When to use what
- GIS and spatial analysis
- Process simulation
- ML and data-driven models
CROSS-CUTTING
Methods & Tools
Which method is fit for this decision—and what can it not tell us?
A cross-cutting teaching area for GIS, process simulation, machine learning, LCA, TEA, optimization, scenarios and uncertainty. It is not a fifth Wang Group research module.
Open lesson →
METHODSDecision analysis
- LCA and TEA
- Optimization and scenarios
- Uncertainty and decision support
CURRICULUM
6 lessons
176 min
01Choosing a method from the decisionBeginner · 24 minStart with the decision, boundary and evidence gap before selecting a familiar model.02GIS and spatial analysisIntermediate · 30 minRepresent where resources, constraints, infrastructure and demand occur—and at what resolution.03Process simulation and balancesIntermediate · 32 minTurn a process concept into connected mass, energy, utility and equipment calculations.04Machine learning for bioeconomy systemsIntermediate · 30 minUse ML when data support prediction or pattern discovery—and keep domain constraints visible.05Scenario and sensitivity analysisIntermediate · 28 minSeparate alternative futures, influential assumptions and uncertain values.06Multi-objective decision supportAdvanced · 32 minExpose feasible trade-offs across cost, carbon, environment and resilience without hiding value choices.
EVIDENCE LIBRARY
Research, models & tools
A curated starting shelf for this teaching area.
TOOLS
Related tools
QGIS / OSGeo
QGIS
An open-source geographic information system for viewing, managing and analysing spatial data.
- Use it when
- Where are resources, constraints, infrastructure and people located?
- Limits
- Spatial precision does not guarantee data accuracy, causality or social acceptance.
Inputs, outputs & scope
- Inputs
- Vector and raster layers, coordinate systems, attributes and analysis rules.
- Outputs
- Maps, spatial metrics, buffers, overlays and exportable datasets.
- Typical applications
- Resource mapping, logistics, site screening, exposure and territorial analysis.
LBNL / JBEI
BioSiting
A geospatial platform for exploring U.S. bioeconomy resources, infrastructure and candidate sites.
- Use it when
- What resources and infrastructure surround a candidate facility location?
- Limits
- A screening map is not a permit, supply contract or deployability decision.
Inputs, outputs & scope
- Inputs
- Location, buffer radius, resource layers, infrastructure and pathway choices.
- Outputs
- Mapped inventories, buffer summaries and downloadable spatial data.
- Typical applications
- Facility screening, resource context and early TEA/LCA framing.
BioSTEAM project
BioSTEAM
An open-source process simulation framework with integrated TEA, LCA and uncertainty workflows.
- Use it when
- How do unit operations combine into material, energy, cost and impact results?
- Limits
- Outputs are only as credible as property methods, scale-up rules and input evidence.
Inputs, outputs & scope
- Inputs
- Thermodynamics, streams, reactions, unit operations, design and economic assumptions.
- Outputs
- Flowsheets, mass and energy balances, equipment sizes, costs and life-cycle indicators.
- Typical applications
- Biorefineries, separation trains, plastic upcycling and early-stage process comparison.
SALib project
SALib
An open-source Python library for global sensitivity analysis, including Sobol, Morris and FAST methods.
- Use it when
- Which uncertain inputs drive model-output variation, interactions or screening priorities?
- Limits
- Method choice, sampling design, model cost and correlated inputs require deliberate treatment; rankings are conditional on specified ranges.
Inputs, outputs & scope
- Inputs
- A defined parameter problem, sampled input sets and corresponding model outputs.
- Outputs
- Method-specific sensitivity indices, confidence intervals and screening rankings.
- Typical applications
- Global sensitivity analysis for process, TEA, LCA and system models.
pymoo project
pymoo
An open-source Python framework for single-, multi- and many-objective optimization, visualization and decision support.
- Use it when
- How can a set of non-dominated alternatives be generated and diagnosed under conflicting objectives?
- Limits
- Algorithm output depends on formulation, scaling and convergence checks; the software does not choose social preferences.
Inputs, outputs & scope
- Inputs
- Decision variables, objective functions, constraints, algorithms and termination criteria.
- Outputs
- Candidate Pareto sets, convergence diagnostics and visualizations for comparison.
- Typical applications
- Technology portfolios, supply networks, process design and other multi-objective system choices.
CASE STUDIES
Examples from Wang Group
2022Outcomes
Novel carbon-negative methane production via integrating anaerobic digestion and pyrolysis of organic fraction of municipal solid waste
Illustrates an integrated organic-waste pathway where process coupling and the counterfactual waste fate determine the carbon claim.
2023Resources
Van Krevelen diagrams based on machine learning visualize feedstock-product relationships in thermal conversion processes
Shows how machine learning and Van Krevelen space can make feedstock–product relationships legible without turning the model into the research question itself.
REFERENCES
References & further reading
- U.S. Department of Energy (2024). Feedstock-Conversion Interface Consortium: Crosscutting Analysis Research.Open source ↗
- International Organization for Standardization (2006). ISO 14040:2006 Environmental management — Life cycle assessment — Principles and framework.Open source ↗
- Hart, Watson and Woodruff (2011). Pyomo: modeling and solving mathematical programs in Python.https://doi.org/10.1007/s12532-011-0026-8 ↗
- Cortes-Peña et al. (2020). BioSTEAM: A Fast and Flexible Platform for the Design, Simulation, and Techno-Economic Analysis of Biorefineries under Uncertainty.https://doi.org/10.1021/acssuschemeng.9b07040 ↗
- Noon and Daly (1996). GIS-based biomass resource assessment with BRAVO.https://doi.org/10.1016/0961-9534(95)00065-8 ↗
- U.S. Department of Energy (2024). 2023 Billion-Ton Report: An Assessment of U.S. Renewable Carbon Resources.https://doi.org/10.23720/BT2023/2316165 ↗
- QGIS / OSGeo (2026). QGIS Documentation.Open source ↗
- Lawrence Berkeley National Laboratory / Joint BioEnergy Institute (2024). BioSiting Tool v2.Open source ↗
- Préat et al. (2020). Identification of microalgae biorefinery scenarios and development of mass and energy balance flowsheets.https://doi.org/10.1016/j.algal.2019.101737 ↗
- BioSTEAM project (2026). BioSTEAM process simulation, TEA and LCA documentation.Open source ↗
- Zhu, Li and Wang (2019). Machine learning prediction of biochar yield and carbon contents in biochar based on biomass characteristics and pyrolysis conditions.https://doi.org/10.1016/j.biortech.2019.121527 ↗
- Gopirajan et al. (2021). Optimization of hydrothermal liquefaction process through machine learning approach: process conditions and oil yield.https://doi.org/10.1007/s13399-020-01233-8 ↗
- Shule Wang, Yiying Wang, Ziyi Shi, Kang Sun, Yuming Wen, Lukasz Niedzwiecki, Ruming Pan, Yongdong Xu, Ilman Nuran Zaini, Katarzyna Jagodzińska, Christian Aragon-Briceno, Chuchu Tang, Thossaporn Onsree, Nakorn Tippayawong, Halina Pawlak-Kruczek, Pär Göran Jönsson, Weihong Yang, Jianchun Jiang, Sibudjing Kawi, Chi-Hwa Wang (2023). Van Krevelen diagrams based on machine learning visualize feedstock-product relationships in thermal conversion processes.https://doi.org/10.1038/s42004-023-01077-z ↗
- Subramanyan et al. (2007). New stochastic simulation capability applied to the GREET model.https://doi.org/10.1065/lca2007.07.354 ↗
- Shule Wang, Yuming Wen, Ziyi Shi, Ilman Nuran Zaini, Pär Göran Jönsson, Weihong Yang (2022). Novel carbon-negative methane production via integrating anaerobic digestion and pyrolysis of organic fraction of municipal solid waste.https://doi.org/10.1016/j.enconman.2021.115042 ↗