Conferences

Dynamic Material Flow Analysis (dMFA) is a cornerstone method in industrial ecology, enabling systematic, time-dependent tracking of materials through their life cycles and quantifying stock accumulation. Accessible, general and reliable tools are a key facilitator for the modeling capabilities of the community. The open-source library ODYM (Pauliuk and Heeren 2020) has provided for the first time a generalized framework for custom MFA modelling, and is widely used in the community. flodym (https://flodym.readthedocs.io) re-implements ODYM, delivering significantly enhanced capabilities in flexibility, usability, and performance.

Built on the principles of ODYM, flodym simplifies multi-dimensional array operations through the FlodymArray class, which automates dimension management and enables intuitive slicing, manipulation, and computation. This allows users to write more understandable and more adaptable code, minimizing errors, facilitating collaborative model development, and easing future modifications. The library tightly integrates multi-dimensional dynamic stock models – handling inflows, outflows, and stock values over arbitrary time steps – with the overall dMFA structure, achieving both conceptual clarity and computational efficiency.

flodym leverages modern Python tools such as pandas for flexible data import/export and pydantic for robust data validation. It includes built-in visualization routines using Matplotlib and Plotly.

Users can choose between different levels of integration or customization: FlodymArray objects can be used stand-alone to simplify dimension management in calculations, filled with data in numpy array or pandas dataframe format from user-defined readers. Or users can choose to use the MFASystem class, optionally with integrated read-in, export and plotting functionality as well as mass balance and other sanity checks.

flodym is comprehensively documented, featuring how-tos, practical examples, and a complete API reference, making it easy to learn and use. Developed with modern software engineering best practices, it is structured for clarity, maintainability, and scalability. It has a quickly growing external user base, and has been externally used in several summer schools and course teaching events. Contributions and collaborative enhancements are highly encouraged and warmly welcomed.

The presentation will begin with an introduction to the core tools provided by the flodym library, demonstrating how to construct and compute MFA systems, as well as how to effectively use `FlodymArray` objects. These demonstrations will draw on examples from the official documentation. The talk will conclude with a short example of REMIND-MFA, a large-scale, open application built using flodym, highlighting the library’s capabilities in real-world research contexts.

The production of the basic materials cement, plastics and steel is responsible for about 20 % of global greenhouse gas emissions and has grown faster than GDP over the last 20 years. Understanding plausible scenario ranges of future demands, but also future social metabolisms is therefore vital for decision-making on the sector’s climate mitigation strategies. Compared to pure production models, the use of material flow analysis (MFA) allows to include recycling potentials, demand-side measures, and end-of-life impacts in the assessment.

REMIND-MFA, a top-down prospective MFA of the basic materials cement, plastics, and steel was developed to complement the scenario analyses of the REMIND integrated assessment model (IAM). It can be coupled to REMIND, but also run stand-alone or coupled to other modeling frameworks. REMIND-MFA has global coverage with country resolution, or – through the use of a flexible and fully automatized input data generation pipeline – with any subset or regional aggregation thereof. There are several other open MFA models, such as MISO2 [1] or ODYM-RECC [2], but to out best knowledge none of them provides economy-wide future scenarios with global coverage.

Building on a method presented in [3], REMIND-MFA features a historical, production-driven MFA to model historical stock build-up starting in 1900, and using recent data up to the year 2023. It extends this historical data into the future using a novel regression method based on time and on SSP scenario trajectories of population and GDP, which ensures that future material demands are consistent with these scenario narratives. Future material flows up to the year 2100 are derived from these stock projections in a stock-driven MFA.

The model includes trade along several stages of the life cycle. A novel simple model is used to project trade flows into the future based on historical trade volumes, as well as future projections of demand and supply.

The model is fully open-source (https://remind-mfa.readthedocs.io), is based on the flodym (https://flodym.readthedocs.io) library, and carefully modularized. This makes the model transparent to external users, easily extensible and adaptable. Input data preparation is fully automatized, and based on a common structure of data read-in, sanitation, further calculations like extrapolations across missing countries and years, and finally an optional regional aggregation based on a provided country mapping. This allows to reproduce and adapt input data generation, and makes all steps in the preparation process fully transparent. Input data preparation is implemented in R, based on the madrat [4] library and published as an own package under the name mrmfa (https://github.com/pik-piam/mrmfa/).

The talk will start off with a presentation of the core methods of the model. It will present model results, including scenarios of future material demands, end-use stock development, and Sankey diagrams of future metabolisms. A brief introduction for potential users of the model will also be given. The talk concludes with an outlook into future development directions, which include service demand-based demand projections and coupled runs with the REMIND IAM.

Improving plastic waste management and circularity in the European Union (EU) requires analytical approaches that go beyond aggregate perspectives and explicitly capture differences across countries, life-cycle stages, and material compositions. However, existing research on plastics in Europe suffers from several key limitations. Most studies focus either on the EU as an aggregate entity or on individual countries, thereby lacking systematic cross-country comparisons. Given the substantial heterogeneity in economic structures, consumption patterns, and recycling technologies across EU member states, comparative analyses are essential. Moreover, current studies typically focus on total plastics or on specific sectors such as packaging, failing to capture the detailed flows of semi-finished and finished products across the full life cycle. In addition, considering the widespread use and complexity of plastic composites, analyses limited to basic chemical components cannot fully reveal actual material flows or recycling potentials. Therefore, a highly disaggregated, polymer-specific analytical framework is required to support the optimisation of circularity strategies.

In this study, we develop a dynamic, country-level, and polymer-specific Material Flow Analysis (MFA) model for plastics, covering 26 EU countries over the period 2018-2024, and define scenarios to 2050. The model explicitly represents plastic flows across all life-cycle stages, including polymer and fibre production, manufacturing of semi-finished products, finished plastic products and plastic-containing products, waste generation, and end-of-life (EoL) treatment, while accounting for international trade. Using the PRODCOM classification system (NACE Rev. 2) as the organising backbone, we identify over 600 product categories across the plastics value chain. Country-level production, import, and export statistics for these categories are compiled from the Eurostat database and integrated with waste generation and treatment data, supplemented by country-specific information on recycling technologies, to reconstruct a comprehensive life-cycle representation of plastic flows. To enable a detailed polymer-specific assessment, product-level plastic fraction coefficients and disaggregated polymer composition shares are compiled to quantify polymer contents across different product categories. This approach allows plastics embedded in intermediate and final products to be consistently allocated to specific polymer types, including PET, PE, PVC, PP, PS, thermosets, and bio-derived plastics, thereby enabling a complete development of polymer-level material flows analysis across Europe, and providing estimates of polymer specific waste generation.

The key innovation of this study lies in the development of a country-resolved, high-resolution framework for analysing polymer-level material flows and resource efficiency across all main sectors of applications, with a particular emphasis on the detailed characterisation of the EoL stage. Importantly, recycling processes are explicitly differentiated by technology type, allowing for a more detailed representation of secondary material production route and for exploring circular pathways. This high-resolution structure enables, for the first time, a systematic comparison of polymer compositions across all life-cycle stages and across EU member states. By resolving polymer-specific flows at the country level, the framework addresses a critical data gap that has constrained previous assessments of plastics demand, waste management, and recycling potential in Europe. More broadly, it provides a consistent and extendable foundation for the design of targeted, polymer-specific waste management and recycling strategies across the EU and offering policy-relevant insights for the governance of the European plastic circular economy.

Hydrogen may play a key role in decarbonizing various economic sectors. However, due to the limited amount of sustainable resources and the resource-intensive nature of its production, it is crucial to prioritize its applications to ensure it delivers the highest societal benefits. Here, we formulate a novel multi-criteria decision analysis (MCDA) framework that considers nine different indicators, comparing fifteen different hydrogen applications. We apply the MCDA framework to Norway. Our results reveal that ammonia, methanol, and seasonal hydrogen storage exhibit the highest overall performance, whereas hydrogen use in cars, trains, and aircraft should generally be discouraged in the Norwegian context. Future work will extend the analysis to other relevant countries towards a global hydrogen ladder.

Global interconnectedness and the expansion of complex global value chains have recently been at the forefront of climate policy design and formulation within an increasingly tense geopolitical context. Such policies (e.g., EU CBAM) typically intend to both address competitiveness concerns and enhance climate mitigation action; yet these are mostly assessed in macroeconomic-energy modelling frameworks lacking appropriate sectoral and geographic granularity pertaining to disaggregated empirical evidence on trade-climate-industry interactions.
To systematically analyse the interplay among trade, climate/energy, and industrial policies across Europe, five case studies are selected (Greece, Germany, Poland, Sweden, and Netherlands) to represent diverse levels of international trade exposure and various economic and sociopolitical outlooks. In each case study, we conduct a narrative analysis following the Narrative Policy Framework (NPF)—an empirical approach, grounded in political science, public policy, sociology, and narrative theory and philosophy, that systematically diagnoses core policy narrative elements (i.e., setting, characters, plot, moral), strategies (i.e., devil/angel shift, scope containment/expansion, narration tone) and dynamics (i.e., demonstrating shifting, evolving, and interacting relationships between elements and strategies), as portrayed in leading national (broadsheet and tabloid) newspapers. Selected articles encompassing the trade-climate-industry nexus are uniformly coded and compared to reflect national analogies and discrepancies vis-à-vis international media discourses.
Amidst global trade conflicts (i.e., tariffs) and overall protectionism tendencies, EU climate policy is scapegoated for deindustrialisation, rising inflation, and competitiveness blows, contradicting phase-out acceleration efforts—often overshadowing deeper industrial challenges and climate crisis impacts. Conversely, grid connectivity and renewable capacity are overplayed as strategic investment opportunities or national security assets.

The paper presentation primarily focused on the Greek case study findings. 

Incorporating climate action, resource efficiency, and circularity performance within the EU’s industrial transition is a well understood necessity—especially in an environment contested by geopolitical developments and competitiveness concerns. However, the transformations and profound energy and material reconfigurations required towards a coordinated industrial transition are often hampered by divergent regional strategies and potential spatial inequalities. Research in support of these policy processes is often constrained by disciplinary boundaries; notably, energy- and climate-economy models typically used to enable assessments of decarbonisation efforts across multiple industrial value-chains and technologies lack the necessary spatial explicitness and often fail to represent the industrial sector with adequate granularity to address the physical realities and diverse needs of different industrial clusters. Here, we adopt a triangulation approach for informing the industrial low-carbon, circular transition in a transdisciplinary setting that revolves around co-creation and Systems of Innovation perspectives, with the aim to output actionable insights for quantitative systems modelling. Our approach is applied to four representative industrial clusters in Europe. We first establish a stakeholder engagement process with regional and EU actors, to produce key policy- and industry-relevant guiding questions. We then apply socio-technical analysis using integrated frameworks comprising the Multi-Level Perspective and Technological Innovation Systems, to uncover enabling mechanisms for, and hurdles to, the transition. Towards informing place-based scenarios that respond to industrial needs, societal expectations, and climate targets, we highlight aspects that modelling scenarios alone cannot capture without spatiotemporally refined inter- and trans-disciplinary methods, including the role of game-changing disruptions, cross-sectoral cooperation, and industrial symbiosis.

Global cement production grew 250% between 1990 and 2019 and today accounts for 5–8% of annual CO₂ emissions. Chinese demand, which represents half of global production, has likely peaked, but growth in Sub-Saharan Africa and India is expected to continue. Cement is “hard-to-abate”: most emissions arise from limestone calcination, and kilns require temperatures above 1,300 °C. If alternative chemistries remain niche, carbon capture and storage (CCS) may be the only route to near-zero emissions, though technology, cost, and storage remain uncertain. Demand reduction—through clinker substitution, efficient design, or recycling of demolition waste—offers a complementary mitigation pathway as part of Circular Economy (CE).

Quantifying CE impacts on cement demand is challenging. Global material flow analysis (MFA) models often rely on stock-saturation assumptions, limiting their ability to capture levers such as material intensity or sufficiency, and complicating alignment with end-use models like buildings. Service-based MFA models exist but are usually regional or sector-specific; a global framework covering all cement-using sectors is lacking.

We propose an integrated MFA framework combining bottom-up service-demand modelling with top-down estimates. Building stocks are modelled bottom-up but harmonized with top-down totals, while industrial and civil engineering demand is inferred top-down. This approach enables the comprehensive coverage required for integrated assessment models (IAMs) while retaining sufficient detail for CE assessment. We construct scenarios of CE penetration across clinker ratios, material intensity, and sufficiency, producing regional projections for clinker, cement, and concrete/mortar demand. Part of the “REMIND Materials” effort, this framework improves industry representation in the IAM REMIND and supports robust assessment of mitigation options for the EU and beyond.

Steel, cement, and chemical sector account for about 16% of total global anthropogenic GHG emissions. Reducing emissions from these hard-to-abate sectors is crucial for meeting climate goals. While technological solutions such as hydrogen- and bio-based processes or carbon capture and storage (CCS) are indispensable, they face challenges of cost, maturity, and sustainable scalability, raising uncertainties about their near- and long-term feasibility. Circular economy (CE) strategies - such as increasing material efficiency, recycling, and reuse - offer complementary pathways, with advantages such as earlier scalability, economic viability, and additional benefits like reduced resource depletion.

Previous modelling studies often assessed either technological substitution or CE strategies in isolation, but their interactions are critical to capture synergies and trade-offs. We present REMIND Materials, a novel integrated framework linking the integrated assessment model REMIND with a dynamic material flow analysis (REMIND-MFA). Embedded in the macroeconomic development and energy system transformation modelled in REMIND, a material production module optimizes production routes. The MFA tracks material demand, use, and recycling dynamics, feeding back information on demand and secondary material availability. Together, they provide a comprehensive lifecycle perspective on industrial decarbonization, spanning production technologies, demand-side mitigation, and end-of-life strategies.

As a global model with the EU as one of 12 modelled regions, REMIND Materials situates European decarbonization pathways within the global context. This is particularly relevant as most basic materials, industrial products, and energy carriers are traded globally, and cost-optimal mitigation pathways will be influenced by global conditions such as biofuel, hydrogen, and e-fuel availability and prices.

Steel, cement, and chemical products account for about 70% of greenhouse gas emissions from basic material production and 16% of total anthropogenic emissions. Deep emission cuts in these hard-to-abate sectors are essential to meet climate goals. While mitigation options for primary production - such as hydrogen- and bio-based processes or carbon capture and storage (CCS) - are indispensable, they face high costs, low maturity, and uncertain scalability. Therefore, circular economy (CE) approaches - which reduce primary material demands through measures like material substitution, light-weighting, and recycling - are promising complementary alternatives partially due to their scalability, economic viability, and, in some cases, potential for early implementation. They may also deliver co-benefits by mitigating other environmental impacts, including water consumption and pollution.

Previous scenario modeling studies have explored material transition opportunities through two main approaches: technological substitution in production processes, and strategies from CE such as improvements in material efficiency, and enhanced recycling and reuse. To capture interactions between these approaches, collaboration between scenario modeling and industrial ecology has intensified.

We present REMIND Materials, a unified framework coupling the integrated assessment model (IAM) REMIND with the dynamic material flow analysis (MFA) model REMIND-MFA. REMIND links a macroeconomic general equilibrium model with a detailed energy system model, extended here with process-based representations of steel, cement, and chemical production. REMIND-MFA tracks material demand, use, and recycling, allowing explicit modeling of CE strategies such as reuse, recycling, and efficiency improvements.

This integration offers a comprehensive lifecycle view of energy-intensive industries’ pathways to carbon neutrality, capturing production technologies, demand-side measures, and end-of-life strategies. This facilitates not only the investigation of each individual strategy’s impacts and potentials, but also the demonstration of strong synergies and economic interactions between different mitigation options. The approach provides decision-makers with quantitative insights to design industrial transition pathways that align climate targets with economic feasibility.

Steel production contributes 7-9% of global GHG emissions, and demand is projected to grow significantly this century. While decarbonization strategies have focused on fuel switching and CCS in primary steelmaking, these raise production costs and may shift incentives toward recycling and international trade.

This study investigates two dynamics: (A) a shift toward higher-quality scrap-based production, and (B) changes in trade patterns driven by regional differences in renewable energy availability. Using REMIND-MFA – a global, 12-region material flow analysis (MFA) model equipped with economic feedbacks and a price-sensitive trade model – we explore the role of price signals in shaping the steel sector’s future.

Part A focuses on circularity. With growing scrap availability and rising primary production costs, the model evaluates how price elasticities in scrap collection and recycling quality (e.g., copper content) influence secondary steel production. Results show an increasing scrap relevance, although in-use stock growth continues. The model illustrates how increased primary steel prices push to improved scrap collection and high-quality recycling, enabling higher shares of scrap-based steel.

Part B introduces a price-sensitive trade module based on a logit model. It captures how regional cost differentials – especially from varying renewable potentials – reshape steel trade. Preliminary demonstrator results show shifting global flows following a modeled price divergence, illustrating how “renewables pull” alters production geography.

This modelling framework supports ongoing efforts to couple REMIND with detailed material system representations, enabling better integration of circularity and trade into global climate policy analysis.

Policymaking for climate change mitigation is increasingly centred on prescribing energy transition pathways to help governments respond to emerging anthropogenic and climatic crises. These pathways are often shaped by geopolitical pressures and competitiveness concerns, which tend to favour aggregated scenario design approaches within the climate-energy-economy nexus. However, a growing segment of the modelling community is calling for more spatially granular scenarios that can better address real-world challenges within specific sectoral and technological contexts, embedded in divergent regional, national, and transnational political economies.

Calls for spatial explicitness and granular documentation into complex scenario building processes—most often involving inter- and trans- disciplinary approaches—has driven the authors to reflect upon the anthropological potential informing place-based decarbonisation scenarios, their co-created and stakeholder-validated nature, and, eventually, their political substance. Therefore, our reflections will mostly target methodological exchanges between the IAM and the anthropology community, aiming to find common grounds towards more transparent and inclusive research approaches. 

Refineries are essential in converting fossil oil into fuels and plastics and thus highly impacted by carbon pricing. Linear carbon accounting, as used in the EU ETS, incentivizes the switch to climate-friendly fuels but not to climate-friendly plastics. It is unclear how this uneven incentivization affects the coupled production of fuels and plastics. We evaluate the impact of linear carbon accounting compared to a method that incentivizes renewable feedstocks on the cost-effective integrated EU refinery system in 2050. Our findings emphasize the necessity of advanced carbon pricing policies to facilitate sustainable refinery transitions.

Steel, cement, and chemical products, which serve as the foundation of economic development and daily life, account for 70% of annual greenhouse gas emissions of global basic material production and 16% of total anthropogenic emissions. Emission reduction is urgently needed to meet climate targets. However, these sectors are considered hard-to-abate: Mitigation options for primary production of material such as hydrogen- and bio-based processes, or carbon capture and storage (CCS) are indispensable, but face challenges such as high costs, often low technological maturity and sometimes limitations in their sustainable potential. This causes substantial uncertainty with respect to their short-term availability and long-term feasibility. Therefore, circular economy (CE) approaches - which reduce primary material demands through measures like material substitution, light-weighting, and recycling - are promising complementary alternatives partially due to their scalability, economic viability, and, in some cases, potential for early implementation. Moreover, they come with the co-benefit of mitigating other adverse effects of primary material production chains such as water use and pollution. Previous scenario modeling studies have explored material transition opportunities through two main approaches: technological substitution in production processes, and strategies from CE such as improvements in material efficiency and enhanced recycling and reuse. To fully capture these transformation options, as well as their interactions, the two research communities of scenario modeling and industrial ecology have increasingly collaborated in recent years. Here, we present REMIND Materials, a unified approach that couples the integrated assessment model (IAM) REMIND with a dynamic material flow analysis (MFA) framework. REMIND links a macroeconomic general equilibrium model with a bottom-up engineering-based energy system model. REMIND Materials adds process-based modeling of steel, cement, and chemical production. The MFA framework, represented by the in-house SIMSON model, captures demand, use, and recycling dynamics, enabling the representation of circular economy (CE) strategies such as recycling, reuse, and material efficiency improvements. This integration provides a comprehensive lifecycle perspective on energy-intensive industry pathways to carbon neutrality, including production process transformations, demand-side mitigation measures, and end-of-life strategies. This facilitates not only the investigation of each individual strategy’s impacts and potentials, but also the demonstration of strong synergies and economic interactions between different mitigation options. The holistic approach provides decision-makers with critical insights to shape transition pathways that balance climate goals with economic feasibility.

The European Union (EU) agreed to achieve climate neutrality by 2050. The industry sector and particularly the production of the basic materials steel and cement are responsible for a quarter of greenhouse gas (GHG) emissions. To decarbonise the production of steel and cement, new production processes must be established, which are related to various challenges, such as high production costs and import dependencies. Building construction is one of the main demand sectors for these materials. The European Green Deal and the Circular Economy Action plan suggest the role of a circular economy in building construction as contributor to efficient industry decarbonisation. This contribution addresses the following research question: “How can a circular economy in building construction contribute to decarbonising the industry sector in the EU?”.

To answer this question, this contribution uses a multi-method approach. For developing future scenarios of a circular economy in EU building construction, we conducted expert interviews to identify influencing factors and quantitative inputs for modelling prospective material flows. The scenarios were implemented by applying a stock-driven material flow analysis (MFA) covering residential and commercial buildings. The MFA was soft-linked to two established energy system models for buildings and industry, namely Invert/EE-Lab and FORECAST. Invert/EE-Lab provided the prospective building stock development as an input to the MFA and FORECAST simulated changes in industrial energy demand, GHG emissions and costs resulting from changes in the prospective material production determined by the MFA.

Based on the expert interviews, we defined three circular economy scenarios for EU building construction reflecting an increasing implementation depth: (I) Low depth focusing on upstream actions, e.g. increased material efficiency and secondary production; (II) Moderate depth extending to midstream actions, e.g. timber construction and innovative design; (III) High implementation depth including downstream actions, such as efficient use of space and lifetime extension. The highest implementation depth leads by 2050 to a reduction in material production for building construction in the EU of 11 Mt steel, 25 Mt cement and 23 Mt clinker. In comparison to a moderate implementation depth, the additional absolute reduction related to the high implementation is moderate. The circular economy in building construction reduces the amount of captured (process) emissions required for achieving the climate targets by up to 12 Mt CO2eq in 2050. Moreover, it significantly decreases the demand for potentially costly energy carriers like electricity and hydrogen. Consequently, the circular economy in building construction can reduce the cumulated costs for decarbonising the EU industry by around 72 bn€ until 2050.

Our study combines qualitative and quantitative approaches for developing scenarios, which contributes to a holistic understanding of complex systems and their future development. Moreover, we directly integrate material flow and energy system modelling allowing to analyse cross-sectoral implications of decarbonisation and to depict the connection of material flows, energy demand and GHG emissions. In conclusion, our results confirm the prospective role of a circular economy for cost-efficient decarbonisation of the basic material industries in the EU.

Incorporating climate action, resource efficiency, and circularity performance within the EU’s industrial transition is a well understood necessity—especially in an environment contested by geopolitical developments and competitiveness concerns. However, the transformations and profound energy and material reconfigurations required towards a coordinated industrial transition are often hampered by divergent regional strategies and potential spatial inequalities. Research in support of these policy processes is often constrained by disciplinary boundaries; notably, energy- and climate-economy models typically used to enable assessments of decarbonisation efforts across multiple industrial value-chains and technologies lack the necessary spatial explicitness and often fail to represent the industrial sector with adequate granularity to address the physical realities and diverse needs of different industrial clusters. Here, we adopt a triangulation approach for informing the industrial low-carbon, circular transition in a transdisciplinary setting that revolves around co-creation and Systems of Innovation perspectives, with the aim to output actionable insights for quantitative systems modelling. Our approach is applied to four representative industrial clusters in Europe. We first establish a stakeholder engagement process with regional and EU actors, to produce key policy- and industry-relevant guiding questions. We then apply socio-technical analysis using integrated frameworks comprising the Multi-Level Perspective and Technological Innovation Systems, to uncover enabling mechanisms for, and hurdles to, the transition. Towards informing place-based scenarios that respond to industrial needs, societal expectations, and climate targets, we highlight aspects that modelling scenarios alone cannot capture without spatiotemporally refined inter- and trans-disciplinary methods, including the role of game-changing disruptions, cross-sectoral cooperation, and industrial symbiosis.