{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,8]],"date-time":"2026-05-08T21:14:37Z","timestamp":1778274877850,"version":"3.51.4"},"reference-count":30,"publisher":"MDPI AG","issue":"5","license":[{"start":{"date-parts":[[2026,5,8]],"date-time":"2026-05-08T00:00:00Z","timestamp":1778198400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Systems"],"abstract":"<jats:p>Accurate demand forecasting across multiple aggregation levels is essential for managing complex supply networks, where operations must balance inventory costs, service levels, and resource coordination under non-stationary and heterogeneous demand patterns. Existing spatiotemporal models typically treat all forecasting units at a single resolution, obscuring inherent hierarchical structures and often producing inconsistent predictions across levels. This study proposes a Hierarchical Hybrid Spatio-Temporal Demand Forecasting (H2SDF) architecture that formulates multi-granularity forecasting as a coupled system-of-systems problem. H2SDF decomposes the task into three coordinated layers. At the macro layer, a frequency-aware model extracts global trends and multi-scale periodicities from aggregate demand, providing a stable system-level reference. At the meso layer, a Transformer-based multi-task learner disaggregates the macro signal into location-specific forecasts while learning dynamic inter-location dependencies via self-attention, avoiding reliance on predefined static graphs. At the micro layer, gradient-boosted tree models refine category-level predictions by fusing upstream signals with contextual covariates to correct residual errors. A top-down coupling mechanism propagates forecasts and consistency constraints across layers. Experiments on a 2976 h real-world dataset with 18 locations and 8 product categories demonstrate that H2SDF reduces RMSE and improves R2 compared with state-of-the-art baselines across all three granularities. The results confirm that hierarchical decomposition with heterogeneous model synergy effectively mitigates demand uncertainty and strengthens decision support for inventory, logistics, and workforce planning.<\/jats:p>","DOI":"10.3390\/systems14050527","type":"journal-article","created":{"date-parts":[[2026,5,8]],"date-time":"2026-05-08T20:11:13Z","timestamp":1778271073000},"page":"527","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["A Coupled Hierarchical Architecture for Multi-Granularity Demand Forecasting"],"prefix":"10.3390","volume":"14","author":[{"given":"Liang","family":"Nie","sequence":"first","affiliation":[{"name":"Logistics Engineering College, Shanghai Maritime University, Shanghai 201306, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Huaixia","family":"Shi","sequence":"additional","affiliation":[{"name":"Logistics Engineering College, Shanghai Maritime University, Shanghai 201306, China"},{"name":"Business School, Shanghai Dianji University, Shanghai 201306, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Qinglei","family":"Zhang","sequence":"additional","affiliation":[{"name":"China Institute of FTZ Supply Chain, Shanghai Maritime University, Shanghai 201306, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jiyun","family":"Qin","sequence":"additional","affiliation":[{"name":"China Institute of FTZ Supply Chain, Shanghai Maritime University, Shanghai 201306, China"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1968","published-online":{"date-parts":[[2026,5,8]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","unstructured":"Zhang, Y., Fan, X., Cao, Y., and Xue, J. (2024). Exploring symbiosis: Innovatively unveiling the interplay between the cold chain logistics of fresh agricultural products and the ecological environment. Agriculture, 14.","DOI":"10.3390\/agriculture14040609"},{"key":"ref_2","doi-asserted-by":"crossref","unstructured":"Shi, H., Zhang, Q., and Qin, J. (2024). Cold Chain Logistics and Joint Distribution: A Review of Fresh Logistics Modes. Systems, 12.","DOI":"10.3390\/systems12070264"},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"107610","DOI":"10.1016\/j.eiar.2024.107610","article-title":"Exploring the path to the sustainable development of cold chain logistics for fresh agricultural products in China","volume":"108","author":"Fan","year":"2024","journal-title":"Environ. Impact Assess. Rev."},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"124693","DOI":"10.1016\/j.eswa.2024.124693","article-title":"Railway cold chain freight demand forecasting with graph neural networks: A novel GraphARMA-GRU model","volume":"255","author":"Peng","year":"2024","journal-title":"Expert Syst. Appl."},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"314","DOI":"10.1108\/K-11-2021-1111","article-title":"Forecasting the demand for cold chain logistics of agricultural products with Markov-optimised mean GM (1, 1) model\u2014A case study of Guangxi Province, China","volume":"53","author":"Tang","year":"2024","journal-title":"Kybernetes"},{"key":"ref_6","first-page":"1","article-title":"Developing Optimal Demand Forecasting Models for a Very Short Shelf-Life Item: A Case of Perishable Products in Online\u2019s Retail Business","volume":"30","author":"Premrudikul","year":"2023","journal-title":"J. Inf. Technol. Appl. Manag."},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"103253","DOI":"10.1016\/j.jretconser.2022.103253","article-title":"Requirement analysis and service optimization of multiple category fresh products in online retailing using importance-Kano analysis","volume":"72","author":"Zhang","year":"2023","journal-title":"J. Retail. Consum. Serv."},{"key":"ref_8","doi-asserted-by":"crossref","unstructured":"Wang, C.C. (2025). Dynamic Dual-Phase Forecasting Model for New Product Demand Using Machine Learning and Statistical Control. Mathematics, 13.","DOI":"10.3390\/math13101613"},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"104317","DOI":"10.1016\/j.jretconser.2025.104317","article-title":"Timing Matters: How pre-and post-holiday promotions affect fresh and frozen product sales in grocery retail","volume":"85","author":"Darbanian","year":"2025","journal-title":"J. Retail. Consum. Serv."},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"2659","DOI":"10.3390\/make6040128","article-title":"Enhancing Hierarchical Sales Forecasting with Promotional Data: A Comparative Study Using ARIMA and Deep Neural Networks","volume":"6","author":"Teixeira","year":"2024","journal-title":"Mach. Learn. Knowl. Extr."},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"29","DOI":"10.56557\/jobari\/2025\/v31i29133","article-title":"Leveraging Data Science for Demand Forecasting and Inventory Management: A Comprehensive Review","volume":"31","author":"Olatunji","year":"2025","journal-title":"J. Basic Appl. Res. Int."},{"key":"ref_12","first-page":"652","article-title":"Retail demand forecasting using Spatial-Temporal gradient boosting methods","volume":"64","author":"Wang","year":"2024","journal-title":"J. Comput. Inf. Syst."},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"216","DOI":"10.1016\/j.ijforecast.2025.03.003","article-title":"Hierarchical neural additive models for interpretable demand forecasts","volume":"42","author":"Feddersen","year":"2025","journal-title":"Int. J. Forecast."},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"170","DOI":"10.32996\/jbms.2024.6.2.17","article-title":"Product Demand Forecasting with Neural Networks and Macroeconomic Indicators: A Comparative Study among Product Categories","volume":"6","author":"Nguyen","year":"2024","journal-title":"J. Bus. Manag. Stud."},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"66","DOI":"10.1007\/s12065-025-01053-7","article-title":"MCDFN: Supply chain demand forecasting via an explainable multi-channel data fusion network model","volume":"18","author":"Jahin","year":"2025","journal-title":"Evol. Intell."},{"key":"ref_16","first-page":"1","article-title":"Improving supply chain planning for perishable food: Data-driven implications for waste prevention","volume":"94","author":"Birkmaier","year":"2024","journal-title":"J. Bus. Econ."},{"key":"ref_17","first-page":"91","article-title":"Combinatorial Optimisation Model for E-Commerce Retail Merchant Demand Forecasting Based on ARIMA and LSTM","volume":"5","author":"Li","year":"2024","journal-title":"Inf. Syst. Econ."},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"1623","DOI":"10.1002\/for.3259","article-title":"Potential Demand Forecasting for Steel Products in Spot Markets Using a Hybrid SARIMA-LSSVM Approach","volume":"44","author":"Huang","year":"2025","journal-title":"J. Forecast."},{"key":"ref_19","doi-asserted-by":"crossref","unstructured":"Douaioui, K., Oucheikh, R., Benmoussa, O., and Mabrouki, C. (2024). Machine Learning and Deep Learning Models for Demand Forecasting in Supply Chain Management: A Critical Review. Appl. Syst. Innov., 7.","DOI":"10.3390\/asi7050093"},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"124409","DOI":"10.1016\/j.eswa.2024.124409","article-title":"A novel attLSTM framework combining the attention mechanism and bidirectional LSTM for demand forecasting","volume":"254","author":"Cui","year":"2024","journal-title":"Expert Syst. Appl."},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"1555","DOI":"10.1080\/13658816.2023.2203218","article-title":"Developing a multiview spatiotemporal model based on deep graph neural networks to predict the travel demand by bus","volume":"37","author":"Zhao","year":"2023","journal-title":"Int. J. Geogr. Inf. Sci."},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"237882","DOI":"10.1016\/j.jpowsour.2025.237882","article-title":"A hybrid deep learning model for load forecasting of electric vehicle charging stations using time series decomposition","volume":"655","author":"Wu","year":"2025","journal-title":"J. Power Sources"},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"2391952","DOI":"10.1080\/17538947.2024.2391952","article-title":"Deep learning for spatiotemporal forecasting in Earth system science: A review","volume":"17","author":"Yu","year":"2024","journal-title":"Int. J. Digit. Earth"},{"key":"ref_24","first-page":"121","article-title":"Employees\u2019 perceived respect and performance in Logistics 4.0: A dyadic perspective of the congruence between employee voice and supervisor listening","volume":"56","author":"Pang","year":"2025","journal-title":"Int. J. Phys. Distrib. Logist. Manag."},{"key":"ref_25","doi-asserted-by":"crossref","first-page":"119706","DOI":"10.1016\/j.renene.2023.119706","article-title":"Week-ahead hourly solar irradiation forecasting method based on ICEEMDAN and TimesNet networks","volume":"220","author":"Zhao","year":"2024","journal-title":"Renew. Energy"},{"key":"ref_26","doi-asserted-by":"crossref","first-page":"110732","DOI":"10.1016\/j.ijepes.2025.110732","article-title":"A novel multi-task learning model based on Transformer-LSTM for wind power forecasting","volume":"169","author":"Zhang","year":"2025","journal-title":"Int. J. Electr. Power Energy Syst."},{"key":"ref_27","doi-asserted-by":"crossref","first-page":"648","DOI":"10.1108\/TR-05-2023-0275","article-title":"Tourism demand forecasting: A deep learning model based on spatial-temporal transformer","volume":"80","author":"Chen","year":"2025","journal-title":"Tour. Rev."},{"key":"ref_28","doi-asserted-by":"crossref","unstructured":"Chen, T., and Guestrin, C. (2016). Xgboost: A scalable tree boosting system. Proceedings of the 22nd Acm Sigkdd International Conference on Knowledge Discovery and Data Mining, ACM Digital Library.","DOI":"10.1145\/2939672.2939785"},{"key":"ref_29","first-page":"4407","article-title":"Research on cross-border e-commerce customer churn prediction based on enhanced XGBoost algorithm with temporal-spatial features","volume":"25","author":"Le","year":"2025","journal-title":"J. Comput. Methods Sci. Eng."},{"key":"ref_30","unstructured":"Wang, Y., Gu, J., Long, L., Li, X., Shen, L., Fu, Z., Zhou, X., and Jiang, X. (2025). FreshRetailNet-50K: A Stockout-Annotated Censored Demand Dataset for Latent Demand Recovery and Forecasting in Fresh Retail. arXiv."}],"container-title":["Systems"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2079-8954\/14\/5\/527\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,5,8]],"date-time":"2026-05-08T20:36:58Z","timestamp":1778272618000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2079-8954\/14\/5\/527"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,5,8]]},"references-count":30,"journal-issue":{"issue":"5","published-online":{"date-parts":[[2026,5]]}},"alternative-id":["systems14050527"],"URL":"https:\/\/doi.org\/10.3390\/systems14050527","relation":{},"ISSN":["2079-8954"],"issn-type":[{"value":"2079-8954","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,5,8]]}}}