AACE CCT cheat sheet

AACE estimate classes run from Class 5 (order of magnitude, 0–2% project definition) to Class 1 (definitive, 65–100% definition); class is assigned by end usage and definition maturity, not by the estimating method used.

Approximate accuracy ranges by class: Class 5 about −20% to −50% low / +30% to +100% high; Class 3 about −10% to −20% / +10% to +30%; Class 1 narrows to about −3% to −10% / +5% to +15%.

Estimate structure: direct costs + indirect costs + contingency + escalation. Management reserve is held outside the project cost baseline and is not part of the estimate of planned work.

Factor scaling: C2 = C1 × (Q2/Q1)^n. The six-tenths rule uses n ≈ 0.6 for capacity-based equipment or plant costs and is valid only over moderate ranges with an empirically validated exponent.

Learning curve: unit cost Tn = T1 × n^b, with b = log(learning rate)/log 2. An 80% curve gives b ≈ −0.322; unit cost and cumulative average cost follow different curves.

Escalation factor = cost index at target date ÷ cost index at base date. Escalation addresses price-level change; contingency addresses scope and risk uncertainty inside approved scope.

Methods ranked by accuracy at a given definition level: bottom-up (highest accuracy, highest effort), detailed/parametric, factored, analogy, order-of-magnitude (lowest accuracy, lowest effort).

Cost versus price: cost is the expenditure required to deliver scope; price = cost + markup (profit, taxes, fees). Markup on cost is not equal to margin on price.

Contingency is quantified through risk analysis - expected value, Monte Carlo simulation, or percentage ranges by estimate class - and is drawn down as identified risks are retired.

Forward pass: ES = maximum predecessor EF; EF = ES + duration. Backward pass: LF = minimum successor LS; LS = LF − duration.

Total float = LS − ES = LF − EF. Free float = minimum successor ES − EF. The critical path is the longest path and carries the minimum total float, normally zero.

PERT expected duration = (O + 4M + P)/6; standard deviation = (P − O)/6; variance = ((P − O)/6)^2. Used for probabilistic duration and schedule risk analysis.

Activity duration = quantity of work ÷ (productivity rate × number of resources); resource-driven durations change whenever crew size changes.

Relationship types are FS, SS, FF and SF, each with optional lag (+) or lead (−). Lag is time added to the logic link, not a separate activity.

Resource leveling may extend project duration and shift the critical path; resource smoothing (time-limited) consumes float but does not delay the project finish.

Schedule hierarchy: Level 1 executive milestone summary, Level 2 summary, Level 3 project control schedule, Levels 4–5 execution detail. The control schedule is the baseline for progress measurement.

Constraints such as start-no-earlier, finish-no-later and mandatory dates override network logic and can produce negative float; open ends and redundant logic reduce schedule quality.

Milestones have zero duration. Hammock or level-of-effort activities derive duration from the activities they span and do not carry their own earned value.

CV = EV − AC and SV = EV − PV (both in cost units). CPI = EV/AC and SPI = EV/PV; negative variances and indices below 1.0 are unfavorable.

EAC options: BAC/CPI when current cost efficiency continues; AC + (BAC − EV) when the variance is atypical; AC + (BAC − EV)/(CPI × SPI) when both cost and schedule drivers persist.

ETC = EAC − AC; VAC = BAC − EAC; TCPI = (BAC − EV)/(BAC − AC) against the original budget, or (BAC − EV)/(EAC − AC) against the current forecast.

Earned schedule: SV(t) = ES − AT and SPI(t) = ES/AT, where ES is the time at which EV should have been earned; avoids the convergence of cost-based SPI to 1.0 at completion.

Fixed-formula earning rules are 0/100, 50/50 and milestone-based; percent complete and equivalent units suit longer or repetitive work. Rule choice should match task duration.

Cost-based percent complete is distorted by front-loaded costs (mobilization, major equipment); physical progress or earned hours is preferred for status reporting.

Forecasting assumes remaining work follows the efficiency that produced CPI (and SPI); any EAC should be validated against a bottom-up re-estimate of the remaining scope.

Only approved changes modify the cost and schedule baselines; unapproved scope is tracked as a pending change and excluded from earned value.

Labor performance: earned hours = budgeted hours × percent complete; comparing earned hours to actual hours gives the labor performance factor used in productivity control.

NPV = Σ CFt/(1 + r)^t − initial investment; accept when NPV > 0 at the hurdle rate. IRR is the discount rate that makes NPV = 0 and is compared with cost of capital.

Payback = initial investment ÷ annual net cash flow; discounted payback uses discounted flows. Neither measures profitability beyond the payback point.

Break-even units = fixed costs ÷ (unit price − unit variable cost); contribution margin ratio = (price − variable cost) ÷ price.

Margin = markup ÷ (1 + markup); a 20% markup on cost equals a 16.7% margin on price. Profit = revenue − total cost.

Straight-line depreciation = (cost − salvage value) ÷ useful life; accelerated methods (MACRS, double-declining balance) change only the timing of after-tax cash flow.

Sunk costs are irrelevant to decisions; relevant measures are incremental cash flow and opportunity cost when comparing alternatives.

Direct costs are traceable to a specific deliverable; indirect costs (overhead, field supervision, home office) are allocated using a defined basis such as labor hours or direct cost.

Life-cycle cost = capital + operating + maintenance + disposal costs over the asset life; the lowest first cost does not imply the lowest total cost.

WACC is the after-tax weighted average of debt and equity costs and is the discount rate used for project NPV when project risk matches company risk.