A Multi-Modal Spatial Risk Framework for EV Charging Infrastructure Using Remote Sensing

Oktay Karaku\c{s}; Padraig Corcoran

REVIEW 4 major objections 5 minor 15 references

A Multi-Modal Spatial Risk Framework for EV Charging Infrastructure Using Remote Sensing

T0 review · 4 major / 5 minor · reviewed 2026-08-07 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read RSERI-EV fuses flood, heat, vegetation, grid, and road layers into a composite resilience score for EV chargers, and the Wales prototype finds 88.8% of charging stations exposed to at least one risk factor.

desk verdict A clear, honest exposure-screening prototype for EV chargers in Wales, but the 'Resilience Score' label oversells an unvalidated composite index. read the letter →

arxiv 2506.19860 v1 pith:ZCXPHO6G submitted 2025-06-10 eess.SP cs.CV

Oktay Karakuş , Padraig Corcoran This is my paper

classification eess.SPcs.CV

keywords electricvehiclechargingresilienceassessmentremotesensingfloodrisklandsurfacetemperaturevegetationindicesspatialgraphWalescasestudy

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper proposes RSERI-EV, a spatial risk-screening framework that assigns each electric-vehicle charging station a composite resilience score from six open-data layers: flood-zone overlap, land-surface temperature extremes, vegetation health (NDVI), land-use/land-cover class, distance to the nearest electrical substation, and proximity to major roads. The central claim is that this interpretable, multi-modal fusion detects environmental and infrastructural vulnerability that placement studies based on demand alone miss. On a Welsh case study of 920 active chargers, the framework finds that 88.8% face at least one risk factor, that heat (72.5%) and vegetation (52.5%) stress dominate, and that the highest composite scores cluster in the South Wales corridor. A spatial k-nearest-neighbour graph over charger locations allows neighbourhood-level reading of the scores, though the paper treats this graph component as exploratory. If the framework holds up, it offers planners a reproducible first-pass screening of where existing or planned charging infrastructure is most exposed to climate and grid stress.

What carries the argument

The load-bearing object is the RSERI-EV composite score, a normalised, equal-weighted sum of six binary risk indicators: flood-zone overlap; land-surface temperature above the regional 90th percentile; low NDVI combined with urban or coastal land-cover class; more than 5 km from a substation; and limited proximity to major roads. Normalisation puts the continuous inputs on a common scale, equal weighting forms the baseline score, and principal-component weighting is offered as a sensitivity check. A spatial $k$-nearest-neighbour graph with $k=5$ connects each charger to its nearest neighbours by Euclidean distance, giving the framework a topology-aware view of risk clustering; in this prototype the graph is descriptive rather than used for propagation or optimisation.

What would settle it

Audit charger outage and restoration records in Wales during recent flood and heat events and test whether chargers flagged by RSERI-EV fail more often, stay down longer, or suffer more physical damage than unflagged chargers in the same period. If the 817 flagged chargers show no higher outage rate or downtime than the 103 unflagged ones, the score's central ranking claim collapses.

Watch

Extended reading notes

Core claim

The paper's central discovery is a prototype method, not a new scientific law: by layering remotely sensed heat and vegetation stress on top of flood maps and grid and road proximity, RSERI-EV produces a single, interpretable resilience score for every charging station. In Wales the method reports that 817 of 920 active chargers (88.8%) carry at least one risk flag; high surface temperature and low vegetation health account for most of the signal, while substation distance and road-access risk are rare (1.8% and 7.8%). The composite scores are not uniform: Neath Port Talbot and Caerphilly lead the local-authority ranking, rural Gwynedd and Pembrokeshire sit at the bottom, and overlap analysis shows that flood, heat, and vegetation risks co-occur at 66 stations while grid and road risks almost never coincide. The paper presents this as evidence that multi-modal fusion captures vulnerabilities a single-layer analysis would miss and as a first step toward predictive resilience modelling.

Load-bearing premise

The score's ability to rank resilience rests on the assumption that the chosen proxies and their binary thresholds (flood-zone membership, a 90th-percentile heat cutoff, low vegetation, urban or coastal land cover, a 5 km substation distance, and limited road access) actually track a charger's vulnerability to failure or slow recovery; the paper does not validate the score against observed outages or damage.

Editorial extensions

If this is right

At least 88.8% of Wales's active charging network sits in a zone flagged for one or more risk factors, meaning most chargers are exposed to some stress even before a specific event.
Heat and vegetation stress are the dominant drivers (72.5% and 52.5%), so region-level resilience planning that ignores thermal and land-cover conditions would miss most exposed chargers.
Flood, heat, and vegetation risks coincide at 66 chargers (7.2% of the network), a small but important set of multi-hazard sites that may deserve priority reinforcement.
Grid and road proximity risk are rare and nearly independent (joint rate 0.9%), so their contribution to the composite is localised and cannot substitute for heat and flood screening.
Because the framework uses only open data and a transparent scoring rule, the same protocol can be re-run in other regions or time periods as soon as the equivalent layers are available.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

We infer the score is best read as a screening exposure index rather than a measured vulnerability: until the binary thresholds are calibrated against outage data, rankings could shift if a threshold such as the 5 km substation distance or the 90th-percentile heat cutoff were varied.
A natural extension is to run RSERI-EV under future climate projections rather than current conditions; that would turn its static heat and flood layers into a forward-looking siting tool, which the paper does not do.
The k-nearest-neighbour graph is currently descriptive, but it could propagate risk labels to chargers with missing data or rank reinforcements by network centrality; those are uses the paper leaves implicit.
A testable extension of the composite logic would be to weight indicators by local evidence of harm, such as flood return period instead of any flood-zone overlap, which would likely sharpen the South Wales findings.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

Paul,

Quick take: this paper delivers a clear, reproducible-in-principle screen of environmental and infrastructural exposure for EV charging stations in Wales, and that is worth something. But the authors call the output a 'Resilience Score' when they have no outcome data linking the proxies to actual charger failure or recovery. The paper is an application, not a methodological advance.

What's new: the Wales-wide multi-layer ranking of individual chargers from open remote sensing and OSM data is a new artifact. The combination of flood zones, LST, NDVI, substation distance, and road proximity into a single index, with a kNN graph for local context, is a sensible modular pipeline. The descriptive statistics (72.5% LST risk, 52.5% vegetation risk, 88.8% at least one risk) are internally consistent, and the maps look like they follow from the stated rules.

What's soft: the central claim that RSERI measures resilience is unsupported. The flags are exposure proxies, not validated failure mechanisms. No observed outages, flood damage, or grid failures are used to test the score. The 'Vegetation Risk' flag is defined as low NDVI with urban/vegetation LULC, while 'LULC Risk' already counts urban/coastal as high risk. So the dominance of LST and vegetation flags, and the high scores in South Wales urban corridors, are partly baked into the input definitions. Several thresholds are not stated (what is 'low NDVI'? what counts as 'limited road accessibility'?). The PCA sensitivity check is mentioned but never shown. No code or data is shipped.

In fairness, the authors explicitly call it a prototype and acknowledge the binary simplification and the exploratory graph. So the gap is not hidden. It just means the title's 'spatial risk framework' is accurate, but the 'Resilience Score' label overpromises. This is an exposure index until validated.

Who this is for: municipal planners and researchers who want a quick, interpretable screening of where chargers face heat, flood, or vegetation stress. They should treat it as a first-pass sieve, not an assessment of operational resilience.

Recommendation for peer review: I would send it to a serious referee, but with a clear request to either (a) rename the output 'exposure score', or (b) add validation against outage or flood event data. The current version is a conditional paper: the application is plausible and the writing is honest, but the core construct needs stricter calibration. I would expect major revision.

Referee Report

4 major / 5 minor

Summary. The paper proposes RSERI-EV, a prototype spatial risk framework for electric vehicle (EV) charging infrastructure. It combines remote sensing layers (flood zones, Landsat 8 land surface temperature, Sentinel-2 NDVI, land use/land cover), infrastructure data (substation proximity, road proximity), and a spatial k-nearest-neighbour graph over charger locations to compute a composite 'Resilience Score' for each charger. The framework is demonstrated on a public EV charger dataset for Wales, reporting that 88.8% of chargers face at least one risk factor, with LST (72.5%) and vegetation (52.5%) being the most prevalent, and that high composite scores concentrate in South Wales districts such as Neath Port Talbot and Caerphilly.

Significance. If the central claim were fully supported, the paper would make a useful, low-cost contribution to climate-resilient EV infrastructure planning: it fuses openly available Earth observation and infrastructure data into a scalable, interpretable screening tool, and its modular pipeline and spatial-graph scaffolding are sensible foundations for future work. The paper is honest about being a prototype and provides descriptive statistics, maps, and a reproducible data-processing path. However, its significance is currently limited by the gap between the 'resilience' label and what is actually measured: the composite score is an exposure index built from binary proxies that are never validated against observed charger failures, flood damage, or grid outages. The framework's potential is clear, but the paper needs to either substantially validate the score or carefully reframe its claims.

major comments (4)

[§3.2 and §4 (Table 1)] The claim that LST and vegetation dominate the risk profile is partly definitional. 'Vegetation Risk' is defined in §3.2 as low NDVI combined with Urban or Vegetation LULC, while 'LULC Risk' already flags urban/coastal chargers as high risk. Because urban land cover is a component of both the vegetation flag and the LULC flag, the reported 52.5% 'Vegetation Risk' overlaps with the urban classification by construction. The dominance finding in Table 1 and the subsequent regional interpretation in §5 therefore do not demonstrate an independent vegetation-health signal. Please report the NDVI-only flag separate from the LULC-conjunction flag, or remove the LULC condition from the vegetation definition.
[§3.2 and §3.4] Several thresholds are not specified, which makes the headline figures (88.8% at-least-one-risk, Table 1) and all RSERI scores non-reproducible. The text states that LST high risk means exceeding the regional 90th percentile, and grid risk means more than 5 km from a substation, but it does not define 'low NDVI', 'limited accessibility to major roads', or the specific LULC classes and data product used (e.g., Corine, ESA WorldCover, or a local layer). These parameters are load-bearing because changing them would change every reported percentage and the LAD ranking in Figure 4. Please state all thresholds explicitly and, ideally, include a sensitivity analysis over these thresholds.
[§3.4 and §4] The central resilience claim is not validated against any observed outcome. The RSERI-EV score is never compared to actual charger outages, flood damage, heat-related failures, or grid service interruptions, despite the literature cited in §2 (e.g., flood resilience of charging infrastructure). In its current form, the score is an exposure index constructed from proxies; calling it a 'Resilience Score' implies a relation to operational resilience that is not demonstrated. To support the paper's main assertion, please either (a) validate against an outcome dataset for a subset of chargers, or (b) explicitly rename the composite to an 'exposure index' and scale back the resilience-language throughout the abstract, §1, §5, and §6.
[§3.4] The handling of missing data is under-reported. The text says only that 'Stations with incomplete data were excluded from the analysis,' but it does not state how many stations were excluded or what 'incomplete' means for each layer. If, for example, all excluded stations lie in rural areas, the reported risk percentages and the 'rural districts low risk' conclusion in §5 could be biased. Please report the number and spatial distribution of excluded stations and discuss how exclusion might affect the findings.

minor comments (5)

[Keywords] The keyword 'Resillience' is misspelled; it should be 'Resilience'.
[§2] The literature review would benefit from distinguishing more clearly between studies of charger placement/optimisation and studies of operational resilience under stressors; currently the two categories are blended, which weakens the positioning of the new contribution.
[§4 (Figures 1-4)] The figures are informative but some are small and the map legends are not fully legible in the printed version; please enlarge the maps and ensure that colour-blind-safe palettes are used for the risk classes.
[§5] The discussion mentions that PCA-based weighting was 'explored for sensitivity analysis' (§3.4), but no results of this sensitivity analysis are presented in §4 or §5. Either report the PCA-weighted results briefly or remove the claim.
[§3.3] The kNN graph construction states k=5, but Section 5 says the graph's potential is 'not yet fully explored'; consider adding a short demonstration of a graph-based diagnostic (e.g., spatial autocorrelation of risk) to justify the graph construction, or explicitly label it as scaffolding for future work.

Circularity Check

0 steps flagged · score 0.0 of 10

No circular derivation: RSERI-EV is a descriptive composite index whose reported frequencies and rankings are arithmetic restatements of its input layers, not predictions of an external outcome.

full rationale

The paper defines each risk indicator in Section 3.2 (flood zone overlap, LST above the regional 90th percentile, substation distance greater than 5 km, limited road access, low NDVI, urban/coastal LULC, and the combined low-NDVI/urban-or-vegetation flag) and then computes the RSERI score in Section 3.4 as a normalized, equally weighted aggregation of these same indicators. The headline results in Section 4—72.5% LST risk, 52.5% vegetation risk, 88.8% with at least one risk, and high average scores in South Wales LADs—are direct tabulations of the indicator definitions rather than outputs of a model that reuses its own fitted target. The paper makes no out-of-sample prediction, fits no parameters to an outcome, and does not claim to infer resilience from data independent of the score. It explicitly frames the work as a prototype screening framework and acknowledges limitations including binary thresholds, unexplored graph reasoning, and scope limited to Wales. No equation equates a fitted quantity to a predicted quantity, and no load-bearing self-citation appears; the reference list contains no works authored by the present authors. The only substantive concern is construct validity—whether the proxy indicators actually capture operational resilience of chargers under stress—but that is an empirical validation gap, not a circular derivation. Accordingly, no circular step meeting the evidentiary standard of the review can be identified.

Assumptions & free parameters 7 free parameters · 6 assumptions · 1 invented entities

The central claim rests on a chain of data-proxy and threshold assumptions: open datasets are reliable, Euclidean distance captures accessibility, the chosen hazard and vegetation indicators map to charger resilience, and binary cuts preserve the signal. The RSERI score itself is defined as the normalized sum of these indicators, so no independent validation is provided.

free parameters (7)

LST high-risk threshold = regional 90th percentile; flags 667/920 (72.5%)
Chosen as a percentile cut, but the high flag rate shows the threshold dominates the result; no sensitivity analysis is reported.
Grid distance threshold = 5 km; flags 17/920 (1.8%)
No justification is given for 5 km as the critical substation distance.
Road accessibility threshold = unspecified; flags 72/920 (7.8%)
The paper does not state the distance or road-class rule used to define limited accessibility.
NDVI low threshold = unspecified; flags 483/920 (52.5%)
No threshold value for low NDVI is reported, yet this indicator drives half of the stations.
LULC high-risk classes = urban and coastal classes
Land cover categories are arbitrarily assumed to imply higher environmental stress.
k for kNN graph = 5
Chosen without sensitivity analysis; graph use is exploratory.
Risk indicator weights = equal weights (PCA variant not shown)
Equal weighting is an arbitrary modeling choice, and the PCA robustness check is mentioned but not presented.

assumptions (6)

domain assumption Euclidean distance in EPSG:27700 approximates meaningful substation and road proximity
Used in Section 3.1 to compute proximity; network distance or routing could change classifications.
domain assumption OpenChargeMap, OpenStreetMap, and EPW flood data are complete and accurate enough for inference
Section 3.1 treats open datasets as ground truth without completeness checks.
ad hoc to paper The chosen environmental proxies indicate charger operational resilience
Section 3.2 maps LST, NDVI, flood, and land cover to risk without validation against outages or failures.
domain assumption Stations with missing data can be excluded without biasing the analysis
Section 3.4 excludes incomplete stations but reports no count or spatial pattern of excluded cases.
ad hoc to paper Binary thresholding preserves the relevant risk signal
Section 3.2 converts continuous indicators to binaries; the authors acknowledge this simplification in Section 5.
standard math Standard statistical tools (Pearson correlation, PCA, KDE) behave appropriately on these spatial samples
Used in Section 4 without distributional or spatial-autocorrelation checks.

invented entities (1)

RSERI composite Resilience Score
purpose: Rank EV charging stations by multi-hazard exposure from weighted binary indicators
No external outcome (outage record, flood damage, grid failure) is used to show the score measures resilience; it is a normalized sum of the inputs by construction.

how reviews work

0 comments

Cite this review

Pith. "Pith review of A Multi-Modal Spatial Risk Framework for EV Charging Infrastructure Using Remote Sensing." pith.science (2026). https://pith.science/paper/ZCXPHO6G

@misc{pith2026250619860,
  author       = {Pith},
  title        = {Pith review of: A Multi-Modal Spatial Risk Framework for EV Charging Infrastructure Using Remote Sensing},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/ZCXPHO6G}},
  note         = {Machine review of arXiv:2506.19860}
}

abstract

Electric vehicle (EV) charging infrastructure is increasingly critical to sustainable transport systems, yet its resilience under environmental and infrastructural stress remains underexplored. In this paper, we introduce RSERI-EV, a spatially explicit and multi-modal risk assessment framework that combines remote sensing data, open infrastructure datasets, and spatial graph analytics to evaluate the vulnerability of EV charging stations. RSERI-EV integrates diverse data layers, including flood risk maps, land surface temperature (LST) extremes, vegetation indices (NDVI), land use/land cover (LULC), proximity to electrical substations, and road accessibility to generate a composite Resilience Score. We apply this framework to the country of Wales EV charger dataset to demonstrate its feasibility. A spatial $k$-nearest neighbours ($k$NN) graph is constructed over the charging network to enable neighbourhood-based comparisons and graph-aware diagnostics. Our prototype highlights the value of multi-source data fusion and interpretable spatial reasoning in supporting climate-resilient, infrastructure-aware EV deployment.

Figures

Figures reproduced from arXiv: 2506.19860 by the authors.

**Figure 2.** (Left) Distribution of EV charging stations by number of concurrent risk factors. [PITH_FULL_IMAGE:figures/full_fig_p009_2.png] view at source ↗

**Figure 3.** (a) Spatial distribution of average RSERI scores across Wales visualized using a hexag [PITH_FULL_IMAGE:figures/full_fig_p010_3.png] view at source ↗

Figures from the paper (1 more)

**Figure 4.** Figure 4: RSERI scores across Wales: (a) Top and bottom 5 LADs by average score, bar colour [PITH_FULL_IMAGE:figures/full_fig_p010_4.png]

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

15 extracted references · 15 canonical work pages

[1]

Sustainability 14(20), 13295 (2022) 10 O

Almutairi, A.: Impact assessment of diverse ev charging infrastructures on overall service reliability. Sustainability 14(20), 13295 (2022) 10 O. Karaku¸ s & P. Corcoran © Carto © OpenStreetMap contributors0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 Avg RSERI Score (a) RSERI Score Distribution © Carto © OpenStreetMap contributors RSERI Class Low Moderate High (b) R...

work page 2022
[2]

Environmental Science and Pollution Research30, 68054– 68083 (2023)

Banegas, J., Mamkhezri, J.: A systematic review of geographic information systems based methods and criteria used for electric vehicle charging station site selection. Environmental Science and Pollution Research30, 68054– 68083 (2023)

work page 2023
[3]

Energy (2025) A Multi-Modal Spatial Risk Framework for EV Charging Infrastructure Using Remote Sensing 11

Batic, M., Breckner, L., Weibelzahl, M.: Geodemographic aware electric vehicle charging location planning for equitable infrastructure deployment. Energy (2025) A Multi-Modal Spatial Risk Framework for EV Charging Infrastructure Using Remote Sensing 11

work page 2025
[4]

Heliyon10(7) (2024)

C ¸ elik, S., Ok, S ¸.: Electric vehicle charging stations: Model, algorithm, simulation, location, and capacity planning. Heliyon10(7) (2024)

work page 2024
[5]

Transportation Research Part D: Transport and Environment139, 104580 (2025)

Halder, B.: Remote sensing-based optimal electric vehicle charging station identification in kolkata urban agglom- eration. Transportation Research Part D: Transport and Environment139, 104580 (2025)

work page 2025
[6]

IEEE Geoscience and remote sensing magazine5(1), 53–65 (2017)

Harb, M.M., Dell’Acqua, F.: Remote sensing in multirisk assessment: Improving disaster preparedness. IEEE Geoscience and remote sensing magazine5(1), 53–65 (2017)

work page 2017
[7]

Renewable and sustainable energy reviews154, 111733 (2022)

LaMonaca, S., Ryan, L.: The state of play in electric vehicle charging services–a review of infrastructure provision, players, and policies. Renewable and sustainable energy reviews154, 111733 (2022)

work page 2022
[8]

Deep Learning in Multimodal Remote Sensing Data Fusion: A Comprehensive Review

Li, J., Zhao, W., Liu, S., et al.: Deep learning in multimodal remote sensing data fusion: A comprehensive review. arXiv preprint arXiv:2205.01380 (2022)

work page Pith review arXiv 2022

Show all 15 references

[9]

Scientific reports14(1), 283 (2024)

Mohammed, A., Saif, O., Abo-Adma, M., Fahmy, A., Elazab, R.: Strategies and sustainability in fast charging station deployment for electric vehicles. Scientific reports14(1), 283 (2024)

2024
[10]

Remote Sensing Applications: Society and Environment37, 101447 (2025)

Prakash, P., Hanafin, J., Sarkar, D., Olszewska, M.: Accelerating electric vehicle (ev) adoption: A remote sensing data driven and deep learning-based approach for planning public car charging infrastructure. Remote Sensing Applications: Society and Environment37, 101447 (2025)

2025
[11]

Nature Communications13(1), 3213 (2022)

Raman, G., Raman, G., Peng, J.C.H.: Resilience of urban public electric vehicle charging infrastructure to flooding. Nature Communications13(1), 3213 (2022)

2022
[12]

npj Sustainable Mobility and Transport1(1), 4 (2024)

Sayarshad, H.R.: Optimization of electric charging infrastructure: integrated model for routing and charging coordination with power-aware operations. npj Sustainable Mobility and Transport1(1), 4 (2024)

2024
[13]

Transportation Research Part D: Transport and Environment128, 104103 (2024)

Torkey, A., Zaki, M.H., El Damatty, A.A.: Transportation electrification: a critical review of evs mobility during disruptive events. Transportation Research Part D: Transport and Environment128, 104103 (2024)

2024
[14]

IEEE Access12, 5404–5461 (2024)

Yuvaraj, T., Devabalaji, K., Kumar, J.A., Thanikanti, S.B., Nwulu, N.I.: A comprehensive review and analysis of the allocation of electric vehicle charging stations in distribution networks. IEEE Access12, 5404–5461 (2024)

2024
[15]

arXiv preprint arXiv:2405.19053 (2024)

Zhang, Z., et al.: Multiscale spatio-temporal enhanced short-term load forecasting of electric vehicle charging stations. arXiv preprint arXiv:2405.19053 (2024)

2024 arXiv

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.

[1] [1]

Sustainability 14(20), 13295 (2022) 10 O

Almutairi, A.: Impact assessment of diverse ev charging infrastructures on overall service reliability. Sustainability 14(20), 13295 (2022) 10 O. Karaku¸ s & P. Corcoran © Carto © OpenStreetMap contributors0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 Avg RSERI Score (a) RSERI Score Distribution © Carto © OpenStreetMap contributors RSERI Class Low Moderate High (b) R...

work page 2022

[2] [2]

Environmental Science and Pollution Research30, 68054– 68083 (2023)

Banegas, J., Mamkhezri, J.: A systematic review of geographic information systems based methods and criteria used for electric vehicle charging station site selection. Environmental Science and Pollution Research30, 68054– 68083 (2023)

work page 2023

[3] [3]

Energy (2025) A Multi-Modal Spatial Risk Framework for EV Charging Infrastructure Using Remote Sensing 11

Batic, M., Breckner, L., Weibelzahl, M.: Geodemographic aware electric vehicle charging location planning for equitable infrastructure deployment. Energy (2025) A Multi-Modal Spatial Risk Framework for EV Charging Infrastructure Using Remote Sensing 11

work page 2025

[4] [4]

Heliyon10(7) (2024)

C ¸ elik, S., Ok, S ¸.: Electric vehicle charging stations: Model, algorithm, simulation, location, and capacity planning. Heliyon10(7) (2024)

work page 2024

[5] [5]

Transportation Research Part D: Transport and Environment139, 104580 (2025)

Halder, B.: Remote sensing-based optimal electric vehicle charging station identification in kolkata urban agglom- eration. Transportation Research Part D: Transport and Environment139, 104580 (2025)

work page 2025

[6] [6]

IEEE Geoscience and remote sensing magazine5(1), 53–65 (2017)

Harb, M.M., Dell’Acqua, F.: Remote sensing in multirisk assessment: Improving disaster preparedness. IEEE Geoscience and remote sensing magazine5(1), 53–65 (2017)

work page 2017

[7] [7]

Renewable and sustainable energy reviews154, 111733 (2022)

LaMonaca, S., Ryan, L.: The state of play in electric vehicle charging services–a review of infrastructure provision, players, and policies. Renewable and sustainable energy reviews154, 111733 (2022)

work page 2022

[8] [8]

Deep Learning in Multimodal Remote Sensing Data Fusion: A Comprehensive Review

Li, J., Zhao, W., Liu, S., et al.: Deep learning in multimodal remote sensing data fusion: A comprehensive review. arXiv preprint arXiv:2205.01380 (2022)

work page Pith review arXiv 2022