7.3 Conceptual & Parametric Estimating Methods

7.3 Conceptual & Parametric Estimating Methods

Quick Summary: Conceptual estimating provides rapid, cost-effective capital forecasts during early stage gates (Class 5 and Class 4) when engineering definition is below 15%. Methodologies range from analogous scaling (adjusting past reference projects for time, location, and scope) to capacity-factored estimating via the non-linear power-sizing exponent equation ($Cost_B = Cost_A \times (Cap_B / Cap_A)^e$), famously known as the Six-Tenths Rule when $e = 0.6$. Parametric estimating leverages statistically validated Cost Estimating Relationships (CERs) that connect physical cost drivers (gross area, volume, megawatts) to cost via regression models ($R^2 \ge 0.80$). Finally, equipment-factored estimating multiplies major delivered equipment costs by empirical ratios—such as the universal Lang factor (3.10 for solids, 4.74 for fluids) or modular Hand factors per equipment class.

---

1. The Role of Conceptual Estimating in Total Cost Management

During the conceptual screening and feasibility stages of an asset's lifecycle, owner organizations must evaluate dozens of competing technology alternatives and project configurations. Commissioning detailed engineering drawings for each alternative is economically unfeasible.

Conceptual estimating solves this challenge by trading micro-level detail for macro-level statistical speed. These methods are predominantly stochastic, deriving project cost from historical empirical relationships rather than itemized takeoffs.

+-----------------------------------------------------------------------------------+
|                    CONCEPTUAL ESTIMATING METHODOLOGY SPECTRUM                     |
|                                                                                   |
|  [ Analogous Estimating ]   --> Simple ratio scaling from a similar past project  |
|                                 (Adjusted for Time, Location, and Capacity)       |
|                                                                                   |
|  [ Capacity-Factored ]      --> Non-linear power-sizing exponent law (e = 0.6)    |
|                                 Cost_B = Cost_A * (Cap_B / Cap_A)^e               |
|                                                                                   |
|  [ Parametric CERs ]        --> Statistical regression models (Y = a + bX)        |
|                                 Driven by physical units (GSF, MW, Horsepower)    |
|                                                                                   |
|  [ Equipment-Factored ]     --> Multiplies major equipment by empirical factors   |
|                                 Lang Factor (Universal) & Hand Factor (By Type)   |
+-----------------------------------------------------------------------------------+

---

2. Analogous Estimating (Ratio & Historical Scaling)

Analogous estimating (or gross ratio estimating) uses the actual cost of a previously completed, similar reference project as the baseline for estimating the target project.

Because no two projects are identical, the historical cost must be systematically adjusted for three fundamental variance vectors:

1. Capacity / Size Delta ($F_{cap}$): Ratio of proposed capacity to reference capacity.
2. Time / Escalation Delta ($F_{time}$): Cost index ratio adjusting for market inflation between the historical construction date and the current project baseline.
3. Geographic Location Delta ($F_{loc}$): Location index adjusting for regional labor wages, productivity, site logistics, and import tariffs.

Mathematical Formulation

$\text{Cost}_{target} = \text{Cost}_{historical} \times \left(\frac{\text{Capacity}_{target}}{\text{Capacity}_{historical}}\right) \times \left(\frac{\text{Index}_{target\_time}}{\text{Index}_{historical\_time}}\right) \times \left(\frac{\text{Location Factor}_{target}}{\text{Location Factor}_{historical}}\right)$

Worked Example 1: Analogous Scaling

• Reference Project: A 50,000 barrel-per-day (bpd) atmospheric crude distillation unit built in Houston in 2020 for $$120,000,000$.
• Target Project: A 50,000 bpd identical distillation unit to be built in Philadelphia in 2026.
• Parameters:
  • Cost Index 2020 = $6,200$; Cost Index 2026 = $7,750$.
  • Location Factor Houston = $1.00$; Location Factor Philadelphia = $1.15$.
• Calculation: $\text{Cost}_{2026} = \$120,000,000 \times \left(\frac{50,000}{50,000}\right) \times \left(\frac{7,750}{6,200}\right) \times \left(\frac{1.15}{1.00}\right)$ $\text{Cost}_{2026} = \$120,000,000 \times 1.00 \times 1.25 \times 1.15 = \$172,500,000$

---

3. Capacity-Factored Estimating & The Six-Tenths Rule

When scaling process plants or industrial equipment across different production capacities, cost does not increase linearly with size. This phenomenon is governed by economies of scale.

The Power-Sizing Exponent Equation

$\text{Cost}_B = \text{Cost}_A \times \left( \frac{\text{Capacity}_B}{\text{Capacity}_A} \right)^e$ Where:

• $\text{Cost}_B$ = Estimated cost of target facility or equipment item.
• $\text{Cost}_A$ = Known historical cost of reference facility or equipment item.
• $\text{Capacity}_B$ = Rated capacity or physical output of target facility/equipment.
• $\text{Capacity}_A$ = Rated capacity or physical output of reference facility/equipment.
• $e$ = Capacity exponent (power-sizing scale factor).

+-----------------------------------------------------------------------------------+
|                         THE POWER-SIZING EXPONENT (e)                             |
|                                                                                   |
|      e < 1.0 (Typical: 0.6)            e = 1.0                     e > 1.0        |
|      ECONOMIES OF SCALE        CONSTANT RETURNS TO SCALE     DISECONOMIES OF SCALE|
|                                                                                   |
|  Unit cost decreases as        Doubling capacity             Unit cost increases  |
|  capacity expands. Surface     doubles cost exactly.         as size creates extreme
|  area grows slower than        Modular parallel trains       structural/metallurgy
|  internal volume.              with no shared infra.         penalties.
+-----------------------------------------------------------------------------------+

The Classic "Six-Tenths Rule" ($e = 0.60$)

In 1947, Roger Williams Jr. published the observation that for many types of chemical process equipment, capital cost varies as the $0.6$ power of capacity. This principle stems from the geometric relationship between the surface area and the volume of three-dimensional storage and pressure vessels:

• Internal volume (capacity) scales as the cube of linear dimensions ($V \propto r^3$).
• Surface area (metal plate weight, fabrication welding, and painting cost) scales as the square ($A \propto r^2$).
• Therefore, $Cost \propto (Capacity)^{2/3} = (Capacity)^{0.67} \approx (Capacity)^{0.60}$.

Typical Empirical Exponents by Industry & Equipment

Worked Example 2: Step-by-Step Capacity-Factored Calculation

Scenario: An industrial chemical producer constructed a $100,000$-gallon chemical processing vessel in 2024 for a direct delivered cost of $$2,400,000$. The company plans to construct a $250,000$-gallon vessel of identical design and metallurgy at the same facility. Engineering records indicate an empirical exponent of $e = 0.65$.

1. State the Formula: $\text{Cost}_B = \text{Cost}_A \times \left( \frac{\text{Capacity}_B}{\text{Capacity}_A} \right)^e$
2. Identify Known Variables:
   • $\text{Cost}_A = $2,400,000$
   • $\text{Capacity}_A = 100,000\text{ gal}$
   • $\text{Capacity}_B = 250,000\text{ gal}$
   • $e = 0.65$
3. Compute the Capacity Ratio: $\text{Ratio} = \frac{250,000}{100,000} = 2.5$
4. Apply the Exponent: $2.5^{0.65} \approx 1.8114$
5. Calculate Target Cost: $\text{Cost}_B = \$2,400,000 \times 1.8114 = \$4,347,360$
6. Unit Cost Analysis (Economies of Scale Demonstrated):
   • Original unit cost: $$2,400,000 / 100,000\text{ gal} = $24.00/\text{gal}$
   • New unit cost: $$4,347,360 / 250,000\text{ gal} = $17.39/\text{gal}$
   • While capacity expanded by $150%$, total cost rose by only $81.1%$, yielding a $27.5%$ reduction in unit capital cost.

[!WARNING] Rule of Thumb for Capacity Ratios: The power-sizing exponent model is valid only within a capacity ratio range of $0.2$ to $5.0$ (a factor of 5). Beyond a 5-fold expansion, equipment faces severe physical transport limits, specialized metallurgy requirements, or the necessity of installing parallel twin units, which breaks the single-vessel scaling curve.

---

4. Parametric Estimating & Cost Estimating Relationships (CERs)

Parametric estimating is an advanced methodology that uses statistical regression to establish mathematical algorithms—termed Cost Estimating Relationships (CERs)—between physical, technical, or performance parameters (independent variables) and project cost (dependent variable).

Structure of Cost Estimating Relationships

1. Simple Linear CER: $Y = a + b(X)$
   • Where $Y$ = Total Cost, $a$ = Fixed baseline setup cost, $b$ = Variable cost per unit, $X$ = Independent cost driver (e.g., Gross Square Feet).
2. Multi-Variable Linear CER: $Y = a + b_1(X_1) + b_2(X_2) + b_3(X_3)$
   • Allows cost to be driven by multiple physical attributes (e.g., building cost driven by area, exterior wall perimeter, and electrical service load).
3. Non-Linear / Log-Linear CER: $Y = a(X)^b \quad \iff \quad \ln(Y) = \ln(a) + b \cdot \ln(X)$

Statistical Validation of CERs

Cost engineers must validate CER models using rigorous descriptive and inferential statistics:

• Coefficient of Determination ($R^2$): Measures the proportion of the total variance in cost explained by the independent variables. A credible CER requires $R^2 \ge 0.80$ ($80%$ of cost variance explained by the model).
• Standard Error of the Estimate ($SE$): Measures the standard deviation of historical data points around the regression line, establishing confidence bands.
• $F$-Significance and $t$-Statistics: Confirm that the relationship between the cost driver and project cost is statistically significant and not the result of random chance ($p\text{-value} < 0.05$).

Worked Example 3: Multi-Variable Parametric CER

Scenario: A healthcare construction firm maintains a calibrated parametric CER for regional medical clinics: $\text{Capital Cost} = \$650,000 + (\$240 \times GSF) + (\$45,000 \times \text{Beds}) + (\$180,000 \times \text{Operating Rooms})$ Estimate the cost of a new 35,000 Gross Square Foot (GSF) surgical clinic featuring 20 patient recovery beds and 3 state-of-the-art operating suites.

• Calculation: $\text{Cost} = \$650,000 + (\$240 \times 35,000) + (\$45,000 \times 20) + (\$180,000 \times 3)$ $\text{Cost} = \$650,000 + \$8,400,000 + \$900,000 + \$540,000 = \$10,490,000$

---

5. Equipment-Factored Estimating Methods

In heavy industrial, power, and process plants, major mechanical equipment represents the physical nucleus of the facility. Direct bulk commodities—piping, structural steel, electrical power, concrete foundations, insulation, and paint—scale in direct proportion to the purchase price of the equipment.

+-----------------------------------------------------------------------------------+
|                    EQUIPMENT-FACTORED ESTIMATING ARCHITECTURE                     |
|                                                                                   |
|   [ Total Delivered Equipment Cost ]  === (Bare Mechanical Equipment F.O.B.)      |
|                  |                                                                |
|                  +--------------------+                                           |
|                  |                    |                                           |
|                  v                    v                                           |
|         [ LANG FACTOR METHOD ]   [ HAND FACTOR METHOD ]                           |
|         Universal Multiplier     Modular Multipliers                              |
|         Applied to Plant Total   Applied Per Equipment Type                       |
|                                                                                   |
|         TPC = F_L * Sum(E)       Installed = Sum( E_i * F_Hi )                    |
|         - Solids Plant: 3.10     - Columns: 4.0                                   |
|         - Mixed Plant:  3.63     - Heat Exchangers: 3.5                           |
|         - Fluid Plant:  4.74     - Pumps: 4.0                                     |
|                                  - Compressors: 2.5                               |
+-----------------------------------------------------------------------------------+

1. The Lang Factor Method (Hans J. Lang, 1947)

The Lang factor method estimates the total capital cost of a complete process facility by multiplying the total delivered cost of all major process equipment by an empirical universal factor:

$\text{Total Plant Cost (TPC)} = F_L \times \sum \text{Cost}_{delivered\_equipment}$

Where the classical Lang factors ($F_L$) are categorized by the physical nature of the process material:

Why Fluid Plants Have Higher Factors

Fluid processing plants require an enormous volume of high-pressure alloy piping, automated control valves, continuous instrumentation loops (DCS), complex structural piperacks, insulation, and fireproofing relative to their bare equipment footprint. In contrast, solid processing plants rely primarily on mechanical conveyors and hoppers with substantially lower piping and instrumentation intensity.

2. The Hand Factor Method (W.E. Hand, 1958)

A primary limitation of the Lang factor is that it applies a single, blunt multiplier across all equipment regardless of whether the plant consists of simple storage vessels or intricate turbo-compressors.

W.E. Hand refined Lang's approach by establishing individual installation factors ($F_H$) for each distinct category of mechanical equipment:

$\text{Total Installed Direct Cost} = \sum_{i=1}^{n} \left( \text{Equipment Cost}_i \times F_{Hi} \right)$

Typical Hand factors for delivered carbon steel equipment include:

• Distillation / Fractionation Columns: $4.0$
• Pressure Vessels & Drums: $4.0$
• Shell & Tube Heat Exchangers: $3.5$
• Direct-Fired Process Furnaces: $2.0$
• Centrifugal Pumps & Drivers: $4.0$
• Reciprocating Compressors: $2.5$ to $3.0$
• Atmospheric Storage Tanks: $2.0$ to $2.5$

Worked Example 4: Hand Factor Estimating

Scenario: A process plant design includes the following delivered major equipment packages:

1. 2 Distillation Columns: $$1,500,000$ (Hand factor = $4.0$)
2. 4 Heat Exchangers: $$600,000$ (Hand factor = $3.5$)
3. 6 Centrifugal Pumps: $$250,000$ (Hand factor = $4.0$)
4. 1 Fired Heater: $$1,000,000$ (Hand factor = $2.0$)

• Calculation: $\text{Total Direct Installed Cost} = (\$1,500,000 \times 4.0) + (\$600,000 \times 3.5) + (\$250,000 \times 4.0) + (\$1,000,000 \times 2.0)$ $\text{Total Direct Installed Cost} = \$6,000,000 + \$2,100,000 + \$1,000,000 + \$2,000,000 = \$11,100,000$
• Notice that total delivered equipment equals $$3,350,000$. The resulting composite plant factor is $$11,100,000 / $3,350,000 = 3.31$, providing far greater fidelity than a generic blanket multiplier.

---

6. Exam Watch: High-Yield Traps & Rules of Thumb

[!WARNING] The "Delivered vs. Installed" Equipment Base: On CCT exam calculations involving Lang or Hand factors, verify whether the provided equipment costs are delivered (F.O.B. job site) or installed. Lang and Hand factors are calibrated to be applied strictly to delivered purchase equipment costs. Applying a Lang factor of 4.74 to installed equipment costs will artificially inflate the estimate by 400%!

[!CAUTION] The Exponent Trap ($e = 0.6$): Do not assume $e = 0.60$ unless the exam question explicitly states "using the Six-Tenths Rule" or provides no exponent. If the problem specifies an exponent (e.g., $e = 0.72$), you must use the specified value!

[!TIP] Linear vs. Factored Extrapolation: If an exam question asks what happens to the capital cost when plant capacity doubles under an exponent of $e = 1.0$, the cost doubles ($2.0^{1.0} = 2.0$). Under $e = 0.60$, the cost increases by only $51.6%$ ($2.0^{0.60} \approx 1.516$).