Vose SoftwareModelChoice decision tree · Oil and gas
Refinery Upgrade Decision
Add a coker, upgrade the hydrotreater, convert to biofuels, or sell the refinery as it is?
From the model’s decision tree
Best first decision: Full Coker Addition ($800M)
The expected value of each first option when every later decision is also made well (the tree rolled back).
| First decision | Expected value | |
|---|---|---|
| Full Coker Addition ($800M) | $795,625,000 | ✔ best |
| Hydrotreater Upgrade ($300M) | $546,125,000 | |
| Convert to Biofuels ($500M) | $584,625,000 | |
| Sell Refinery As-Is | $415,000,000 |
| Measure | Value |
|---|---|
| Expected value of the best strategy | $795,625,000 |
| Chance of a loss on the best strategy | 6.2% |
| Best outcome on the best strategy | $2,000,000,000 |
| Worst outcome on the best strategy | −$450,000,000 |
S-curve (cumulative probability) of Full Coker Addition ($800M) from the saved simulation.
The decision
A 150,000 bbl/day refinery built in the 1970s faces increasing pressure from tightening emissions regulations, aging equipment, and shifting crude supply patterns toward heavier, more sour crudes. The owner must decide on a major capital investment strategy for the next 20 years. The energy transition adds uncertainty about long-term demand for refined petroleum products.
Objective
Maximize expected NPV over a 20-year horizon. The decision involves tradeoffs between traditional refining profitability, energy transition risk, and the option value of selling before committing major capital.
The options
- Option 1: Full Coker Addition ($800M)
Install a delayed coker unit to process heavy sour crude at a significant discount to light sweet benchmarks. This is the highest-cost option but captures the heavy-light crude differential and widens the refinery's crude slate. Returns depend heavily on crack spreads and the H-L differential.
- Option 2: Hydrotreater Upgrade ($300M)
Upgrade the hydrotreating capacity to meet Tier 3 gasoline and IMO 2020 marine fuel specs. Enables processing of medium sour crudes. Lower capital requirement but doesn't capture the heavy crude discount.
- Option 3: Convert to Biofuels Terminal ($500M)
Repurpose the refinery for renewable diesel and sustainable aviation fuel production. An energy transition play that bets on strong biofuel demand and supportive government policy (RFS credits, LCFS, blender's tax credit).
- Option 4: Sell Refinery As-Is
Sell the facility to another operator or financial buyer. Sale price depends on current refining market conditions and buyer appetite.
The decision tree
The tree has 1 decision, 13 chance events and 30 outcomes. The best path is open: at every decision the best choice is marked, and the expected value beside each branch counts every cash flow from the start. Open any other branch to see what follows it.
DecisionChance event+$ cash flow on the branch
Full Coker Addition ($800M) −$800,000,000expected $795,625,000 ✔ best
Chance: Crack Spread Environment30% Wide Crack Spreadsexpected $1,460,000,000
Chance: Heavy-Light Differential (Wide)- 35% Large H-L Differential +$2,800,000,000payoff $2,000,000,000
- 40% Moderate H-L Differential +$2,200,000,000payoff $1,400,000,000
- 25% Small H-L Differential +$1,600,000,000payoff $800,000,000
45% Normal Crack Spreadsexpected $760,000,000
Chance: Heavy-Light Differential (Normal)- 35% Large H-L Differential +$2,100,000,000payoff $1,300,000,000
- 40% Moderate H-L Differential +$1,500,000,000payoff $700,000,000
- 25% Small H-L Differential +$900,000,000payoff $100,000,000
25% Narrow Crack Spreadsexpected $62,500,000
Chance: Heavy-Light Differential (Narrow)- 35% Large H-L Differential +$1,300,000,000payoff $500,000,000
- 40% Moderate H-L Differential +$800,000,000payoff $0
- 25% Small H-L Differential +$350,000,000payoff −$450,000,000
Hydrotreater Upgrade ($300M) −$300,000,000expected $546,125,000
Chance: Crack Spread (Hydrotreater)30% Wide Crack Spreadsexpected $922,500,000
Chance: Regulatory Environment (Wide)- 30% Strict New Regulations +$1,400,000,000payoff $1,100,000,000
- 45% Moderate Regulation +$1,200,000,000payoff $900,000,000
- 25% Status Quo Regulation +$1,050,000,000payoff $750,000,000
45% Normal Crack Spreadsexpected $520,000,000
Chance: Regulatory Environment (Normal)- 30% Strict New Regulations +$950,000,000payoff $650,000,000
- 45% Moderate Regulation +$800,000,000payoff $500,000,000
- 25% Status Quo Regulation +$700,000,000payoff $400,000,000
25% Narrow Crack Spreadsexpected $141,500,000
Chance: Regulatory Environment (Narrow)- 30% Strict New Regulations +$550,000,000payoff $250,000,000
- 45% Moderate Regulation +$420,000,000payoff $120,000,000
- 25% Status Quo Regulation +$350,000,000payoff $50,000,000
Convert to Biofuels ($500M) −$500,000,000expected $584,625,000
Chance: Biofuel Demand Outlook25% Strong Biofuel Demandexpected $1,275,000,000
Chance: Policy Support (Strong Demand)- 40% Strong Policy Support +$2,200,000,000payoff $1,700,000,000
- 35% Moderate Policy Support +$1,700,000,000payoff $1,200,000,000
- 25% Weak Policy Support +$1,200,000,000payoff $700,000,000
45% Moderate Biofuel Demandexpected $572,500,000
Chance: Policy Support (Moderate Demand)- 40% Strong Policy Support +$1,400,000,000payoff $900,000,000
- 35% Moderate Policy Support +$1,000,000,000payoff $500,000,000
- 25% Weak Policy Support +$650,000,000payoff $150,000,000
30% Weak Biofuel Demandexpected $27,500,000
Chance: Policy Support (Weak Demand)- 40% Strong Policy Support +$800,000,000payoff $300,000,000
- 35% Moderate Policy Support +$450,000,000payoff −$50,000,000
- 25% Weak Policy Support +$200,000,000payoff −$300,000,000
Sell Refinery As-Isexpected $415,000,000
Chance: Refinery Sale Market- 30% Strong Buyer Interest +$600,000,000payoff $600,000,000
- 40% Moderate Interest +$400,000,000payoff $400,000,000
- 30% Weak Market +$250,000,000payoff $250,000,000
Change the numbers. Open the workbook with ModelChoice, change a probability or a cost in the tree, and the best decision updates at once.
Key assumptions
| Assumption | Value | Notes |
|---|---|---|
| Coker Addition Capital | $800,000,000 | Delayed coker + utilities, 3-year build |
| Hydrotreater Capital | $300,000,000 | New HDS unit + hydrogen plant |
| Biofuels Conversion Capital | $500,000,000 | Pretreatment + hydrotreating reactors |
| Refinery Capacity | 150,000 bbl/day | Current nameplate, Gulf Coast location |
| Evaluation Horizon | 20 years | NPV of operating cash flows |
| Crack Spread (Wide) | $18-22/bbl | Strong refining margins cycle |
| Crack Spread (Normal) | $12-15/bbl | Mid-cycle refining margins |
| Crack Spread (Narrow) | $6-9/bbl | Weak margins, overcapacity |
| Heavy-Light Differential | $5-20/bbl | Key driver for coker economics |
| Biofuel Policy Risk | High | Credits range $0.50-2.00/gal, uncertain |
| Sale Price Range | $250-600M | Depends on buyer market conditions |
The figures are illustrative: the companies and numbers are invented for teaching.
Try it in ModelChoice
- Strategy Table. Compare coker vs hydrotreater vs biofuels vs sell
- Sensitivity Analysis. Impact of crack spreads and H-L differential on coker NPV
- Risk Profile. Compare downside exposure of each investment path
- Two-Way Sensitivity. Biofuel demand vs policy support for conversion option
Find similar models
- Product
- Type
- Application
- Uses
Further reading
Download the workbook and open it with ModelChoice. The free trial opens every example.
Related example models
- ModelChoice★★★★★Mature Oil Field: Enhanced Oil Recovery DecisionFor a mature oil field: CO₂ injection, water alternating gas, a polymer pilot, primary recovery, or decommissioning?Three decisions and 11 chance events. Best first decision: Continue Primary Recovery, expected value $48.0m.
- ModelChoice★★★★★Offshore Drilling Campaign DecisionDrill an exploration well now, shoot 3D seismic first, farm out the licence, or give it up?Six decisions and 14 chance events. Best first decision: Acquire 3D Seismic First, expected value $40.0m.
- ModelChoice★★★★★LNG Export Terminal InvestmentBuild a full-scale LNG export terminal, a modular first phase, a floating unit, or nothing?Two decisions and ten chance events. Best first decision: Build Modular (Phase 1), expected value $1.32bn.
An illustrative example: the figures are invented for teaching. Last updated 9 October 2026.