Food waste (FW) remains a major challenge for institutional foodservice systems, undermining resource efficiency and the transition toward more sustainable production and consumption. Evidence linking foodservice system design to FW across both production and consumption stages is limited. This study evaluated the impact of transitioning from a paper-based meal ordering system (PMOS) to an electronic meal ordering system (eMOS) on hospital FW. Standardised before-and-after audits conducted across a full menu cycle quantified kitchen FW (spoilage, preparation waste, unserved food) and plate waste, using leftovers as a proxy for patient consumption, across 2214 PMOS and 2388 eMOS meals. Absolute kitchen FW increased from 134.1 to 161.4 kg/day, reflecting higher meal production following hospital expansion between audit periods. However, kitchen FW per meal produced decreased by 27% (170.4 to 124.0 g/meal), driven by reductions in preparation waste (20.7 to 8.7 g/meal) and unserved food (149.3 to 113.7 g/meal), indicating improved production efficiency and stronger demand-production alignment. Plate waste remained largely unchanged at approximately 0.7 kg/patient/day (41% PMOS; 40% eMOS), with only modest improvements for breakfast and among patients receiving special diets. These findings suggest that eMOS may improve production efficiency by reducing kitchen FW intensity (g/meal) but is unlikely, on its own, to substantially influence plate waste and patient consumption. Additional, complementary strategies may be required to meaningfully reduce plate waste. By evaluating both absolute and efficiency-normalised kitchen FW alongside plate waste, this study provides evidence to inform more resource-efficient hospital foodservice systems and support progress toward SDG 12.3.
Nitrous oxide (N2O) is a long-lived greenhouse gas (GHG) and the most important ozone-depleting substance, making its mitigation a critical global priority. However, long-term evidence on N2O emissions from a consumption-based perspective remains limited. From both production- and consumption-based perspectives, this study assesses global N2O emissions across 164 economies and 120 sectors over 1990–2023 and applies the decoupling model and source-differentiated structural decomposition analysis (SDA) to examine the decoupling of N2O emissions from economic growth and economic drivers. As global N2O emissions increased substantially, the share embodied in trade rose from 19.06% to 28.82%, while the share of South–South trade expanded from 17.16% to 36.31%, with Brazil–China becoming the largest bilateral flow. Fewer economies achieved strong decoupling for consumption-based emissions (27) than for production-based emissions (37). The contribution of demand growth to rising emissions weakened and converged across regions, while Asia and Pacific was the only region in which the mitigating contribution of emission intensity strengthened from −12.36% to −18.94%. Among major developed economies, the intermediate input structure and demand structure effects associated with developing economies shifted from increasing to reducing consumption-based N2O emissions, revealing cross-border mitigation spillovers. After 2018, the demand-structure effect in China's livestock sector was the largest contributor to emission growth, at 306.34 Gg. Our findings provide a scientific basis for targeted climate mitigation policies and international cooperation.
Aquafeeds play a central role in determining the environmental, nutritional, and socio-economic sustainability of aquaculture systems. This review provides an integrated assessment of aquafeed sustainability by synthesising evidence from Life Cycle Assessment (LCA), fish health studies, and socio-economic analysis. A sample of 61 articles was reviewed to identify methodological trends (including functional unit and the most assessed impact category—climate change), evaluate how aquafeed formulations affect environmental performance, and explore the effects of alternative ingredients on fish health and socio-economic performance. Environmental quantitative analysis revealed variability in climate change (CC) impacts, from 0.84 to 21.31 kg CO₂ eq/kg live weight, mainly influenced by aquafeed composition and moderate association with feed conversion ratio (FCR). This suggests that feed efficiency alone cannot explain environmental performance. Socio-economic analysis similarly indicates that FCR and diet cost should not be assessed independently, as the economic profit index showed weak associations with both. Evidence from animal health studies suggests that balanced, digestible nutrient formulations generally support fish growth, health, and feed efficiency. However, these findings depend on ingredient origin, production system, species, inclusion level, digestibility, and LCA methodological choices. Cross-study synthesis highlighted trade-offs among environmental, health and socio-economic dimensions, showing that improvements in one dimension may not ensure overall sustainability. Overall, aquafeed composition and FCR were the variables most consistently linking environmental impacts, fish performance and economic outcomes. These findings support harmonised assessment frameworks integrating environmental, animal health and socio-economic indicators to guide sustainable aquafeed development and inform robust, context-specific decisions across aquaculture production systems.
Agrivoltaic systems, which co-locate photovoltaic energy generation with crop cultivation on the same land area, offer a direct pathway to address the intersecting challenges of renewable energy transition, food production, and acutely scarce freshwater resources. The present study investigated the energy–water–food nexus performance of a 30 kWp elevated bifacial photovoltaic array installed over drip-irrigated Capsicum annuum L. cultivation under a hyper-arid desert climate in Al-Ahsa, Eastern Province, Saudi Arabia (25.4°N, 49.5°E). A paired experimental design compared the agrivoltaic treatment with an open-field control over a full growing season (February–July, 183 days). The bifacial photovoltaic system generated 30,287 kWh of AC electricity, equivalent to a specific growing-season yield of 1010 kWh kWp−1, with a mean performance ratio of 77.4% and a bifacial energy gain of 9.8% relative to a monofacial reference configuration on bare desert soil. The elevated array reduced mean subcanopy air temperature by 2.2 °C and soil temperature by 3.0 °C, whilst elevating relative humidity by 6.8 percentage points and reducing vapour pressure deficit by 0.82 kPa. Seasonal irrigation water consumption in the agrivoltaic treatment was 1053 mm, a 28.7% reduction compared with the open-field control (1478 mm), equivalent to a water saving of 4250 m3 ha−1 season−1. The marketable fresh pepper yield in the agrivoltaic full-irrigation treatment reached 44.1 t ha−1, a decrease of 9.3% relative to the control (48.6 t ha−1). Water use efficiency improved by 27.5% from 4.22 kg m−3 in the control to 5.38 kg m−3 in the agrivoltaic treatment. The land equivalent ratio of 1.75 confirmed that the dual-use system generated 75% more combined land output than equivalent areas of separate photovoltaic and crop installations.
Understanding the emotional value generated by the neighborhood built environment (NBE) during the use stage is essential for the sustainable management of urban service systems. However, existing research still has two limitations. First, the NBE is commonly regarded as a physical spatial context, while its emotional and social performance as an urban service system is often overlooked. Second, most studies assume linear and spatially consistent relationships between the NBE and emotions and seldom distinguish its different associations with positive and negative emotions. To address these gaps, this study conceptualizes the NBE as a spatially embedded neighborhood-scale urban service system and applies an interpretable spatial machine-learning (ISML) framework to investigate its complex associations with location-based emotional expressions in Wuhan, China. The results show that: (1) The ISML framework that combines geographically weighted random forest (GWRF) and SHapley Additive exPlanations (SHAP) outperforms conventional models under both conventional evaluation and spatial block cross-validation. (2) Positive and negative emotions exhibit distinct NBE association structures, with building density and NDVI emerging as the most important factors associated with positive emotion, whereas recreational facility accessibility and building density as the leading factors associated with negative emotion. (3) Most NBE factors display pronounced nonlinear associations, indicating that emotional value does not increase monotonically with service provision, as both insufficient and excessive provision may generate adverse contributions. (4) The associations between the NBE and both emotional dimensions exhibit substantial spatial heterogeneity and core–periphery differences. By extending sustainable production and consumption research to the use-stage emotional value of neighborhood-scale urban services, this study demonstrates that urban service provision should shift from the simple expansion of quantity toward differentiated optimization based on nonlinear responses and local contexts. Such a shift may enhance the emotional and social performance of existing urban service systems.
Circular economy (CE) policies are often presented as gender-neutral, yet a growing body of literature highlights gender inequalities in everyday engagements with the household consumption work the CE transition will entail. In this paper we examine gendered dimensions of CE-related behaviours in Scotland, drawing on a nationally representative survey of 1516 adults. Using the Circular Behaviours Scale, we analysed nine domains of circular behaviours, including aspects of avoiding unnecessary purchasing, reuse and sharing behaviours, repair and recycling. Analysis explored how these behaviours vary by sex, and the extent to which differences between females and males can be explained by environmental concern, working patterns and time availability. Results show that female respondents report higher engagement than men in most circular behaviour domains, with gender gaps greatest among reuse and sharing behaviours, while males report higher avoidance of unnecessary purchases. While some gender differences are partially explained by differences in environmental concern, employment and time pressure, significant gender gaps persist after controlling for these factors, particularly in divestment and sharing behaviours. Furthermore, gender gaps vary depending on household composition, with the presence of children associated with narrower gender gaps in behaviours relating to borrowing and acquiring used items. These findings highlight how gender and household composition intersect to shape circular practices and suggest policies promoting CE engagement risk reinforcing existing inequalities if they fail to account for gendered divisions of labour and consumption work. Embedding gender-sensitivity into CE policymaking is therefore essential as part of a broader just circular economy agenda.
Building materials lacking reliable and accessible information are often classified as waste, thereby limiting their potential for circularity. Although the concept of material passports is gaining traction in the construction sector, there is no globally standardised mechanism defining the specific material data required to support effective circular decision-making. This study addresses this gap by identifying the critical building material data necessary to support decisions on reuse, recycling, repurposing, and recovery. An initial set of 75 data points was identified through a comprehensive literature review and grouped using the Kawakita Jiro Method. Following this, 59 data points were assessed through 28 structured interviews and analysed using a mixed approach. The findings indicate that reuse decisions require the most comprehensive material data, followed by repurposing and recycling, whereas recovery decisions depend on comparatively fewer data inputs. Of the 59 data points, 52 were identified as critical for decision-making on reuse. Drivers for reuse include physical connection details, market demand, intended reuse path, and environmental indicators, with the utilisation data category receiving the closest attention. Repurpose, recycling, and recovery decisions require 36, 26, and 15 data points, respectively. Overall, sustainability-related data strongly influence recycling and repurposing decisions, whereas product information plays a critical role in recovery decisions, as reflected in the number of significant. By defining the essential material data required for circular decision-making, this study provides guidance for stakeholders on the data that must be captured and managed throughout the lifecycle to support reliable and effective circular practices in the construction sector.
This study develops an extended life cycle assessment (LCA) framework to evaluate nine urban bus fuel pathways across resource, environmental, and economic dimensions. The framework integrates LCA with multi-objective optimization to identify optimal fleet configurations under environmental-priority, economic-priority, and balanced-development scenarios. The LCA results show that gas-electric hybrid buses (G-EB) exhibits the lowest life cycle energy consumption, 82.2% lower than that of conventional diesel buses. Battery electric buses (BEB) and green hydrogen fuel cell buses (GHFCB) have the lowest global warming potential, whereas liquefied natural gas buses exhibit the lowest total life cycle cost. The optimization results show that fleet structures are both scenario- and city-specific. Under environmental priority, G-EB and gray hydrogen fuel cell buses (HFCB) dominate in Beijing, G-EB leads in Shanghai, and BEB and G-EB dominate in Guangzhou. Under economic priorities, Beijing favors diesel-electric hybrid buses (D-EB) and BEB; Chongqing favors BEB; Xi'an favors G-EB and D-EB; and Shenyang is dominated by G-EB. The balanced-development scenario yields diversified multi-fuel portfolios shaped by local energy resources, infrastructure, terrain, and climate. These findings extend LCA from pathway comparison to region- and scenario-specific fleet decision support.
Mechanically recycled polyester (rPES) is promoted as a textile-circularity measure, yet repeated mechanical reprocessing can degrade fibre integrity and increase microplastic fibre (MPF) shedding. Existing life cycle assessments (LCAs) commonly treat recycling events independently and rarely characterise MPF emissions. This study develops a transparent parametric multi-cycle LCA framework that couples loop-dependent manufacturing burdens and MPF release with conventional midpoint indicators, MarILCA-based freshwater microplastic physical-effect characterisation and an exploratory circularity indicator. The framework is implemented as an executable model and released as Supplementary Information 2 (SI2), with the complete parameter documentation in Supplementary Information 1 (SI1). The functional unit is 1 kg of fabric-equivalent service over a defined wash-cycle programme; a 30% recycled blend is the baseline and ideal 1:1 substitution is analysed as a bounding case. Manufacturing GWP falls from 3.05 to 2.60 kg CO2-eq kg−1 at loop 1 and then rises to 2.62 and 2.65 kg CO2-eq kg−1 at loops 2 and 3. Cumulative MPF emissions rise from 8.25 to 11.55, 35.47 and 51.15 g kg−1. Loops 2 and 3 are therefore Pareto-dominated by loop 1 on manufacturing climate burden and MPF mass in 100% of 10,000 Monte Carlo iterations within the stated screening ranges. The first loop is a genuine cross-category trade-off: it saves 0.45 kg CO2-eq kg−1 of manufacturing GWP while adding 3.30 g kg−1 of MPF over the baseline wash programme. Using the current MarILCA freshwater PET-microfiber midpoint factor for 10 μm fibres gives an incremental 4.26 × 103 PAF m3 day at the recommended factor, but the published factor spans several orders of magnitude; no cross-category single-score verdict is imposed. Contribution analysis shows that tumble drying accounts for approximately 50% of life-cycle GWP. The framework therefore provides a reproducible, weighting-free test for dominance beyond the first mechanical recycling loop while making the remaining first-loop trade-off explicit.
Decarbonizing the iron and steel industry is critical for sustainable economic development. This study applies Life Cycle Assessment (LCA) to evaluate the environmental and energy implications of four novel carbon mitigation retrofit pathways compared to a conventional blast furnace-basic oxygen furnace (BF-BOF) plant (S0). These scenarios integrate retrofit technologies into existing infrastructure: amine-based carbon capture for storage (S1), Power-to-Gas technology for methanol synthesis (S2), a methanation process for internal gas substitution (S3), or direct hydrogen utilization via blast furnace injection (S4). Results indicate that S1 achieves the highest reduction in global warming potential (32%), constrained by the residual coke requirements of the blast furnace, while fossil resource scarcity decreases by 4% in S2 and S4. Although mineral resource scarcity remains stable across all scenarios, water consumption in electrolyzer-based systems triples, contributing 22% to the direct electrolyzer demand and 78% to the indirect raw materials and energy mix. Non-renewable energy use drops by 26% in scenarios with a high renewable energy integration (S2, S3 and S4). Sensitivity analysis reveals that electricity sourcing is the primary driver of environmental performance; a highly non-renewable European energy mix increases global warming potential, fossil resource scarcity, non-renewable cumulative energy demand and water consumption by up to 115%. Conversely, the 100% wind-powered alternative drastically reduces carbon emissions but slightly increases mineral resource scarcity (4.5%), due to infrastructure requirements. Overall, while Power-to-Gas and hydrogen integration significantly decouple existing steelmaking from fossil fuel dependence, these findings define the realistic boundaries of retrofit decarbonization, demonstrating that low-carbon electricity infrastructure is a strict prerequisite for circular carbon and hydrogen deployment in integrated BF-BOF systems.
Fossil fuel-powered freight trucks substantially increase environmental pollution during the last-mile delivery process. An underground freight transport system (UFTS) has been proposed as a potential environmentally friendly alternative for future deliveries. Consumers' willingness to pay for UFTS services is crucial to the widespread adoption and development of UFTS. Drawing on the Stimulus-Organism-Behavior-Consequence structure and self-determination theory, this paper develops a research model to explore important factors affecting individuals' willingness to pay for UFTS. A dataset comprising 551 qualified survey responses from Singapore is analyzed through structural equation modeling techniques. The results reveal that environmental concern, ascription of responsibility, awareness of green consequences, and social image positively and indirectly affect public willingness to pay for UFTS through pro-environmental values and behavioral intention. Moreover, pro-environmental values indirectly increase pro-environmental behavioral intention through green trust. Collectively, these components have notable explanatory power (49.7%) on public's willingness to pay for UFTS. The analysis of total effects further suggests that social image is the strongest determinant of consumers' willingness to pay for UFTS services. This paper advances current behavioral research by exploring drivers of consumers' willingness to pay for UFTS from a green perspective and provides insights into policy and management strategies in last-mile logistics.
Sustainable industrial manufacturing requires decision-support tools that integrate environmental, economic, and technical considerations within real production systems. However, conventional life cycle assessment (LCA) approaches are often static, evaluate only a limited number of predefined alternatives, and lack integration with enterprise resource planning (ERP) data, restricting their usefulness for operational decision-making. This study develops an ERP-driven modular LCA framework incorporating prospective life cycle modelling, uncertainty analysis, and multi-criteria decision analysis to automatically generate and evaluate technically feasible manufacturing pathways over time.To illustrate this framework, a proof-of-concept study on a specific furniture component that compared 61 material–process pathways and three decision criteria (environmental impacts, production cost, and mechanical performance). Magnesium alloy ZM21 was found to be the strongest candidate, with human health and ecosystem impacts reduced by up to 80%, production costs lowered by more than 60%, and mechanical performance elevated by roughly 50% relative to the aluminium casting baseline. These advantages were sustained across 10,000 Monte Carlo simulations and future prospectives of electricity decarbonization scenarios, demonstrating that the accumulative benefits observed were due to decreased material demand or fewer synthetic routes, rather than solely background assumptions. The results also highlight how lightweight material substitution with low energy manufacturing processes can simultaneously optimize environmental and economic performance. Even though the developed framework has been tested by a single product only, it proposes an approach that can be applied in diverse products and industrial sectors for introducing sustainability assessment to ERP supported manufacturing decision making.
Repair is promoted as a climate strategy for clothing. Yet the consistency of this benefit across garment quality, production impacts, use patterns, and repair logistics remains undertested. We used probabilistic life cycle assessment to compare repairing versus replacing jeans in a Northern European urban context. The model covered 18 service configurations spanning collection models, customer travel modes, and logistics vehicle types. Across paired Monte Carlo draws, repair reduced life cycle climate emissions by 46–52% relative to replacement and outperformed replacement in nearly all scenarios. Baseline garment durability and production footprint, not transport, explained most outcome variation. Transport distances had negligible influence. Even dedicated customer trips by combustion-engine car tolerated median distances of 84–97 km (total travel distance) before repair lost its climate advantage. The largest per-unit benefits arose for short-lived, high-footprint jeans. We define this inverse relationship as the fast fashion repair paradox. The paradox does not imply that fast fashion should be promoted. It shows that current linear products offer the largest immediate climate dividend from repair, while circular-by-design products remain the long-term goal. These findings imply two complementary recommendations. In the current linear economy, policy and market design should expand repair access for high-impact linear garments. For a circular economy, producers should design jeans for durability, repairability, and lower production impacts.
Understanding how carbon footprints differ across households is essential for designing fair and effective carbon pricing mechanisms. However, such assessments can vary depending on whether carbon footprints are allocated based on the monetary value of goods and services purchased or physical quantities consumed, because expenditure differences may reflect not only differences in consumption volumes but also differences in product characteristics, quality, and prices. Here, using micro-level data from China's Household Survey linked with an environmentally extended input–output database, we compare monetary- and physical-based carbon accounting for consumption categories with information on both expenditures and physical quantities and examine the implications for carbon taxation. We find systematic differences in estimated emissions across household groups, ranging from −9% to +7% at the expenditure-quintile level. Differences are larger within individual categories, reaching −54% to +59% for footwear. These differences substantially alter the estimated distribution of carbon tax burdens across household groups. Physical-based accounting indicates higher carbon tax burdens for the poorest quintile, especially for essential goods such as food and residential energy. The comparison demonstrates how accounting choices shape assessments of carbon tax burdens and climate policy fairness.
