New Framework Brings Order to the Hidden Uncertainties of Future-Focused Life Cycle Assessment

Every emerging technology carries a promise: that it will be cleaner, greener, and more sustainable than what came before. But verifying that promise before a technology reaches the market is one of the trickiest problems in environmental science. Prospective life cycle assessment, or pLCA, is the tool researchers use to estimate the future environmental impacts of product systems that do not yet exist at scale. Now, a team of researchers from the University of Duisburg-Essen and Leiden University has published a structured, stepwise guidance in the Journal of Industrial Ecology that tackles a long-standing blind spot in these assessments: what happens when a single process produces more than one useful output, and the future changes which of those outputs actually matter.

The problem the researchers address is known as multifunctionality. Many industrial processes are inherently multitaskers. A waste incineration plant does not simply dispose of trash; it may generate electricity and heat, and its leftover bottom ash may contain recoverable metals and minerals. In life cycle assessment, any process with more than one functional flow—whether a product output or a waste input that is being treated—must be handled carefully, because the environmental burdens and benefits of the process need to be distributed among its multiple functions. Conventional LCA has developed a toolkit for this over decades, including allocation methods that partition impacts by mass or economic value, and substitution approaches that credit a coproduct for displacing an alternative product elsewhere in the economy.

What makes pLCA fundamentally harder is time. In a conventional assessment, the analyst can observe the process as it exists today and classify each flow with reasonable confidence. In a prospective assessment, the process may not yet exist at industrial scale, and the economic and regulatory environment around it may shift dramatically before it does. The authors highlight a striking example: waste heat that is considered a nonfunctional byproduct at laboratory scale may become a valuable, marketable product once the process is upscaled and integrated into district heating networks. A flow that is a waste today can become a product tomorrow, and that single reclassification can change whether a process counts as multifunctional at all—and therefore how its impacts are calculated.

Earlier work had already established foundations for dealing with these questions. Guinée and colleagues proposed a four-step approach for identifying and handling multifunctional processes in LCA, moving from distinguishing product and waste flows, to identifying functional flows, to spotting multifunctional processes, to handling them with a chosen procedure. Separately, Langkau and colleagues introduced the SIMPL approach, a stepwise framework for scenario-based inventory modeling in prospective LCA. But the SIMPL approach explicitly excludes the uncertainties tied to multifunctionality decisions, and reviews have found that the identification and handling of multifunctional processes in pLCA studies is often inconsistent, incomplete, or missing altogether. The new guidance is designed to fill precisely that gap by weaving the four-step multifunctionality approach into the three iterative steps of SIMPL.

The resulting framework is built around sub-steps attached to the first and fourth steps of the Guinée approach. In the identification phase, practitioners are guided to identify the external key factors—political, economic, social, technological, environmental, and legal influences—that could change the type of each flow or the number of functional flows in the future. They then develop explicit future assumptions about these changes and integrate them into consistent future scenarios. In the handling phase, the same logic is applied to the parameters that govern how multifunctionality is resolved: the substituted product, the substitution ratio, and the allocation factors. For each handling procedure—system expansion, substitution, physical allocation, or economic allocation—the guidance provides tailored recommendations for identifying what could change and building those changes into scenarios.

Each procedure carries its own distinctive vulnerabilities to the future. System expansion, which redefines the system boundaries to include all functions of the process, is comparatively insensitive to future variability in a non-comparative study, though the authors note that analysts should still verify that goals and scopes align with future scenarios. Substitution, by contrast, is highly sensitive: the substitution ratio, defined as the functionality of the coproduct divided by the functionality of the substituted product, can shift as coproduct quality, input composition, and production technology evolve. Physical allocation depends on the physical relationships between functional flows, which may change through process upscaling, learning curves, and economies of scale. Economic allocation is arguably the most volatile of all, because it rests on relative revenues that fluctuate with commodity prices, patents, technology diffusion, and inflation—yet many published pLCAs apply static revenue ratios that freeze today’s market conditions into a model of tomorrow.

To demonstrate the framework in action, the researchers applied it to a novel process for valorizing incinerator bottom slag, the residual ash left after municipal waste combustion in Germany. Today, most of this slag ends up in landfill construction, but it contains minerals and metals that could be recovered as high-quality secondary raw materials. The process under study uses selective milling followed by magnetic, eddy current, and density separation to yield three fractions: minerals, iron, and non-ferrous metals. The recovered minerals can substitute natural resources in cement production, with the pre-calcinated calcium oxide content reducing carbon dioxide emissions by replacing calcium carbonate. Because the process has only been tested discontinuously at laboratory scale and is unlikely to reach industrial operation before 2030, it is a textbook candidate for prospective assessment.

The team convened a scenario workshop with seven industry representatives spanning waste incineration, slag processing, plant engineering, and cement production to apply the guidance. The workshop revealed how regulatory and quality considerations could flip flow classifications: a minerals stream currently classified as waste under German construction regulations could become a product if metal separation improves or regulations change, while an iron-rich stream’s status depends on whether it is landfilled or used as copper ore. The researchers ultimately built ten foreground scenarios—covering system expansion, three substitution variants, three physical allocation variants, and three economic allocation variants—and combined each with three background scenarios for 2030 generated with the premise tool based on ecoinvent data and the REMIND integrated assessment model, yielding thirty scenario combinations evaluated with the superstructure approach.

The results carry a message that should unsettle anyone who assumes background energy scenarios are the dominant source of uncertainty in future-facing assessments. Within a single foreground scenario, switching between background datasets changed the climate impact by at most 12 kilograms of carbon dioxide equivalents per metric ton of treated slag. But varying the foreground assumptions within the same handling procedure produced far larger swings: 109 kilograms for substitution and 30 kilograms for economic allocation. Most strikingly, in the substitution case, the choice of foreground scenario could reverse the sign of the result entirely, flipping the process from a net carbon sink to a net emitter. In other words, the methodological assumptions about how multifunctionality is handled can matter more than the choice of future energy system.

The authors are careful to note that this finding comes from a single case study and should not be generalized, since sensitivity patterns depend on system characteristics and temporal scope. They also acknowledge practical limitations: the framework depends on expert and stakeholder participation, generates a combinatorial explosion of potential scenarios that requires subjective judgment to prune, and currently keeps the multifunctionality handling procedure fixed across background scenarios. Yet the core contribution stands. By making the invisible choices visible—documenting which assumptions about flow types, substitution ratios, and allocation factors underlie a result—the guidance supports more transparent and reproducible pLCA, particularly for dynamic systems such as energy-intensive electrification projects. For decision-makers weighing investments in emerging green technologies, the study is a reminder that the environmental verdict on the future depends not only on what the world will look like, but on the small, consequential choices analysts make when modeling it.

Subject of Research: A stepwise methodological framework for handling multifunctionality and its uncertainties in prospective life cycle assessment

Article Title: Stepwise guidance for tackling multifunctionality challenges in prospective life cycle assessment

Article References: Zacharopoulos, L., Thonemann, N., Steubing, B., Guinée, J., & Geldermann, J. (2026). Stepwise guidance for tackling multifunctionality challenges in prospective life cycle assessment. Journal of Industrial Ecology, 30(4), 1885-1901. https://doi.org/10.1007/s44498-026-00128-7

Image Credits: AI Generated

DOI: 10.1007/s44498-026-00128-7

Keywords: prospective life cycle assessment, multifunctionality, life cycle assessment, uncertainty analysis, scenario development, substitution, allocation, waste incineration, slag valorization, industrial ecology, circular economy, integrated assessment models