A primer on life cycle assessment
Life cycle assessment (LCA) accounts for the resources used and the emissions released over the life of a product or service: raw material extraction, manufacturing, transport, use and end of life. ISO 14040 and 14044 define four phases: goal and scope, inventory, impact assessment and interpretation. The inventory lists flows such as kilograms of carbon dioxide or megajoules of energy. Impact assessment converts flows into indicators such as global warming potential. Interpretation checks the results against the goal and states their limits.
Approaches to LCA
Process-based LCA builds the product system from unit processes, each with measured inputs and outputs. It is precise about the product but stops where the analyst stops: purchases that are not modeled are left out. In this course the process approach is SimaPro.
Economic input-output LCA (EIO-LCA) uses the input-output accounts of a whole economy. Each sector's purchases from every other sector are known in dollars, so a demand for one sector's output can be traced through every supplier, and the suppliers' suppliers, without a boundary cut. The cost is resolution: every product in a sector shares the sector's average intensities.
Hybrid LCA combines the two. The foreground of the product is modeled with process data; parts with no process data are estimated with EIO intensities from their cost. This is the method for term projects in this course. Record which parts were estimated how, and avoid counting an input twice.
The EIO-LCA method
Let A be the direct requirements matrix: Aij is the dollars of sector i needed to produce one dollar of sector j. For a final demand vector y, total output across the economy is x = (I − A)−1 y = L y, where L is the Leontief inverse. The inverse sums every round of indirect purchases. Let B hold each sector's emissions and resource use per dollar of its output, and C the characterization factors that turn flows into impact indicators. Impacts are then C B L y. The model stores D = C B (impact per dollar of each sector's output) and N = D L (impact per dollar of final demand, including the supply chain). The tool multiplies the column of N for your sector by your dollars.
Assignment 3 Part I builds exactly this with ten sectors: a transactions table, the A matrix, the Leontief inverse, and an emissions vector. The model here has 411 sectors and 23 indicators; the mathematics is the same.
Contributions by sector are the terms of the sum: sector j contributes Dij Ljs to indicator i for one dollar of demand in sector s. The ranked table in the results lists the largest of these terms.
Assumptions and uncertainty
- Linearity. Impacts are proportional to dollars. Twice the demand gives twice the impact; there are no economies of scale.
- Homogeneity. Every product in a sector carries the sector's average intensities. A low-carbon concrete and an ordinary one are the same sector.
- Price as the measure of quantity. Dollars enter the model, so the price basis and price year matter. A purchaser price includes trade and transport margins that belong to other sectors; the tool converts to producer prices with the model's adjusters.
- Vintage. Economic structure is the 2012 U.S. benchmark; emission and resource data are from about the same period. Results describe the 2012 economy, not today's.
- Domestic technology for imports. Imported inputs are represented with U.S. production technology.
- Boundary. Cradle to gate. Use, maintenance and disposal are not included; add them separately.
- Uncertainty. The tool reports point values without ranges. Sector-average intensities can differ from a specific product by a factor of two or more; treat results as screening estimates and say so.
Limitations
- No distinction between products within a sector, and no way to model a design change inside one.
- A single-region model: only U.S. production is represented.
- Impact indicators are characterized potentials (for example kilograms of CO₂ equivalent), not damages.
- The "Energy" indicator counts primary energy of extracted resources and is not the delivered energy that older tools reported. See Documentation.
- Results depend on the price you enter. A wrong price year or basis changes every number in proportion.
Interpreting results
- Totals include the whole supply chain. The supply-chain share tells you how much occurs outside the sector you chose.
- The contributing-sectors table shows where in the chain an impact occurs. A large share in Electricity or Unrefined oil and gas points at energy inputs; a large share in the sector itself points at on-site processes.
- The driving flows under an indicator show which emissions produce it; for example methane and carbon dioxide under Greenhouse gases.
- Compare sectors within one category and one model. Do not compare these numbers with results from a different model year or a different tool without adjusting for definitions.
- Report the model version, the sector, the dollars, the price year and the price basis with every number. The copy button on the tool page produces this block.
References
- Leontief, W. (1970). Environmental repercussions and the economic structure: an input-output approach. Review of Economics and Statistics 52(3), 262–271.
- Hendrickson, C., Horvath, A., Joshi, S., and Lave, L. (1998). Economic input-output models for environmental life-cycle assessment. Environmental Science & Technology 32(7), 184A–191A.
- Hendrickson, C. T., Lave, L. B., and Matthews, H. S. (2006). Environmental Life Cycle Assessment of Goods and Services: An Input-Output Approach. Resources for the Future Press.
- Yang, Y., Ingwersen, W. W., Hawkins, T. R., Srocka, M., and Meyer, D. E. (2017). USEEIO: A new and transparent United States environmentally-extended input-output model. Journal of Cleaner Production 158, 308–318.
- Ingwersen, W. W., Li, M., Young, B., Vendries, J., and Birney, C. (2022). USEEIO v2.0, The US Environmentally-Extended Input-Output Model v2.0. Scientific Data 9, 194.
- ISO 14040:2006 and ISO 14044:2006, Environmental management — Life cycle assessment.