seeded Monte Carlo cases generated in Docker
Docker-Generated Monte Carlo Screen
Uncertainty Lab
A 3D de-risking lab for yield, MFSP, CI, hydrogen price and carbon intensity, heat recovery, aqueous treatment, catalyst life, feedstock credits, and model reconciliation.
internal pass-all rate against MFSP, CI, catalyst, and fuel-yield gates
top de-risking control by generated score
highest-scoring scenario package for first investment review
First investment screen. Tests the variables most likely to move yield, MFSP, CI, catalyst life, and model credibility at the same time.
Continuous yield and fuel-yield data package Required pilot evidence
The largest economics swing is whether continuous sludge HTL and upgrading actually land inside the biocrude and fuel-yield bands.
Continuous mass balance, carbon balance, oil recovery correction, fuel yield after upgrading, downtime, and sample-to-sample variability.Hydrogen price and carbon-intensity meter Required before deep upgrading claims
Hydrogen price and CI can flip both MFSP and GHG savings, especially for renewable diesel and SAF cases.
kg H2 per kg oil, delivered or on-site H2 price, H2 CI, temporal matching, oxygen co-product credit, and sensitivity to deep HDN.Heat recovery and fouling demonstration Pilot-critical
Heat recovery is central to continuous HTL economics and CI, but fouling can degrade the assumed energy balance.
Recovered heat fraction, fouling drift, cleaning interval, startup fuel, thermal-fluid losses, and net heat duty.Aqueous treatment cost and CI demonstration Required integration gate
Aqueous treatment can either reduce MFSP and GHG through product recovery or add aeration, COD, ammonia, and inhibition penalties.
COD removal, ammonia fate, toxicity, aeration energy, electrocatalytic H2 balance, nutrient byproducts, and WRRF return-load response.1000 h catalyst lifetime and guard-bed run Required deactivation evidence
Catalyst replacement, guard-bed loading, pressure drop, and HDN depth can turn an attractive oil into a costly upgrading feed.
Run hours, pressure drop, guard-bed breakthrough, spent-catalyst autopsy, product nitrogen, hydrogen demand, and downtime.Reduced-order model reconciliation Required review gate
Monte Carlo results are only useful if process, discounted-cash-flow, and LCA models agree on functional units and boundaries.
Model versioning, mass/energy closure, uncertainty distributions, correlation assumptions, LCA boundary, and independent reproduction.The Docker-generated selector picked Continuous yield and fuel-yield data package + Hydrogen price and carbon-intensity meter + Aqueous treatment cost and CI demonstration + Reduced-order model reconciliation + Heat recovery and fouling demonstration + Executive sensitivity dashboard. Pyomo-formulated six-control uncertainty package with yield, hydrogen, aqueous, technical, economic, climate, model, time, and cost gates; solved by deterministic enumeration.
These results are a screening model for research planning. The pass rate is not a commercial-readiness claim until measured sludge assays, continuous HTL runs, hydrotreating data, product analysis, and third-party TEA/LCA close the same gates.
Current readiness decision
The model says the pathway is promising, not ready for commercial fuel claims
The generated screen carries published PNNL uncertainty bands into a site-facing decision model. The leading action is a measured package: continuous yield and fuel-yield data, hydrogen price and CI, aqueous treatment, heat recovery, model reconciliation, and a manager-facing sensitivity dashboard. The pass rate is useful for planning, but it does not replace measured sludge assays, pilot runs, product-quality data, or independent TEA/LCA.
Triangular screen centered on 38.7% finished fuel yield from wet-waste HTL and upgrading.
The internal claim gate is held at $3.45/GGE until site costs and credits are measured.
Median result from 800 generated cases; P50 fuel yield is 38.72%.
First investment screen
63.1Minimum de-risking package
Tests the variables most likely to move yield, MFSP, CI, catalyst life, and model credibility at the same time.
- Economic
- 85.7
- Climate
- 81
- Technical
- 85.3
- Dependency
- 50.3
Pilot readiness focus
61.7Pilot confidence package
Prioritizes measured pilot behavior and site feedstock reality before adding market optionality.
- Economic
- 86.7
- Climate
- 77.7
- Technical
- 81.7
- Dependency
- 52
CI downside focus
58Climate downside package
Tests whether low-carbon claims survive hydrogen CI, heat loss, return-load emissions, and avoided-burden uncertainty.
- Economic
- 78
- Climate
- 85
- Technical
- 64.3
- Dependency
- 61.3
MFSP downside focus
56.8Cost downside package
Checks whether the pathway survives low yield, weak tipping credit, high capex, high hydrogen price, or lower credits.
- Economic
- 86.3
- Climate
- 68
- Technical
- 59.7
- Dependency
- 56.7
Revenue flexibility focus
52Market optionality package
Tests whether the project can switch product slate, co-feed, and logistics strategy without losing margin or CI.
- Economic
- 79
- Climate
- 61.3
- Technical
- 57.7
- Dependency
- 64.7
Generated de-risking package
Pyomo-gated six-control uncertainty package
Continuous yield and fuel-yield data package
Continuous mass balance, carbon balance, oil recovery correction, fuel yield after upgrading, downtime, and sample-to-sample variability.
Hydrogen price and carbon-intensity meter
kg H2 per kg oil, delivered or on-site H2 price, H2 CI, temporal matching, oxygen co-product credit, and sensitivity to deep HDN.
Aqueous treatment cost and CI demonstration
COD removal, ammonia fate, toxicity, aeration energy, electrocatalytic H2 balance, nutrient byproducts, and WRRF return-load response.
Reduced-order model reconciliation
Model versioning, mass/energy closure, uncertainty distributions, correlation assumptions, LCA boundary, and independent reproduction.
Heat recovery and fouling demonstration
Recovered heat fraction, fouling drift, cleaning interval, startup fuel, thermal-fluid losses, and net heat duty.
Executive sensitivity dashboard
P10/P50/P90 metrics, tornado drivers, pass/fail gates, selected de-risking experiments, and source-linked assumptions.
Monte Carlo drivers
Variables most correlated with MFSP and CI in the generated screen
capexFactor
+0.547absolute correlation 0.547
fuelYieldPct
-0.5absolute correlation 0.5
feedstockCreditUsdDryTon
-0.401absolute correlation 0.401
hydrogenPriceUsdKg
+0.349absolute correlation 0.349
ciGCo2eMj
+0.254absolute correlation 0.254
ghgReductionPct
-0.254absolute correlation 0.254
ghgReductionPct
-1absolute correlation 1
hydrogenCiKgCo2eKg
+0.673absolute correlation 0.673
heatRecoveryFraction
-0.527absolute correlation 0.527
fuelYieldPct
-0.316absolute correlation 0.316
mfspUsdGge
+0.254absolute correlation 0.254
aqueousPenaltyFactor
+0.214absolute correlation 0.214
Molecular uncertainty markers
Representative species used to keep the decision model tied to real chemistry
PFOA C7HF13O2
- Score
- 100
- MW
- 364.057
- Hetero
- 15
- F atoms
- 13
Phosphoric acid H3O4P
- Score
- 63
- MW
- 97.994
- Hetero
- 5
- F atoms
- 0
Carbon dioxide CO2
- Score
- 34
- MW
- 44.009
- Hetero
- 2
- F atoms
- 0
Acetic acid C2H4O2
- Score
- 34
- MW
- 60.052
- Hetero
- 2
- F atoms
- 0
Ammonia H3N
- Score
- 31
- MW
- 17.031
- Hetero
- 1
- F atoms
- 0
Pyridine C5H5N
- Score
- 31
- MW
- 79.102
- Hetero
- 1
- F atoms
- 0
Water H2O
- Score
- 23
- MW
- 18.015
- Hetero
- 1
- F atoms
- 0
Phenol C6H6O
- Score
- 23
- MW
- 94.113
- Hetero
- 1
- F atoms
- 0
Dodecane C12H26
- Score
- 15.5
- MW
- 170.34
- Hetero
- 0
- F atoms
- 0
Hydrogen H2
- Score
- 12
- MW
- 2.016
- Hetero
- 0
- F atoms
- 0
Methane CH4
- Score
- 12
- MW
- 16.043
- Hetero
- 0
- F atoms
- 0
Control ledger
Every promising run must survive yield, cost, climate, catalyst, aqueous, and model-boundary uncertainty together
Yield uncertainty66.7
Continuous yield and fuel-yield data package
The largest economics swing is whether continuous sludge HTL and upgrading actually land inside the biocrude and fuel-yield bands.
- Economic
- 96
- Climate
- 74
- Technical
- 92
- Measured
- 88
Hydrogen uncertainty65.5
Hydrogen price and carbon-intensity meter
Hydrogen price and CI can flip both MFSP and GHG savings, especially for renewable diesel and SAF cases.
- Economic
- 86
- Climate
- 98
- Technical
- 68
- Measured
- 90
Aqueous uncertainty65.3
Aqueous treatment cost and CI demonstration
Aqueous treatment can either reduce MFSP and GHG through product recovery or add aeration, COD, ammonia, and inhibition penalties.
- Economic
- 88
- Climate
- 88
- Technical
- 90
- Measured
- 84
Model uncertainty65.2
Reduced-order model reconciliation
Monte Carlo results are only useful if process, discounted-cash-flow, and LCA models agree on functional units and boundaries.
- Economic
- 78
- Climate
- 82
- Technical
- 80
- Measured
- 90
Thermal uncertainty60.2
Heat recovery and fouling demonstration
Heat recovery is central to continuous HTL economics and CI, but fouling can degrade the assumed energy balance.
- Economic
- 82
- Climate
- 84
- Technical
- 86
- Measured
- 86
Decision surface58.6
Executive sensitivity dashboard
Managers need to see which variables make the project pass or fail, not just a single best-case result.
- Economic
- 66
- Climate
- 66
- Technical
- 66
- Measured
- 96
Feedstock uncertainty56.7
Feedstock contract and avoided-disposal baseline
Tipping fees, avoided disposal, counterfactual sludge management, and seasonal sludge quality are often decisive.
- Economic
- 92
- Climate
- 78
- Technical
- 46
- Measured
- 86
Market uncertainty56
Marine, renewable diesel, and SAF market switch
The best product slate can change with hydrogen price, fuel credit, HDN depth, hydrocracking yield, and local offtake.
- Economic
- 78
- Climate
- 70
- Technical
- 70
- Measured
- 84
Upgrading uncertainty55.9
1000 h catalyst lifetime and guard-bed run
Catalyst replacement, guard-bed loading, pressure drop, and HDN depth can turn an attractive oil into a costly upgrading feed.
- Economic
- 84
- Climate
- 60
- Technical
- 96
- Measured
- 82
Co-feed uncertainty49.8
FOG, food waste, and co-feed availability envelope
Co-feeds can improve oil yield but add supply, contamination, contract, logistics, and product-quality uncertainty.
- Economic
- 68
- Climate
- 56
- Technical
- 74
- Measured
- 80
Logistics uncertainty47.4
Sludge, biocrude, H2, and product logistics test
A centralized upgrader, delivered hydrogen, or hauled wet sludge can change MFSP and CI enough to alter the deployment model.
- Economic
- 64
- Climate
- 52
- Technical
- 48
- Measured
- 92
Policy uncertainty47
LCFS, RIN, SAF, and avoided-disposal credit sensitivity
Project economics may depend on credits that vary by product, jurisdiction, CI score, verification method, and policy durability.
- Economic
- 88
- Climate
- 64
- Technical
- 24
- Measured
- 74
Capital uncertainty46.3
Class-3 capex and installation factor estimate
N-th-plant TEA does not include all first-of-kind contingencies, redundancies, startup delays, or site-specific installation costs.
- Economic
- 90
- Climate
- 28
- Technical
- 62
- Measured
- 78
Climate baseline uncertainty44.8
WRRF methane and N2O emissions measurement
Avoided methane and N2O can dominate CI savings but are site-specific and can be fragile under audit.
- Economic
- 46
- Climate
- 94
- Technical
- 38
- Measured
- 74
uncertainty-pnnl-2023
Uncertainty analysis for wet-waste HTL TEA/LCA
Monte Carlo wet-waste HTL source reporting biocrude yield, fuel yield, MFSP, and GHG-reduction ranges.
Open sourceuncertainty-rsc-wet-waste-2026
Wet-waste HTL TEA/LCA with upgrading strategies
2026 source comparing sludge, manure, and food waste across marine, renewable diesel, and SAF product strategies.
Open sourceuncertainty-pnnl-site-2024
Site-specific wet-waste HTL design case
PNNL site-specific design case for wet-waste HTL, biocrude upgrading, economics, and GHG assessment.
Open sourceuncertainty-aqueous-monetization-2024
Aqueous-phase treatment and monetization options
PNNL source showing aqueous-phase treatment choices can reduce MFSP and life-cycle GHG relative to baseline.
Open sourceuncertainty-energy-mss-2026
Municipal sewage-sludge HTL TEA/LCA
2026 municipal sewage-sludge HTL source reporting continuous-mode TEA/LCA, 19 g CO2e/MJ, and sensitivity framing.
Open sourceuncertainty-ml-lca-tea-2025
Machine-learning-aided LCA and TEA for sludge HTL
Integrated ML/LCA/TEA source for optimizing sludge HTL energy, carbon emissions, and economic viability.
Open sourceuncertainty-pnnl-prelim-2016
Preliminary HTL and upgrading TEA for municipal sludge
PNNL preliminary sludge HTL TEA source with feedstock disposal credit and centralized upgrading assumptions.
Open sourceuncertainty-pnnl-sot-2022
Wet Waste HTL and Biocrude Upgrading to Hydrocarbon Fuels: 2022 SOT
PNNL state-of-technology report for wet-waste HTL conversion, upgrading, LCI, energy, and carbon efficiency data.
Open sourceuncertainty-pnnl-electrooxidation-2025
Electrocatalytic oxidation of HTL aqueous phase for H2
PNNL source indicating HTL aqueous COD may supply hydrotreating H2 while treating the aqueous phase.
Open sourceuncertainty-rsc-wwtp-2024
HTL integrated with wastewater treatment plants
RSC source comparing co-located WWTP HTL process configurations, TEA, and LCA assumptions.
Open sourceuncertainty-nrel-lci-2023
Biofuel life-cycle inventory comparison
LCI source for comparing biofuel pathways, aqueous treatment boundaries, and upgraded wet-waste fuel cases.
Open sourceuncertainty-pnnl-lifetime-2021
Extended catalyst lifetime testing for HTL biocrude
PNNL source for long-run hydrotreating and catalyst deactivation uncertainty.
Open sourceuncertainty-catalyst-slurry-2024
Continuous slurry hydrotreating of sewage-sludge HTL biocrude
Pilot-scale slurry/fixed-bed comparison with operability and long-run stability still requiring proof.
Open sourceuncertainty-energy-fuels-tea-2023
Sewage-sludge HTL liquid biofuels TEA
Sewage-sludge HTL TEA source for heat integration, centralized upgrading, and economic context.
Open sourcethe pathway is not investment-ready until uncertainty bands show the same case passes yield, MFSP, CI, catalyst, aqueous, and feedstock gates together