the contractor get a time extension"; compensable answers "does the contractor get paid for it." Weather beyond the contract allowance, market-wide strikes, and — in most treatments — true concurrent delay land here. The contractor gets days and both parties eat their own costs.

C. Zero. Consuming float is not extending the project. The subcontractor spent 9 of the path's 15 days; the path now has 6 left; the completion date did not move and nobody incurred extended general conditions. This is the most common error in delay analysis and it runs in both directions — contractors claiming time for float consumption, and owners alleging concurrency against it.

But note the second half: the float is now gone and somebody should write that down. A contemporaneous note that the path fell from 15 to 6 costs nothing and is worth a great deal if it later becomes controlling.

B. Starting from the baseline ignores everything that actually happened before the event — including, on Northgate, the fact that the procurement chain's float had already been reduced. Starting from today's schedule contaminates the answer with everything that happened after. The accepted update in effect at the time was prepared by people who did not know the event was coming and therefore had no motive to shade it, which is exactly what makes it credible.

B. It is the cost of buying one calendar day from that activity. Northgate's steel erection: ($4,222,000 − $4,180,000) ÷ (84 − 77) = $42,000 ÷ 7 = $6,000 per calendar day.

B. Recalculate. Crashing shifts the critical path. Money spent compressing an activity that is no longer critical buys exactly zero days. In the chapter's worked example, four days of compression on the enclosure branch made the MEP branch co-critical — without anybody touching a single MEP activity — and every dollar after that point had to be spent upstream of the split or on both branches at once.

C. 720 attendance hours at an assumed 0.85 sustained efficiency deliver 612 productive hours against 480 at 40 hours — +27.5 percent output for +69.8 percent cost. You are buying schedule at roughly two and a half times the normal price of the work. And of the extra cost, the fatigue loss ($7,092 on the chapter's crew) is larger than the entire overtime premium ($5,520), which is the half almost nobody prices.

C. A written directive is directed acceleration, which is normally a straightforward change. Constructive acceleration is what happens without a directive: an excusable delay, a proper extension request, a denial or unreasonable deferral, insistence on the original date, and actual acceleration with actual cost. The doctrine and its elements vary by jurisdiction and by contract form, and some contracts try to disclaim it.

C. All four are gates, but the one contractors lose on most often is the last one, because it is the only one that requires action while the work is happening. If premium time for the acceleration sits in the same payroll codes as the overtime the job was already running, no analyst on earth can separate them. Open separate cost codes on the day you start accelerating — thirty minutes of setup for a six-figure position.

False. It is equally consistent with float being consumed silently on non-controlling paths, or with the schedule not being updated at all — remaining durations derived from the baseline produce a model that is arithmetically incapable of changing its answer. The completion date is a lagging indicator; read it only alongside the float trend.

False. Liquidated damages compensate for late completion. A delay that consumes float does not move the completion date and therefore causes no compensable lateness. The owner's legitimate concern is different and real: the float is now gone, and the next slip on that path — anybody's slip — will move the date.

True, and it is a trap worth naming out loud. Do both things separately: cooperate on the recovery and reserve your entitlement position in writing, in the same transmittal. Two sentences. The contractor who refuses to submit until entitlement is settled is being obstructive; the contractor who submits silently is giving away an argument for free.

True. Negative float is the arithmetic statement this chain must be shortened by that many days, or the constrained date moves by that many days. It appears whenever the calculated finish is later than a required or imposed date. It is not "less than no float" — it is the size of the recovery you owe.

False, and this is the chapter's most expensive point. The cost slope prices the compression. It does not price the consequence. On Northgate, activity L — deck-edge detailing — had the second cheapest slope on the page at $7,000/CD, bought 4 days for $28,000, and produced roughly $494,000 of deck-edge rework and curtain-wall field modification. The crash table cannot see tolerance, trade density, or fatigue.

True as a general principle, with a real caveat. Neither party can show the other's conduct was the operative cause, because each party's own conduct was independently sufficient — so the contractor gets time but not money, and the owner gets no liquidated damages. The treatment of concurrency varies substantially by jurisdiction and by contract form, and some contracts define or allocate it expressly. Never state the general principle as if it were the law where you are.

Float belongs to the path, not to the activities on it, and it can be spent exactly once by whoever gets to it first. Kestrel's six days took the pool from +7 to +1; Caldwell's four then took it to −3 — three days of negative float, which is precisely the margin by which the April 18 mill release was missed. Because the constraint at the end of that chain was discrete rather than continuous, three days of lateness became a 35-day wait for the next rolling opening, and after Ironbridge's 12-day recovery, a 23-day project delay worth $244,950.

Identification (including retained logic or progress override) · progress this period · current forecast and variance · the controlling path · near-critical paths with their trend · changes to the model · delays and impacts with the entitlement position · recovery actions with dates · requirements of others · attachments.

The section most often omitted is changes to the model. What goes wrong: somebody re-sequences three relationships to fix a real problem, the completion date improves by nine days, and nobody writes it down. Four months later an owner's consultant compares two consecutive files, finds eighteen changed relationships and no explanation, and every number you produce for the rest of the job carries an asterisk. It is usually carelessness rather than dishonesty — which does not help you at all when the comparison report is on the table.

The 26-day path. First, rate over level: losing 14 days a month means it is controlling in under eight weeks, and a path burning float at that rate has a mechanism behind it that is still running — you do not have a float problem, you have an unidentified production or procurement problem currently spending your float. Second, the cheap fix lives where there is still room: intervening on a path with 26 days is a management decision, while intervening on a path with 2 days is an acceleration, and acceleration costs roughly ten times as much.

The 2-day path is stable, and stable at a low level means something is holding it there. Watch it; do not spend your morning on it.

(a) Forward pass from the data date.

   B:  in progress, 34 CD remaining →  180 + 34 = 214
   C:  in progress, 44 CD remaining →  180 + 44 = 224
   D:  ES = max(B 214, C 224) = 224  →  224 + 58 = 282
   E:  ES = 282                      →  282 + 32 = 314

Projected completion: day 314, against a required day 300 — 14 calendar days late, and 14 days later than the baseline's 300.

(b) Controlling path: A → C → D → E. Note that it is not the enclosure. In the baseline, C finished at 210 and B at 190, so C was already governing D; the update has widened that. Anyone managing this job by watching the curtain wall — which is the loud problem, with a supplier and a story attached — is watching the wrong chain.

(c) Total float against the required day 300.

   E:  LS = 300 − 32 = 268
   D:  LF = 268, LS = 268 − 58 = 210
   C:  LF = 210, LS = 210 − 44 = 166   TF = 166 − 180 = −14   ◄── controlling
   B:  LF = 210, LS = 210 − 34 = 176   TF = 176 − 180 =  −4

C is at −14; B is at −4. Both are negative, which means both chains have to be shortened — but shortening B alone buys nothing, because C governs the merge at D until B and C are within 10 days of each other.

(d) Exposure: 14 CD × $7,300/CD = $102,200.

(e) A recovery plan, and no notice.

Both causes named in the problem are yours: your steel erector and your curtain-wall supplier. A subcontractor's or supplier's performance is non-excusable as between you and the owner — you gave no notice because there is nothing to notice. What you owe the owner is disclosure in the narrative and a recovery plan.

Two further moves a good project manager makes here. First, the recovery money goes on C, not on B — the loud problem is not the controlling one, and this is exactly the situation where a project spends $80,000 accelerating the thing everybody is talking about and gains zero days. Second, look backward at the same time: pull the last three updates and find out when C's float went negative, because if it has been eroding since day 90 there is a management failure here that is separate from, and larger than, either subcontractor.