Sustainable manufacturing of electrical and electronic equipment (EEE) has become imperative in response to growing concerns over resource depletion, waste generation, and environmental degradation. Among the strategies proposed to address these challenges, remanufacturing has emerged as a particularly effective approach. Remanufacturing refers to an industrial process by which an item is returned to a like-new condition from both a quality and performance perspective.

While several initiatives have emerged to promote repair and product longevity of electronics such as iFixit’s repairability scoring system, the European Union’s Ecodesign for sustainable products regulation (ESPR) on repair and recycling, and sustainability certification schemes like EPEAT, there remains a notable absence of comparable frameworks addressing other circular economy strategies, particularly remanufacturing. While first-party remanufacturing is generally assumed to provide inherent incentives for design-for-remanufacturing, third-party remanufacturing could be supported through design regulations to effectively facilitate remanufacturing. As laptop remanufacturing is predominantly carried out by third-party firms rather than original equipment manufacturers (OEMs), it is the subject of this enquiry.

This study employs mixed methods to examine the design factors that influence the remanufacturing of laptops. Online interviews and correspondence were conducted with Circular Computing, United Kingdom, the largest third-party remanufacturer of laptops in the world. Interviews and correspondence included qualitative inquiries about specific design choices that influence ease of remanufacturing across different laptop models.The company also shared quantitative data sets that recorded disassembly and assembly times for eight different laptop models. These measurements were taken across a sample ranging from one hundred units to twenty thousand units representing four original equipment manufacturers. To validate and expand the data set provided by Circular Computing, a laptop repair specialist in Ireland, assessed the laptop models, documenting the disassembly and assembly process and provided further qualitative and quantitative insights on design challenges. Analytical approaches such as the ease of Disassembly Method (eDIM) focus on predicting disassembly time from design information. In contrast, the present study adopts an empirical time study approach to capture actual disassembly and assembly time behaviour governed by different product designs. Therefore, disassembly and assembly time of the laptops was used as a metric to measure the remanufacturability of the laptops.

The analysis shows that disassembly time across all laptops varies substantially, ranging from 15 to 25 minutes, and the assembly time is longer, ranging from 50 to 75 minutes. It was observed that the keyboard and screen required the most time consistently across all laptop models, highlighting them as the components with the greatest impact on overall remanufacturability. The proposed design regulations include fastening keyboards with metal screws rather than permanent plastic fastening, and avoiding integrated RAM to enable easier and more cost-effective replacement. LCD modules should be secured using metal screws instead of adhesive bonding and increasing screen-to-body ratios and thinner bezels hinder non-destructive disassembly, wider bezels could improve remanufacturing efficiency. These will be further developed into precise design regulations.

Space heating and cooling are dominant contributors to residential energy use and greenhouse gas (GHG) emissions, making the decarbonization of thermal services a critical systems challenge. Industrial ecology and building energy modeling provide high-fidelity tools for evaluating retrofit options through life-cycle assessment and detailed simulation. However, these approaches require extensive data, computational resources, and technical expertise, which limits their applicability for the rapid and transparent comparison of retrofit options at the household decision-making scale. Household adoption is influenced by trust, risk perception, affordability, and access to information; these factors can exacerbate inequities in the transition to clean heating and cooling systems. As a result, households tend to rely on contractor recommendations or calculators with narrow technology coverage, static assumptions, or limited treatment of uncertainty, creating a gap between sustainability assessment methods and early-stage decision-making.

This work presents a physics-based, parsimonious modeling framework developed as a web-deployed decision-support tool for households in New England. The tool enables users to shortlist and compare residential heating and cooling retrofit options, including conventional fossil systems, heat pumps, and hybrid configurations, based on cost and life-cycle GHG emissions over typical system lifetimes. Rather than targeting detailed system sizing or predictive accuracy, the framework is designed to support early-stage technology screening through comparative ranking and interpretable outputs.

The methodology follows a staged, decision-oriented modeling process. First, a reduced-order thermal demand model is developed using a lumped resistance–capacitance network. Model complexity is increased incrementally by adding additional components. At each stage, model performance is evaluated against physics-based reference ResStock archetypes and EnergyPlus simulations to identify a parsimonious representation that balances decision-grade accuracy with low computational burden. Here, parsimony is defined as retaining only those model components that materially influence comparative heating and cooling outcomes. Second, the resulting thermal demand model is parameterized using regionally representative building characteristics. Sensitivity analysis is applied to identify the parameters that most strongly influence heating and cooling demand. The thermal demand model is then coupled with cost and life-cycle–based GHG emissions models to enable comparative evaluation and ranking of heating and cooling technologies. Uncertainty is assessed through scenario-based weather analysis using representative warmest and coldest years to bound interannual climate variability. The modeling framework is implemented in a graphical user interface that supports scenario exploration, allowing users to evaluate how changes in tariffs, incentives, envelope performance, and operating assumptions affect comparative cost and emissions estimations.

Expected outcomes include transparent, uncertainty-aware comparisons of energy demand, operational costs, GHG emissions, and payback metrics suitable for early-stage option screening. To link knowledge with action, we will evaluate the tool with New England residents using a participatory, decision-science approach. Findings will be used in the iterative redesign of the interface and communication of uncertainty.

By aligning model parsimony, validation strategy, and interface design with household decision-making contexts, this work demonstrates how sustainability assessment methods can be translated into computationally efficient, interpretable tools to support residential decarbonization.

Freshwater scarcity is intensifying across the United States, especially in the west, as population growth, expanding agricultural and energy demands, and climate change place increasing pressure on limited water resources in regions such as the Colorado River Basin. Sustainable resource planning and life cycle assessment (LCA) often rely on midpoint indicators to assess water‑related impacts. The Available Water Remaining (AWARE) method is the globally recommended midpoint indicator for water scarcity assessment, but its reliance on global hydrologic models and large basins spanning multiple states limits its applicability for U.S. decision making. AWARE‑US addressed some of these limitations by using measured U.S. Geological Survey (USGS) runoff and water‑use data at the county level. However, both AWARE and AWARE‑US share a critical limitation: neither account for how water consumption in an upstream subbasin reduces the water available to downstream users when basins are subdivided. This omission is particularly consequential in AWARE‑US, where basins are divided into many subbasins. Recent methodological advances in AWARE2.0 address this limitation through a hydrologically meaningful approach to basin subdivision, but the dataset remains global and does not incorporate U.S. observational data.

Central to all AWARE formulations is the metric availability minus demand (AMD), which represents how much renewable freshwater remains for additional users after meeting environmental flow needs. In AWARE‑US, AMD is calculated using only local conditions: the runoff generated within a basin minus local human water consumption and environmental water requirements, divided by the area of that basin. AWARE2.0 instead calculates AMD for each subbasin using the flow measured at the outlet of the furthest downstream‑connected basin and the total area of all downstream basins, ensuring that upstream regions are evaluated within the context of the larger river system they feed. In this study, we develop AWARE2.0‑US, a U.S.‑specific implementation of AWARE2.0 subbasin delineation logic that applies these improved concepts using measured USGS runoff and water‑use data. To represent hydrologic connectivity withing the USGS data, we construct a directed acyclic network of basins linking Hydrologic Unit Codes (HUC)s with National Hydrography Dataset flowlines, retaining only the dominant cross‑boundary flow path for each basin. This simplification removes less than 3% of total inter‑HUC flow volume while preserving major hydrologic connections.

AWARE2.0‑US produces substantially higher characterization factors (CF)s indicating greater scarcity or lower AMD in the headwaters of major river systems, particularly across the upper Colorado River Basin and headwater counties in Wyoming and Montana, where CFs increase by up to 100%. In contrast, semi‑arid upstream regions of the central Great Plains show minimal change because AMD values were already low under AWARE‑US. Downstream basins exhibit the opposite pattern: CFs decrease by up to 40% in the Lower Mississippi and South Atlantic‑Gulf regions due to the inclusion of accumulated upstream inflows in AMD. By combining methodological advances from AWARE2.0 with U.S. observational datasets, AWARE2.0‑US provides a more hydrologically realistic and spatially explicit basis for evaluating freshwater scarcity impacts across the United States. This improved realism strengthens the interpretability of CFs and enhances their relevance for decision making contexts.

With a need to reduce the overreliance on fossil fuel-based plastics, novel biobased and biodegradable plastics have been developed. Among the developed ones, polyhydroxyalkanoates (PHAs) are one of the commonly sought bioplastics due to their reduced impact on the environment and human health. Polyhydroxybutyrate (PHB) is one of the commonly used polymers under the PHA category and this is alluded to the fact that it is sustainable and has wide applicability. Nevertheless, it’s crucial to understand the overall environmental impact of PHB production.

This study investigates the use of forest residue biomass (FRB) and shrub willow (SW) as sustainable feedstocks for Poly(3-Hydroxybuyrate) (PHB) production. Specifically, it aims to evaluate the environmental impact of PHB derived from SW and a SW-FRB mixture. This cradle-to-gate analysis assesses the environmental impact across biomass production, collection, transportation to the biorefinery, and conversion process, that include pretreatment, hydrolysis, fermentation, and product recovery. The inventory analysis is based on results from the process simulations informed by lab-scale experimental data and previous literature. Life cycle impact assessment is performed in OpenLCA using the TRACI 2.1 method. From the analysis, the global warming potential (GWP) obtained for the conventional scenario is lower as compared to that of the green solvent. When the energy source was changed to biomethane, the GWP was reduced by 78%. However, this change also led to increases in certain impact categories. For instance, the freshwater ecotoxicity increased by 88% mainly due to substantial water demand associated with biomethane purification Notably heat and enzymes are the main contributors across most of the impact categories. This is alluded to the fact that heat and enzyme production are very energy intensive processes. Overall, the use of lignocellulosic feedstocks presents a more sustainable alternative to both food crops and fossil resources given that the energy source is renewable and the amount of heat and enzymes used is reduced.

Copper demand is expected to increase substantially due to global electrification and renewable energy expansion, while the average ore grade continues to decline. This study conducts a comprehensive techno-economic analysis (TEA) and life cycle assessment (LCA) comparing the conventional pyrometallurgy process with an emerging alternative, heap bioleaching, across ore grades from 0.6% to 0.2%. Energy and material requirements were modeled using grade-dependent scaling of process inventories, and uncertainty was quantified through pedigree-based and expert-elicited parameters. The results show that heap bioleaching achieves 10–40% lower production costs and 30–50% lower life cycle greenhouse gas (GHG) emissions compared with the pyrometallurgical route, with the advantage increasing as ore grade declines. Environmentally, heap bioleaching substantially reduces global warming and fossil fuel depletion impacts but increases ecotoxicity and eutrophication potentials due to greater tailings mass. These trade-offs emphasize the need for improved tailings treatment and reagent management in future bio-hydrometallurgical systems. Overall, the findings provide quantitative insight into how declining ore grades reshape the environmental and economic competitiveness of alternative copper production routes.

This keynote argues that system-level sustainability decisions in electronics increasingly depend on standardized, portfolio-scale chip-level carbon benchmarks, and that without such benchmarks, common design, sourcing, and policy conclusions about embodied emissions are systematically distorted.

This creates a structural gap in how system-level sustainability in electronics is assessed, particularly in platforms where integrated circuits account for a substantial share of embodied emissions. As a result, designers, purchasers, and policymakers often rely on bespoke life-cycle assessments of a limited number of components or assume semiconductor uniformity, constraining credible comparisons, obscuring intervention points, and introducing persistent uncertainty into upstream value-chain system boundaries.

This work presents a large-scale cradle-to-gate benchmarking study of product carbon footprints for more than 30,000 integrated circuits spanning multiple technology nodes, wafer sizes, manufacturing locations, and packaging types. By treating integrated circuits as first-class sustainability objects rather than opaque inputs, we show that conclusions about dominant emission drivers, sourcing priorities, and design trade-offs can change materially once semiconductor variability is made explicit. The analysis challenges the widespread assumption that chips can be treated as interchangeable components in electronics LCAs without distorting system-level results.

The study combines detailed technical attributes for each integrated circuit with process- and facility-level environmental models spanning wafer fabrication, assembly, and test to estimate cradle-to-gate greenhouse gas emissions in kilograms of CO₂e per die and per packaged part. Product carbon footprints are generated in alignment with ISO 14067 using a defined functional unit, consistent upstream value-chain system boundaries, explicit cut-off criteria, and mass- and area-based allocation of shared process energy. Integrating primary manufacturing data where available with secondary life-cycle inventory datasets enables portfolio-scale comparability while preserving sufficient resolution to reveal decision-relevant differences across nodes, packaging types, and manufacturing geographies. Model development reflects applied use and technical review with multiple leading semiconductor manufacturers and with Carbon Trust Assurance.

Using a flagship smartphone platform as a decision-relevant case study, we propagate chip-level product carbon footprints through bill-of-materials-based assemblies to demonstrate how common assumptions of semiconductor uniformity can reverse conclusions about dominant emission drivers, supplier priorities, and the relative effectiveness of design versus sourcing interventions. The results highlight the roles of fabrication energy and process gases at advanced nodes, the relative contributions of packaging and testing at mature nodes, and the influence of geographic grid mixes and fab efficiencies on overall cradle-to-gate emissions.

This keynote concludes that the absence of portfolio-scale chip-level carbon benchmarks is not merely a data gap, but a structural limitation on credible sustainability decisions in the electronics industry. Without explicit treatment of semiconductor variability, system boundaries are misdrawn, supplier and design trade-offs are misranked, and policy-relevant assessments of embodied emissions risk false confidence. Moving from one-off, bespoke LCAs to data-driven, portfolio-scale product carbon footprints fundamentally changes how system boundaries are defined, how interventions are prioritized, and how decarbonization progress is evaluated. More broadly, the approach offers a transferable sustainability measurement framework for high-impact, high-complexity components, demonstrating how portfolio-scale, component-level carbon benchmarks can restore transparency, clarify system boundary definitions, and materially strengthen decision integrity across diverse global value chains.