Chapter 11 — Quiz

C. You cannot out-effort missing information. If the drawings do not say what the exterior wall is, no amount of care tells you what it costs. This is the premise of the entire estimate-class system: Class 5 (program only) carries a range of roughly −30/+50%, Class 1 (bid documents) roughly −3/+10%, and the difference between them is information, not diligence.

B. $/SF and cost per functional unit are available earlier, at programming, and remain useful as cross-checks. Full takeoff with subcontractor quotes requires construction documents. At SD you know the systems — steel frame, curtain wall, VAV — without the details, which is exactly the state UniFormat's element-based structure is built for.

C. Money is spent across the construction period, not on day one, so the weighted-average date of the spend is close to the midpoint. On Northgate, escalating to NTP instead of the midpoint would have understated the estimate by $1,049,957 — and it would not have looked like a mistake, which is what makes it dangerous.

B. The recorded cost is dollars spent across that project's whole duration, so its effective "as of" date is roughly the middle of construction. Fairhaven completed 26 months ago and took 16 months, so its midpoint is 34 months ago. Using completion instead understates the adjusted cost by roughly 2.4% — about $1.2 million on a Northgate-sized project.

B. Value = Function ÷ Cost. You raise value by delivering the same function for less, or more function for the same. Delivering less function for less money is a scope reduction — which may be the right decision, but must be named as what it is and chosen by the owner with open eyes.

B. (1 + 0.06 + 0.03 + 0.0643) × 1.04 = 1.1543 × 1.04 = 1.2005. Practically: every dollar of work removed takes about $1.20 off the bottom line, so closing a $4,800,591 gap requires finding about $3,998,920 of actual work, not $4.8 million of it.

C. Function is held exactly — the building resists lateral load just as well — and the saving comes from a more efficient way of delivering it. A reduces function (durability and infection control) and is a scope cut. B changes nothing about the project and only removes the funding for identified risk. D may be legitimate phasing, but usually returns at a 30–40% premium and must be priced with the return cost written down.

C. Contingency is not money; it is an admission. Deleting the reserve does not delete the risk — it removes the funding and transfers the exposure to whoever is standing closest when it lands, usually as a budget overrun, a change order, or a scope cut made under time pressure. A CM who allows this has done the owner harm even though the owner asked for it.

B. UniFormat organizes by building element — substructure, shell, interiors, services, equipment, special construction, building sitework — which is what you know at SD. MasterFormat organizes by work result and is the structure for specifications and detailed bids, which is what you know at CD.

C. $47,500,000 ÷ $61,000,000 = 77.9%. The other 22.1% — design fees, $3,640,000 of owner-furnished imaging equipment, FF&E, permits, testing and commissioning, financing, and $1,525,000 of owner's contingency — is where three of the most common CM advice failures live.

B. A GMP is guaranteed only against the scope defined in the GMP documents, and the drawings alone never define it completely. The qualifications page converts "based on the documents and our understanding of the project" into enforceable sentences. On Bellhaven (case study 11.2), four missing qualification sentences cost about $480,000 — roughly $120,000 a sentence.

C. The member exists and can be fabricated; it simply cannot reach the site as drawn. Access and availability questions — can the material, equipment, and people physically get to the work, and does the specified product exist in this market at a deliverable lead time — catch this class of finding. Cost on Northgate if missed: $36,000 and a detail revision.

True. Gross area, floor-to-floor height, stories, perimeter, skin-to-floor ratio, structural bay, program mix, and occupancy are known early and largely determine system costs — and across thirty-five elemental lines, high and low rate errors partially cancel. The corollary matters: when a building is genuinely unusual, the errors stop cancelling and start compounding.

False. Both can be wrong in the same direction — same market assumptions, same estimator, same anchor projects. Convergence tells you that you have not made a blunder. It is the absence of one kind of alarm, not evidence of accuracy.

False. Escalate only un-bought scope. Subcontractors holding fixed-price agreements now own their own escalation. This is why Northgate's escalation line fell from $2,801,338 at SD (nothing bought) to $575,200 at GMP (84% bought).

False. It is worth less, because less markup rides on it — the burden multiplier falls as contingency and escalation come down. And that is the small half of the effect; the large half is that a facade change costs a redline at SD and a redesign plus a resubmittal at 90% CD.

False. The equipment is the owner's, but the provisions for it — slab depression, structural framing, shielding, electrical feed — are in the contractor's scope. Northgate's CO #14 is exactly this: the imaging vendor selected a different MRI after the GMP was set, and the resulting work cost $186,400 against an owner's verbal understanding of "about $60,000."

False. A scope reduction the owner chooses knowingly, with the trade-off documented, is entirely legitimate — Northgate's café change (VE-16, $321,700) is one, and it was recorded as a program decision rather than as VE. What is not legitimate is a scope reduction presented as value engineering, so the owner discovers what they gave up later.

The remainder came from retiring uncertainty, not from removing work: the design contingency fell from $2,614,001 to zero as the documents completed, and the escalation allowance fell from $2,801,338 to $575,200 as the work was bought out at fixed prices. Meanwhile the cost of the work actually went up $233,310 — VE removed $2,521,700 of raw work and design development added $2,755,010 back.

It matters for two reasons. First, it is honest: most of what closes a budget gap between SD and GMP was never in the building. Second, only the VE is something you can promise an owner in advance. Contingency comes back only if the named risks do not happen; escalation comes down only if the buyout goes well. Those are forecasts you make and own, not commitments you can hand over in March.

Removing a scope item changes the project: something that was going to be built will not be built, and everyone can see what was given up. Removing the design contingency changes nothing about the project — the same building, the same systems, the same risks — it only removes the funding for cost the drawings are going to add anyway.

Concretely: Northgate carried $2,614,001 of design contingency at SD. Design development then added $2,755,010 to the cost of work. Had that contingency been cut in half to close the gap, the $2,755,010 would still have arrived; it would simply have arrived as an unfunded overrun in month nine, when the only remedies left are expensive. Deleting a scope item deletes a cost. Deleting a contingency deletes only the plan for one.

1. Does the function change? If yes, it is a scope cut, no matter how good the number is. Function is first because it is disqualifying.
2. Does life-cycle cost stay level or improve? A first-cost saving that raises operating cost is a transfer from the owner's capital budget to their operating budget, not a saving.
3. What schedule and quality risk does it carry, and is that risk written in the log? Those are real costs the estimate does not show.
4. Did the owner choose it, knowing what they give up, in writing? Last, because a scope reduction the owner makes deliberately is legitimate and one the owner discovers later is a betrayal.

The order encodes a hierarchy: disqualify first, then price the full cost, then price the risk, then confirm consent.

(a) (1 + 0.04 + 0.03 + 0.052) × 1.04 = 1.122 × 1.04 = 1.16688

(b) Gap = $30,640,000 − $28,000,000 = $2,640,000. Raw work required = $2,640,000 ÷ 1.16688 = $2,262,443.

(c) The cost of construction is $30,640,000 ÷ 1.16688 = $26,258,056. Cutting design contingency from 4% to 1% removes 3% × $26,258,056 = $787,742, plus fee at 4% = $31,509 — a "saving" of about $819,251, not $1,100,000. (To reach $1,100,000 you would have to drive the design contingency below zero.)

Why you decline: at design development the drawings are roughly 40% complete and will add real cost between now and permit; the contingency is the funded estimate of that cost, and cutting it does not cancel the cost, it only guarantees the cost arrives unfunded. Recreation centers are particularly exposed here — gym structure, pool or kitchen equipment coordination, and locker-room MEP are all classic late-growth areas.

(d) Three defensible avenues, each with the function held:

• Gymnasium roof structure — function: span the gym clear and carry roof and equipment loads. Compare steel joist and joist-girder, long-span joists, glulam, and panelized systems; the structural depth also interacts with building height and enclosure area.
• Mechanical zoning strategy — function: condition zones with radically different loads (gym, kitchen, offices, locker rooms). Consolidating units, adding energy recovery, and right-sizing rather than replicating a single approach across incompatible zones.
• Enclosure system rebalance — function: exclude weather, admit daylight, resist wind, meet the energy code U-value and SHGC. Shifting the ratio of glazing to opaque wall while holding thermal performance is often the largest single lever on a building with a high skin-to-floor ratio.

Weak answers here name products ("cheaper brick," "smaller windows") instead of functions — which is precisely the failure the function-analysis discipline exists to prevent.