{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,11]],"date-time":"2026-03-11T21:25:35Z","timestamp":1773264335649,"version":"3.50.1"},"reference-count":43,"publisher":"MDPI AG","issue":"3","license":[{"start":{"date-parts":[[2026,3,9]],"date-time":"2026-03-09T00:00:00Z","timestamp":1773014400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IJGI"],"abstract":"<jats:p>Urban commercial restructuring, driven by the closure of traditional supermarkets and the expansion of new-format superstores, creates a large-scale spatial reallocation challenge requiring scientific location-allocation methods. Traditional heuristic algorithms such as Genetic Algorithm (GA) struggle with discrete spatial optimization under 400+ candidate sites and complex geographic mask constraints: they converge slowly and easily fall into local optima. This study proposes a Deep Reinforcement Learning (DRL) framework named GeoPPO (Geospatial Proximal Policy Optimization) to address this gap. Using Xi\u2019an\u2019s retail restructuring as a case setting\u2014427 candidate locations and multidimensional geographic features\u2014the approach models spatial constraints via a gridded environment encoded as a five-channel state tensor. Key innovations include a dynamic action-constraint mechanism that masks invalid actions based on boundary rules and competition avoidance, and a curriculum learning strategy that enables stable convergence. The framework fills the need for methods that handle hard spatial constraints in large-scale location-allocation. Tests demonstrate rapid convergence within 1,000 epochs, achieving 75% average demand coverage\u20142.7% and 5.5% higher than GA and Particle Swarm Optimization (PSO), respectively. Ablation experiments confirm that Vanilla PPO without dynamic action masking fails to produce feasible solutions. The framework offers a feasible technical path for handling highly dynamic urban facility spatial configuration with geographic mask constraints.<\/jats:p>","DOI":"10.3390\/ijgi15030114","type":"journal-article","created":{"date-parts":[[2026,3,9]],"date-time":"2026-03-09T15:49:42Z","timestamp":1773071382000},"page":"114","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["GeoPPO\u2014A Location-Allocation Method of Superstores Based on Deep Reinforcement Learning\u2014A Case Study of Xi\u2019an"],"prefix":"10.3390","volume":"15","author":[{"given":"Yuxuan","family":"Hu","sequence":"first","affiliation":[{"name":"School of Remote Sensing and Information Engineering, Wuhan University, Wuhan 430079, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4088-1637","authenticated-orcid":false,"given":"Kun","family":"Qin","sequence":"additional","affiliation":[{"name":"School of Remote Sensing and Information Engineering, Wuhan University, Wuhan 430079, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8651-9505","authenticated-orcid":false,"given":"Shaohua","family":"Wang","sequence":"additional","affiliation":[{"name":"State Key Laboratory of Remote Sensing Science, Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100101, China"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1968","published-online":{"date-parts":[[2026,3,9]]},"reference":[{"key":"ref_1","unstructured":"State Council of the People\u2019s Republic of China (2025, February 01). Approval of Xi\u2019an Territorial Spatial Master Plan (2021\u20132035). State Council Document No. 13, Available online: https:\/\/www.gov.cn\/zhengce\/zhengceku\/202501\/content_7000441.htm."},{"key":"ref_2","unstructured":"Caixin Global (2025, February 01). In Depth: E-Commerce Giants Shun Supermarkets as Sales Slump. Available online: https:\/\/www.caixinglobal.com\/2025-01-16\/in-depth-e-commerce-giants-shun-supermarkets-as-sales-slump-102279780.html."},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"395","DOI":"10.1016\/j.ijresmar.2005.09.005","article-title":"An exploratory look at supermarket shopping paths","volume":"22","author":"Larson","year":"2005","journal-title":"Int. J. Res. Mark."},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"184","DOI":"10.1177\/0739456X9301200302","article-title":"Locational models, geographic information and planning support systems","volume":"12","author":"Harris","year":"1993","journal-title":"J. Plan. Educ. Res."},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"343","DOI":"10.1177\/0885412210364589","article-title":"Evaluation in urban planning: Advances and prospects","volume":"24","author":"Oliveira","year":"2010","journal-title":"J. Plan. Lit."},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"26","DOI":"10.1109\/MSP.2017.2743240","article-title":"Deep reinforcement learning: A brief survey","volume":"34","author":"Arulkumaran","year":"2017","journal-title":"IEEE Signal Process. Mag."},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"2299211","DOI":"10.1080\/17538947.2023.2299211","article-title":"Sponet: Solve spatial optimization problem using deep reinforcement learning for urban spatial decision analysis","volume":"17","author":"Liang","year":"2024","journal-title":"Int. J. Digit. Earth"},{"key":"ref_8","first-page":"104454","article-title":"AIAM: Adaptive interactive attention model for solving p-Median problem via deep reinforcement learning","volume":"138","author":"Liang","year":"2025","journal-title":"Int. J. Appl. Earth Obs. Geoinf."},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"387","DOI":"10.1007\/s10458-005-2631-2","article-title":"Cooperative multi-agent learning: The state of the art","volume":"11","author":"Panait","year":"2005","journal-title":"Auton. Agents Multi-Agent Syst."},{"key":"ref_10","first-page":"5998","article-title":"Attention is all you need","volume":"Volume 30","author":"Vaswani","year":"2017","journal-title":"Advances in Neural Information Processing Systems"},{"key":"ref_11","first-page":"1","article-title":"A review of population-based meta-heuristic algorithms","volume":"5","author":"Beheshti","year":"2013","journal-title":"Int. J. Adv. Soft Comput. Appl."},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"1205","DOI":"10.3846\/16111699.2015.1113198","article-title":"Geomarketing models in supermarket location strategies","volume":"17","year":"2016","journal-title":"J. Bus. Econ. Manag."},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"04021062","DOI":"10.1061\/(ASCE)UP.1943-5444.0000780","article-title":"Optimizing road network density considering automobile traffic efficiency: Theoretical approach","volume":"148","author":"Cui","year":"2022","journal-title":"J. Urban Plan. Dev."},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"102610","DOI":"10.1016\/j.cities.2020.102610","article-title":"Visualizing and exploring POI configurations of urban regions on POI-type semantic space","volume":"99","author":"Liu","year":"2020","journal-title":"Cities"},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"20","DOI":"10.1007\/s43762-022-00047-w","article-title":"Points of Interest (POI): A commentary on the state of the art, challenges, and prospects for the future","volume":"2","author":"Psyllidis","year":"2022","journal-title":"Comput. Urban Sci."},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"103126","DOI":"10.1016\/j.apgeog.2023.103126","article-title":"Retail location modeling of supermarket chains in Taipei city","volume":"161","author":"Lin","year":"2023","journal-title":"Appl. Geogr."},{"key":"ref_17","unstructured":"Engstrom, L., Ilyas, A., Santurkar, S., Tsipras, D., Janoos, F., Rudolph, L., and Madry, A. (2020). Implementation matters in deep policy gradients: A case study on ppo and trpo. arXiv."},{"key":"ref_18","first-page":"2017","article-title":"Spatial transformer networks","volume":"Volume 28","author":"Jaderberg","year":"2015","journal-title":"Advances in Neural Information Processing Systems"},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"122","DOI":"10.1108\/AA-02-2014-018","article-title":"A genetic algorithm for supermarket location problem","volume":"35","author":"Alnahhal","year":"2015","journal-title":"Assem. Autom."},{"key":"ref_20","unstructured":"Guo, W., Xu, Y., and Jin, Y. (2023). Swap-based deep reinforcement learning for facility location problems in networks. arXiv."},{"key":"ref_21","first-page":"3274","article-title":"Triple-q: A model-free algorithm for constrained reinforcement learning with sublinear regret and zero constraint violation","volume":"Volume 151","author":"Wei","year":"2022","journal-title":"International Conference on Artificial Intelligence and Statistics (AISTATS 2022)"},{"key":"ref_22","doi-asserted-by":"crossref","unstructured":"Su, H., Zheng, Y., Ding, J., Jin, D., and Li, Y. (2024). Large-scale Urban Facility Location Selection with Knowledge-informed Reinforcement Learning. Proceedings of the 32nd ACM International Conference on Advances in Geographic Information Systems (ACM SIGSPATIAL 2024), 29 October\u20131 November 2024, ACM.","DOI":"10.1145\/3678717.3691254"},{"key":"ref_23","doi-asserted-by":"crossref","unstructured":"Wang, S., Xu, D., Zhou, J., Su, C., Li, X., Liang, X., Cao, C., Liu, C., and Zhong, Y. (2024). Convenience stores geospatial location optimization analytics using deep reinforcement learning. Proceedings of the 3rd ACM SIGSPATIAL International Workshop on Spatial Big Data and AI for Industrial Applications, ACM.","DOI":"10.1145\/3681766.3699605"},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"2723","DOI":"10.1177\/0042098010377368","article-title":"Introduction. Urban networks and network theory: The city as the connector of multiple networks","volume":"47","author":"Pflieger","year":"2010","journal-title":"Urban Stud."},{"key":"ref_25","doi-asserted-by":"crossref","first-page":"235","DOI":"10.1111\/0034-6527.00283","article-title":"Supermarket choice and supermarket competition in market equilibrium","volume":"71","author":"Smith","year":"2004","journal-title":"Rev. Econ. Stud."},{"key":"ref_26","doi-asserted-by":"crossref","unstructured":"Nowor\u00f3l, A., Kopyci\u0144ski, P., Ha\u0142at, P., Salamon, J., and Ho\u0142uj, A. (2022). The 15-minute city\u2014The geographical proximity of services in Krakow. Sustainability, 14.","DOI":"10.3390\/su14127103"},{"key":"ref_27","doi-asserted-by":"crossref","first-page":"658","DOI":"10.1080\/13658816.2016.1220561","article-title":"Measurements of POI-based mixed use and their relationships with neighbourhood vibrancy","volume":"31","author":"Yue","year":"2017","journal-title":"Int. J. Geogr. Inf. Sci."},{"key":"ref_28","doi-asserted-by":"crossref","first-page":"540","DOI":"10.1177\/030913259602000408","article-title":"GIS, spatial analysis and spatial statistics","volume":"20","author":"Unwin","year":"1996","journal-title":"Prog. Hum. Geogr."},{"key":"ref_29","doi-asserted-by":"crossref","first-page":"227","DOI":"10.1016\/j.sbspro.2014.07.038","article-title":"The usage of Geographical Information Systems (GIS) in the marketing decision making process: A case study for determining supermarket locations","volume":"148","author":"Turk","year":"2014","journal-title":"Procedia-Soc. Behav. Sci."},{"key":"ref_30","unstructured":"Oak Ridge National Laboratory (2025, February 01). LandScan Global 2023: Silver Edition [Data Set], Available online: https:\/\/landscan.ornl.gov\/."},{"key":"ref_31","doi-asserted-by":"crossref","first-page":"12","DOI":"10.1109\/MPRV.2008.80","article-title":"OpenStreetMap: User-generated street maps","volume":"7","author":"Haklay","year":"2008","journal-title":"IEEE Pervasive Comput."},{"key":"ref_32","doi-asserted-by":"crossref","unstructured":"Weng, Y., Xu, M., Chen, X., Peng, C., Xiang, H., Xie, P., and Yin, H. (2024). An Efficient Algorithm for Extracting Railway Tracks Based on Spatial-Channel Graph Convolutional Network and Deep Neural Residual Network. ISPRS Int. J. Geo-Inf., 13.","DOI":"10.3390\/ijgi13090309"},{"key":"ref_33","unstructured":"Xi\u2019an Municipal Bureau of Statistics (2025, February 01). Communiqu\u00e9 on Major Data of the Seventh National Population Census of Xi\u2019an City, Available online: https:\/\/tjj.xa.gov.cn\/."},{"key":"ref_34","doi-asserted-by":"crossref","first-page":"2512","DOI":"10.1109\/TKDE.2020.3007194","article-title":"Where to go next: A spatio-temporal gated network for next poi recommendation","volume":"34","author":"Zhao","year":"2020","journal-title":"IEEE Trans. Knowl. Data Eng."},{"key":"ref_35","doi-asserted-by":"crossref","first-page":"246","DOI":"10.1038\/s42256-022-00462-y","article-title":"Quantifying the spatial homogeneity of urban road networks via graph neural networks","volume":"4","author":"Xue","year":"2022","journal-title":"Nat. Mach. Intell."},{"key":"ref_36","doi-asserted-by":"crossref","first-page":"436","DOI":"10.1038\/nature14539","article-title":"Deep learning","volume":"521","author":"LeCun","year":"2015","journal-title":"Nature"},{"key":"ref_37","doi-asserted-by":"crossref","first-page":"2912","DOI":"10.1109\/TASE.2023.3271666","article-title":"A DRL-based reactive scheduling policy for flexible job shops with random job arrivals","volume":"21","author":"Zhao","year":"2023","journal-title":"IEEE Trans. Autom. Sci. Eng."},{"key":"ref_38","doi-asserted-by":"crossref","unstructured":"Lin, C.H., Yumer, E., Wang, O., Shechtman, E., and Lucey, S. (2018). St-gan: Spatial transformer generative adversarial networks for image compolocation selection. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, IEEE.","DOI":"10.1109\/CVPR.2018.00985"},{"key":"ref_39","doi-asserted-by":"crossref","first-page":"340","DOI":"10.1111\/j.1538-4632.1999.tb00989.x","article-title":"Local forms of spatial analysis","volume":"31","author":"Fotheringham","year":"1999","journal-title":"Geogr. Anal."},{"key":"ref_40","doi-asserted-by":"crossref","first-page":"17","DOI":"10.1111\/pirs.12216","article-title":"Urban networks: Connecting markets, people, and ideas","volume":"95","author":"Glaeser","year":"2016","journal-title":"Pap. Reg. Sci."},{"key":"ref_41","doi-asserted-by":"crossref","first-page":"748","DOI":"10.1038\/s43588-023-00503-5","article-title":"Spatial planning of urban communities via deep reinforcement learning","volume":"3","author":"Zheng","year":"2023","journal-title":"Nat. Comput. Sci."},{"key":"ref_42","doi-asserted-by":"crossref","first-page":"2537","DOI":"10.1109\/TKDE.2017.2741484","article-title":"Spatial-aware hierarchical collaborative deep learning for POI recommendation","volume":"29","author":"Yin","year":"2017","journal-title":"IEEE Trans. Knowl. Data Eng."},{"key":"ref_43","doi-asserted-by":"crossref","first-page":"811","DOI":"10.1287\/mksc.1080.0398","article-title":"Supermarket pricing strategies","volume":"27","author":"Ellickson","year":"2008","journal-title":"Mark. Sci."}],"container-title":["ISPRS International Journal of Geo-Information"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2220-9964\/15\/3\/114\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,3,11]],"date-time":"2026-03-11T05:24:32Z","timestamp":1773206672000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2220-9964\/15\/3\/114"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,3,9]]},"references-count":43,"journal-issue":{"issue":"3","published-online":{"date-parts":[[2026,3]]}},"alternative-id":["ijgi15030114"],"URL":"https:\/\/doi.org\/10.3390\/ijgi15030114","relation":{},"ISSN":["2220-9964"],"issn-type":[{"value":"2220-9964","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,3,9]]}}}