Urban Science

Background: To date, there are no known published studies in Portugal that link the perceived neighborhood walkability of a specific geographical area to the functional capacity and social participation of older adults living in the community. Thus, the main objective of this study, conducted in Portugal, is to examine whether the perceived neighborhood walkability is associated with functional capacity, walking confidence, social participation, and exercise self-efficacy among community-dwelling older adults, thereby contributing to the national evidence base on the influence of contextual factors on human functioning, with particular relevance to active and healthy aging. Methods: An exploratory, cross-sectional study was conducted using data from older adults (aged 50 and over) living in the community, in Portugal. A member of the research team assessed physical function using objective measures of upper and lower body function, balance and mobility. To assess the perceived neighborhood walkability, a validated questionnaire for Portuguese adults living in the community with good internal consistency (α = 0.848) was used. The walkability dimensions assessed were safety, facilities/surroundings, comfort/esthetics and accessibility/connectivity. The Pearson correlation coefficient was used to analyze the relationships between continuous variables. Differences between groups were analyzed using independent sample Student’s t-tests or analysis of variance (ANOVA) as appropriate. Results: Perceived neighborhood walkability was associated with several sociodemographic, health, and psychosocial variables. Higher educational levels and better self-perceived health were associated with more positive perceptions of walkability. Participants without urinary incontinence and osteoarthritis reported significantly higher walkability scores, particularly in the safety, facilities/surroundings, and comfort/esthetics dimensions. Older age was associated with higher scores in the facilities/surroundings dimension, whereas slower gait speed and higher anxiety levels were associated with lower perceptions in this dimension. Few significant associations were found between perceived neighborhood walkability and objective physical function measures, and these associations were generally weak. Conclusions: Overall, the findings suggest that neighborhood walkability is a multidimensional construct influenced not only by environmental characteristics but also by health status, psychological factors, and individual perceptions. These results reinforce the importance of considering contextual and environmental factors when promoting mobility, participation, and healthy aging among community-dwelling older adults. Full article

Urban traffic systems are increasingly vulnerable to non-recurrent congestion and abnormal traffic fluctuations, posing significant challenges to intelligent traffic management and resilient transportation operations. Existing traffic anomaly detection methods often struggle to simultaneously characterize heterogeneous anomaly patterns under dynamically evolving traffic states, while severe class imbalance and limited data plausibility further constrain detection reliability. To address these challenges, this study proposes a traffic-prior-guided state-aware framework for robust urban traffic anomaly detection. A Multi-Scale Natural Neighborhood (MS-NaN) module transforms one-dimensional traffic flow sequences into a nine-dimensional representation integrating sequence dynamics, multiscale statistical deviations, and spatiotemporal phase characteristics, thereby embedding traffic state priors into the detection process. Building upon these representations, the Dual-Branch Context-Gated Network (DB-CGNet) separately captures instantaneous traffic disruptions and trend-evolving congestion patterns. An adaptive context-aware gated fusion mechanism then combines the branch features to enhance robustness under complex and non-stationary traffic conditions. To improve evaluation realism, high-fidelity baseline traffic data are generated through B-spline smoothing and first-order autoregressive residual modeling, and anomaly patterns are constructed under Highway Capacity Manual (HCM)-constrained capacity reduction mechanisms. Experiments conducted on a 91-day urban expressway dataset demonstrate that the proposed method achieves the best overall performance among eight benchmark models under a 72 min observation window, attaining an F1-score of 0.7757 and an area under the receiver operating characteristic curve (AUC) of 0.9112. Ablation studies further reveal the critical role of traffic prior features in detecting short-duration evolving anomalies. The proposed framework provides a robust and interpretable solution for intelligent urban traffic monitoring, anomaly warning, and resilient traffic operation management. Full article

The increase in global carbon dioxide emissions is intensifying climate change and severely affecting diverse communities. Addressing these impacts requires sustainable planning approaches that promote environmental and social balance. This research aims to develop a transferable planning methodology incorporating ex ante evaluation for sustainable productive rural settlements, the application of which is examined through a design proposal in Pocollay, Tacna, used as a case study. The methodology comprises six sequential phases: integrated territorial diagnosis, construction of the multicriteria evaluation model, assessment of the baseline scenario, formulation of the planning model, application of the methodology to the case study, and comparative multicriteria evaluation of the projected scenario. As an application of the methodology, a Master Plan and a productive housing module were developed, integrating the proposed planning principles within a circular economy approach, together with bioclimatic design strategies aimed at resource efficiency, passive comfort, water management, renewable energy integration, and reduced environmental pressure. Compared with the baseline scenario, the projected scenario obtained higher potential performance scores across the architectural, environmental, productive, and socioeconomic indicators evaluated. These results represent expected performance according to predefined criteria rather than impacts observed after construction. In conclusion, the main contribution of the research is the proposed planning and evaluation methodology, which can be replicated or adapted in future applications by recalibrating its indicators, weights, climatic parameters, productive systems, material strategies, and community priorities. Full article

Public spaces are essential for urban well-being, social interaction, environmental quality, and territorial equity; however, their presence does not necessarily ensure adequate quality, accessibility, or multifunctionality. Although previous research has emphasized the importance of green and public spaces, fewer studies have integrated multidimensional quality assessment, GIS-based spatial analysis, and multicriteria prioritization in intermediate Latin American cities. This study aimed to evaluate spatial inequalities in the quality of urban public spaces in Loja, Ecuador, through a GIS-based multi-attribute composite index. A total of 183 public spaces were assessed using structured field observation of observable conditions and organized into six dimensions: functional, environmental, cultural, recreational, safety, and economic. The data were cleaned, validated, integrated with spatial layers, and analyzed through descriptive, inferential, multivariate, spatial autocorrelation, and multicriteria procedures. The results showed that public space quality was mainly concentrated at low and medium levels, with functional and safety dimensions playing a stronger role in the global index, while cultural, recreational, and economic components showed weaker performance. Spatial analysis revealed significant territorial differences, local clusters, and spatial outliers. The proposed framework provides a replicable tool to diagnose observable public space quality, identify priority intervention areas, and support evidence-based urban planning in other intermediate cities. Full article

Rapid globalization has caused the architectural and urban homogenization phenomenon, weakening regional characteristics and architectural and urban identity, especially in many historic cities. To address this phenomenon, this paper explores how urban dwellers perceive these processes, offering a user-centered perspective. The research mainly uses a questionnaire-based survey and selects Hangzhou, China, as the research site. It distributed 500 questionnaires and obtained 435 responses, of which 413 valid adult responses were included in the final analysis. Applying both quantitative and qualitative approaches, the research focuses on two phases: the first phase’s results show that the dwellers primarily identify architectural identity through architectural characteristics and daily spatial experiences, while their dissatisfaction with contemporary architecture is mainly based on cultural disconnect, disharmony with the surrounding environment, and formalism. They prefer architecture that can inherit traditional and local characteristics while offering new interpretations and expressions in the contemporary context. Grounded in this questionnaire evidence, the second phase concentrates on translating these perceptions into design insights and approaches, which are as follows: integrate tradition into the design semantics and concepts; reinterpret traditional spatial composition and reasonably insert formal innovations consistent with the local context; incorporate the use of traditional materials and construction; and enhance the connection between the architecture and the surrounding environment and urban tissue. The paper not only expands the discipline and relevant theories but also provides a user-centered, replicable framework for architectural and urban design that strengthens the regionally specific architectural and urban identity to respond to the challenges of homogenization. Full article

Urbanisation and climate change have intensified the Urban Heat Island (UHI) effect in European cities, increasing thermal stress and health risks, especially for vulnerable people. Nature-based Solutions (NbS), such as green infrastructures (GI), offer effective mitigation strategies. A notable example is the Horizon 2020 URBAN GreenUP project in Valladolid, Spain, where a green shade shelter infrastructure was installed and monitored between 2019 and 2022 to evaluate its cooling performance and its effects on temperature-based urban heat indicators through the use of Key Performance Indicators (KPIs). Local thermal conditions were monitored across two adjacent narrow streets—one with the green shade shelter and one as a control. The installation of the green shade shelter in Valladolid produced measurable thermal benefits. During summer peaks, it reduced ambient temperatures by approximately 0.9 °C, with daily maximum temperatures decreasing by 1.0–3.0 °C. In winter, daily minimum temperatures dropped by around 1.0 °C, while autumn saw an increase of 0.7 °C. Summer minimum temperatures varied, ranging from a 0.2 °C decrease in early summer to a 0.7 °C increase in late summer, with a 0.3 °C rise during peak summer in mid-July. The intervention also reduced the frequency and delayed the onset of days exceeding 35 °C and nights above 20 °C, although the delay in peak tropical nights was limited to seven days. These findings indicate that green shade shelters can contribute to improving street-level thermal conditions and reducing temperature-based heat exposure indicators in dense urban environments. Full article

Urban competitiveness has traditionally been associated with infrastructure, economic performance, and innovation capacity, while the role of financial governance has received considerably less attention. This study examines how citizens form perceptions of urban financial integrity and whether such perceptions influence evaluations of urban attractiveness. Drawing on a governance-based framework, the study investigates the relationships among Digital Governance, Financial Transparency, Anti-Fraud Capacity, Financial Integrity, and Urban Attractiveness. Data were collected from 2764 residents of 18 digitally advanced cities across Northern and Western Europe and North America. To enhance methodological rigor, the sample was randomly divided into two equal subsamples and analyzed using Exploratory Factor Analysis, Confirmatory Factor Analysis, and Structural Equation Modeling. The results indicate that Digital Governance is positively associated with both Financial Transparency and Anti-Fraud Capacity, while Financial Transparency and Anti-Fraud Capacity are positively associated with stronger perceptions of Financial Integrity. Financial Integrity emerged as a significant predictor of Urban Attractiveness and served as the principal mechanism through which governance practices influenced broader evaluations of cities. The findings suggest that citizens do not perceive cities as attractive simply because they are digitally advanced; rather, digital governance becomes valuable when it enhances confidence that public financial resources are managed transparently, responsibly, and in the public interest. These results identify Financial Integrity as the key mechanism through which citizens translate governance quality into perceptions of urban attractiveness. Full article

Smart city and smart region governance increasingly relies on data-informed decision-making, stakeholder participation, digital tools, and public feedback. However, these mechanisms do not automatically create institutional learning. This article examines sustainable smart region governance as an institutional learnability problem and asks whether participation, local knowledge, and feedback are converted into adaptive action and retained as institutional memory. The study applies a Preferred Reporting Items for Systematic reviews and Meta-Analyses extension for Scoping Reviews (PRISMA-ScR)-informed structured evidence mapping combined with an embedded regional case diagnosis from the Tokaj Wine Region, Hungary. The analysis integrates the literature on smart governance, learning regions, higher education quality assurance, territorial resilience, and adaptive governance with regional governance observation materials, stakeholder survey data, coding tables, calculation workbooks, and analytical figures deposited in a public Figshare dataset. The results identify the Act/Remember gap as the central learning-cycle disruption. Planning, implementation, monitoring, and consultation may be present, but feedback often fails to become adaptive action, and action is weakly retained as institutional memory. The comparison with higher education quality assurance shows that structured feedback and continuous improvement principles are transferable only as learning logic, not as procedural models. The findings also show that single-profile territories are especially vulnerable to delayed learning, strategic lock-in, and weak community knowledge-building. The article contributes to smart governance research by proposing institutional learnability as a diagnostic capacity of sustainable smart regions. It argues that digital tools should function as learning infrastructure supporting traceability, feedback-to-action mechanisms, and institutional memory, rather than as substitutes for human deliberation, trust, and collective responsibility. Full article

Planning long-term national development under uncertainty is difficult for small states, where indicators are fragmented, frequently revised, and span tightly coupled social, economic, and environmental domains. Scenario tools can support such decisions without committing to a single forecast, but they often lack historical grounding, cross-domain consistency, reproducible update pipelines, and explicit links between policy choices and outcomes. We present the Andorra Scenario Engine, a transparent, data-grounded framework that harmonizes multi-source national indicators into a consistent state vector and propagates four contrasting pathways—Continuity, Overgrowth, Degrowth, and Density—to 2049 through bounded, coupled update rules. The engine belongs to the exploratory-modelling tradition: an upstream, auditable layer toward a national digital twin rather than a forecasting system, whose outputs are internally consistent conditional futures. A central structural modelling choice is to derive population endogenously from GDP growth via a lagged labour-immigration elasticity (0.40 at a one-year lag, 0.10 at two years), grounded in IMF (2025) and World Bank data. A back-cast benchmark quantifies the accuracy cost of this choice (9.5% population RMSE against 1.7% for a linear trend), accepted in exchange for a policy-relevant causal lever. Three macroeconomic parameters and a housing-affordability anchor are calibrated to 2010–2024 official statistics (Nelder–Mead; RMSE = 1.77% on real GDP per capita). Scenario endpoints for 2049 range from 70,049 (Degrowth) to 174,002 (Overgrowth); a sensitivity audit identifies the housing-affordability threshold and the lag-1 elasticity as the load-bearing assumptions on which the headline signals depend. The framework provides a reproducible, human-in-the-loop baseline for comparing policy-relevant trade-offs and binding constraints in data-scarce microstates. Full article

Economic development is multidimensional, yet provincial comparisons in Thailand often rely on individual indicators or broad sustainability indices. This study constructs separate TOPSIS rankings of economic prosperity and distributional outcomes for all 77 Thai provinces. Six prosperity indicators capture economic output, human capital, spatial agglomeration, labor market outcomes, and structural transformation, while four distributional indicators capture income inequality, economic vulnerability, informal employment, and household dependency. The analysis uses 2024 provincial accounts and Labor Force Survey microdata, with comparisons for 2022–2024. Prosperity is concentrated in Bangkok, the surrounding metropolitan region, and the Eastern Seaboard, while Phuket represents a distinct tourism-led pathway. Favorable distributional outcomes are concentrated within the metropolitan–industrial corridor, whereas disadvantage is more prevalent in the North and Northeast. The rankings are associated but conceptually distinct, and the prosperity ranking differs meaningfully from one based solely on GPP per capita. The prosperity ranking is highly stable over time, while the distributional ranking exhibits substantial stability after harmonizing income coverage. Both rankings are robust to alternative normalization procedures, weighting schemes, and aggregation methods. Spatial statistics confirm clustering. The findings demonstrate the value of separating economic prosperity from distributional disadvantage when benchmarking provincial development and assessing urban–regional disparities. Full article

This study aims to examine the role of Relational Stakeholder Management (RSM) on the success of sustainable construction projects in urban areas of Pakistan. The study also examines the mediating role of trust (competence trust, integrity trust, and intuitive trust) and the moderating role of project managers’ leadership skills. Data were collected from 303 professionals working on sustainable construction projects in Pakistan and analyzed using Structural Equation Modeling (SEM). The results show that RSM has a positive and significant direct impact on the success of sustainable construction projects, while competence trust and integrity trust mediate this relationship. However, Intuitive Trust did not mediate the relationship between RSM and project success. Furthermore, the project manager’s leadership competencies strengthen relationships with relational stakeholders and foster trust, thereby improving overall project success. This research highlights that, in addition to technical factors, human and relational aspects are crucial to the success of energy-efficient and low-carbon construction projects in urban areas of Pakistan. The study provides integrated empirical evidence on the interplay between stakeholder relationships, trust, and leadership, specifically in the context of sustainable construction projects in urban areas of Pakistan. The findings of the study provide practical insights to improve RSM, develop trust, and strengthen leadership competencies to enhance success in sustainable construction projects. Full article

Rapid urbanization can shrink and fragment cultivated land while raising agricultural productivity, creating mismatches among spatial configuration, functional performance, and management modes. These dimensions are rarely assessed in a unified, value-neutral framework. We developed a Spatial–Functional–Modal (SFM) framework to evaluate multidimensional synchronicity in 46 county-level units of the Nanjing Metropolitan Area, China, using Period 6 data spanning 2000–2022. We combined 18 entropy-weighted indicators with coupling and coupling coordination degree (CCD) models at county and municipal scales. Regional SFM-CCD rose from 0.304 to 0.327 but remained in the model-defined near-misalignment category, while average coupling declined from 0.747 to 0.740. All pairwise CCDs increased, but coupling trajectories differed: spatial–functional coupling weakened, spatial–modal coupling diverged between Jiangsu and Anhui, and functional–modal coupling strengthened. Anhui’s SFM-CCD increased more consistently and surpassed Jiangsu in later years. Nanjing remained a relative synchronicity trough, reflecting weaker alignment among normalized dimensions rather than lower absolute agricultural capacity. Alternative weights and ±10% and ±20% perturbations largely preserved relative rankings and the overall temporal pattern, but absolute values and categories remained method-sensitive. The framework diagnoses asynchronous transitions and can inform differentiated governance in other rapidly urbanizing regions. Full article

The rapid transition toward electric mobility is widely recognized as a key strategy for improving urban environmental quality. However, while electric vehicles eliminate tailpipe emissions, they continue to contribute to environmental pressures through non-exhaust sources such as tire wear, road surface abrasion, and particle resuspension. This study develops a composite index framework for assessing road-segment-level operational environmental burdens associated with electric traffic, focusing on traffic operations, electric vehicle load characteristics, non-exhaust emission potential, and meteorological dispersion conditions. The framework does not constitute a life-cycle assessment and does not include battery production, electricity-generation mix, or other upstream environmental impacts. The framework combines four dimensions of influence: traffic operations, electric vehicle characteristics, non-exhaust emission processes, and meteorological dispersion conditions. Indicator selection was performed using the Delphi method, while indicator weights were determined through the Analytic Hierarchy Process (AHP). The methodological contribution lies not in the individual methods applied, but in their integration into a road-segment-level assessment framework specifically designed to identify and prioritize environmentally sensitive locations under traffic electrification scenarios. The resulting model incorporates sixteen indicators aggregated into a single environmental impact index that enables the ranking, classification, and prioritization of urban road segments according to their environmental burden. A case study conducted on selected urban streets demonstrates that non-exhaust emission indicators, particularly tire wear and particle resuspension, represent the most influential factors in the assessment process. Within the illustrative five-segment case study, the relative road-segment ranking remained unchanged under the electrified-traffic scenario, while the structure of the assessed burden shifted toward non-exhaust processes. The proposed framework provides a practical decision-support tool for urban planners and transport authorities by enabling the identification of environmentally sensitive locations, prioritization of infrastructure interventions, and support for sustainable mobility strategies in increasingly electrified urban transport systems. Full article

Freely accessible weather APIs are an attractive input for photovoltaic (PV) power prediction, yet how their reference-based error carries through—under temporal aggregation—into day-ahead PV energy is rarely quantified before model development. This study audits eight freely accessible weather-data services—seven forecast services and the Meteostat observational archive—and finds that, for the endpoints, subscription tiers, and collection period evaluated here, only two of the forecast services expose hourly global horizontal irradiance (GHI), while a third advertised irradiance field returns empty—itself a material result for municipal integrators. The two irradiance-capable services are propagated through a common physical PV model of the 548 kWp east–west rooftop plant being built on a university campus in Bratislava, Slovakia, against independent references (CAMS irradiance; NASA POWER temperature and wind). Throughout, “error” denotes discrepancy against the reference, not measurement truth. Over a common 62-day spring–summer window, the day-ahead daily-energy mean absolute error (MAE, relative to mean reference daily energy) is 10–11% for both sources; Open-Meteo’s winter-inclusive own window raises its value to about 13%. On the matched window, Open-Meteo has the lower hourly GHI RMSE (121 vs. 128 W·m⁻²), yet this advantage does not carry through to day-ahead energy—the point-estimate ordering even changes—and neither matched-window difference is statistically resolved. Because daily aggregation rewards low bias over low scatter, this ordering change is already present in daily GHI energy—before the PV conversion—so hourly irradiance accuracy alone does not determine day-ahead energy. Propagated input-data auditing is therefore an advisable step before ML-based PV forecasting and day-ahead scheduling of urban distributed PV. Full article

Smart mobility strategies often emphasize digital information, applications, and data-driven transport management, while giving less attention to the stop-level conditions that shape passengers’ everyday experience. This study develops and applies a Smart Bus Stop Readiness Index (SBSRI) to assess seven interchange bus stops in Jeddah, Saudi Arabia. The index integrates five weighted dimensions: Passenger Information and Digital Readiness; Physical and Thermal Comfort Provision; Pedestrian Accessibility and Universal Design; Safety and Security; and Land-Use and Activity Integration. Data were collected through field audits, spatial mapping, passenger observations, and a short survey of 71 users. The results indicate that the selected stops have operational interchange importance but generally limited readiness. The mean SBSRI score was 39.07/100; under the adopted planning-oriented classification scheme, only Al-Balad Main Station A achieved moderate readiness, while the remaining stops were classified as showing low or very low readiness. Physical and Thermal Comfort Provision was the weakest dimension, particularly in relation to shade, seating, shelter, and shaded waiting areas. Passenger information and pedestrian accessibility also showed substantial deficiencies. Sensitivity analysis indicated that the principal stop rankings remained stable under alternative weighting scenarios, although category labels were more responsive to threshold selection. This study concludes that smart bus stop readiness should be assessed as a socio-technical condition integrating digital systems with climate-responsive waiting provision, pedestrian accessibility, safety, and the surrounding urban context. Full article

This research examines the impact of urbicide on urban memory, using the city of Sarajevo as a case study. Urbicide is defined as the deliberate destruction of the built environment, including both its tangible and intangible aspects. This study proposes a place-based, multilayered methodological framework to analyze the effects of urbicide on the erasure of urban memory among inhabitants, as well as on traumatic and transformed urban memory. The methodological approach proposed and used in this research includes desk research; urban analysis to reveal the material evidence of destruction; sampling to categorize the level of destruction; visual documentation; surveying to measure the impact of urbicide on urban memory; thematic analysis to interpret the results; and mapping to locate both the tangible and intangible aspects of the survey results. The suggested methodological approach is tested through a pilot study in the city of Sarajevo, 31 years after the end of the 1992–95 siege. The results of the pilot survey presented in this paper, although partial due to the sample size, allowed us to explore the feasibility of the method and to implement the necessary improvements to measure the impacts of urbicide on city memory. Furthermore, the pilot study highlighted the need for a holistic, mixed-methods approach to investigate both the tangible and intangible aspects of urbicide, with a specific emphasis on the use of visual questionnaires. Full article

Three-dimensional (3D) building reconstruction plays an important role in urban planning, disaster resilience, and sustainable infrastructure development. Conventional approaches based on airborne Light Detection and Ranging (LiDAR) data can be computationally intensive and often require large labeled datasets, particularly for large-scale applications. This study presents a semi-automated workflow for reconstructing multi-level 3D building models by integrating airborne LiDAR point cloud data with building footprints extracted from National Agriculture Imagery Program (NAIP) imagery using a Mask Region-Based Convolutional Neural Network (Mask R-CNN). The extracted footprints were used to spatially isolate building-specific LiDAR subsets for 3D reconstruction. The proposed methodology generated building models at multiple Levels of Detail (LOD), ranging from two-dimensional (2D) footprints to volumetric representations with detailed roof structures. Building footprint extraction was quantitatively evaluated against LiDAR-derived footprints generated using Density-Based Spatial Clustering of Applications with Noise (DBSCAN), which served as the reference dataset and detected buildings obscured by tree canopy. The reconstruction workflow was implemented in Open3D and incorporated a boundary-aware mesh refinement strategy based on ear-clipping triangulation to improve rooftop continuity in LOD2 models. The proposed footprint extraction framework achieved a mean Intersection over Union (IoU) of 0.8226 relative to LiDAR-derived reference building footprints, indicating reliable building delineation that supports the proposed multi-level 3D building reconstruction workflow. Full article

The management of municipal solid waste (MSW) in a region driven by tourism is a challenge. The dynamics of the waste produced can fluctuate significantly. Islands are particularly vulnerable due to geographic constraints and limited infrastructure. Using data to forecast and evaluate MSW management strategies would help make policy decisions. This research introduces a data-driven framework that integrates tourism dynamics with a predictive baseline for MSW generation and facilitates ongoing evaluation of system performance. The framework comprises four operational zones, namely data acquisition, baseline modelling, deviation analysis, and dashboard-based communication. It also follows the waste stream from upstream waste generation, through midstream collection and transfer, to downstream treatment and disposal. Phuket Island, Thailand, was chosen as a case study. The island has strong tourism-driven waste variability, limited waste management infrastructure, and a clear upstream–midstream–downstream waste-management chain. Tourism was explicitly integrated as an exogenous factor to capture transient demand effects. The amount of MSW was modelled using monthly tourism data. Support Vector Regression was selected to establish the expected waste baseline owing to its high predictive accuracy (MAE = 1143 t, RMSE = 1352 t and MAPE = 3.3%) outperforming other tested models (i.e., SARIMAX, Mean Baseline, Seasonal Naive, Random Forest and Histogram-Based Gradient Boost). Then, it was used to establish the baseline under normal conditions. The residual was then calculated as the difference between the observed values, and the performance of the maintenance system can be further assessed using the deviation analysis. A normal range defined within 3$\sigma$ of the residual percentages from a baseline; values outside this range are flagged as anomalies. The results were integrated into a prototype dashboard to support operational monitoring, policy evaluation, and stakeholder-orientated interpretation. The proposed framework provides a practical basis for an evidence-based adaptive policy in tourism-dependent regions. Future work should incorporate waste composition and spatially explicit data to further improve the MSW management protocol. Considering that tourism demand exhibits seasonal patterns influenced by external factors, the results of this research offer an invaluable tool for waste management. Full article

Machine learning (ML) models like XGBoost have gained traction in housing valuation as they outperform classical hedonic models in predictive accuracy. However, their black box nature has limited their interpretability, until recent advances like SHAP values enabled deeper insights into variable contributions. This study explores whether XGBoost not only predicts housing prices more accurately but can also help identify latent submarket structures, without imposing predefined spatial or socioeconomic boundaries. Using a raw dataset of 7092 housing listings from Barcelona (reduced to 6527 unique listings and finally to 6112 observations after outlier removal), we employ a sequential methodological framework: first, OLS and Spatial Durbin Models identify statistically significant predictors while accounting for spatial dependence; second, an XGBoost model is trained with these validated variables to extract SHAP values quantifying each attribute’s model-attributed contribution to individual predicted prices; third, clustering these SHAP values reveals three distinct submarkets with systematic differences in model-attributed valuation patterns. We validate the resulting segments through five complementary procedures: out-of-sample cluster assignment verifying segment generalizability; nested OLS interaction models that formally test differences in model-attributed valuation patterns across clusters; spatial block bootstrap resampling that assesses cluster stability under geographic subsampling; non-parametric tests against the 2017 cadastral value—an independent administrative proxy for spatial value stratification; and spatial autocorrelation statistics confirming non-random geographic structure. Together, this multi-pronged validation provides convergent evidence that the identified segments capture stable and externally coherent patterns in the model’s learned valuation structure. Unlike prior studies focused solely on prediction, our approach bridges ML’s technical rigor with econometric interpretability. We argue that this capability may stem from XGBoost’s tree-based architecture, which naturally partitions the feature space to accommodate heterogeneous valuation patterns. Future research could extend this framework to more recent tree-based models, further advancing interpretable ML for housing market analysis. Full article