[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"item-3613":3,"related-3613":113},{"id":4,"title":5,"url":6,"summary":7,"summary_zh":8,"content":9,"source_name":10,"source_url":6,"published_at":11,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":15,"score_detail":16,"sources":24,"tags":26,"primary_topic":32,"secondary_topics":38,"classification":51,"search_phrases":92,"slug":95,"view_count":87,"doi":96,"paper":97,"created_at":112},3613,"AI-Assisted Demand Forecasting and Production Scheduling for High-Volume Food Manufacturing in Saudi Arabia: A Food Security Framework for Vision 2030","https:\u002F\u002Fdoi.org\u002F10.64388\u002Firev10i3-1723457","High-volume food manufacturing in Saudi Arabia is increasingly exposed to volatile demand, short product life cycles, promotion-driven peaks, imported input dependencies, capacity bottlenecks, and the operational consequences of seasonal and religious events. These conditions make demand forecasting and production scheduling inseparable rather than sequential planning tasks. This review synthesizes research published between 2020 and 2025 on artificial intelligence-assisted demand forecasting, perishable-product planning, finite-capacity scheduling, digital-twin-enabled decision support, and food supply-chain resilience. The study develops an integrated framework in which demand sensing produces probabilistic forecasts that are translated into production decisions through capacity, shelf-life, sanitation, changeover, labour, inventory, and service constraints. Evidence indicates that machine-learning and deep-learning methods can improve forecast accuracy when promotions, weather, calendar events, regional variation, and censored demand are represented appropriately, while optimization models can convert these signals into feasible schedules that reduce changeovers, waste, shortages, and unstable resource use [1-8]. The review also identifies a persistent implementation gap: forecasting studies often stop at predictive accuracy, whereas scheduling studies frequently assume demand inputs are fixed and reliable. The proposed framework closes this gap by linking forecast uncertainty, rolling-horizon scheduling, execution feedback, and food-security outcomes. For Saudi manufacturers, the value of this integration lies not only in lower cost, but also in improving product availability, freshness, localized processing capacity, shock recovery, and evidence-based management under Vision 2030. The paper concludes with a staged implementation roadmap and research priorities for explainability, cross-site scheduling, event-aware forecasting, data governance, and resilient planning.","沙特阿拉伯的大规模食品制造日益面临需求波动、产品生命周期短、促销驱动的峰值、进口原料依赖、产能瓶颈以及季节性和宗教事件的运营影响。这些条件使得需求预测与生产调度成为不可分割的任务，而非顺序规划任务。本文综述了2020年至2025年间发表的相关研究，涵盖人工智能辅助需求预测、易腐产品规划、有限产能调度、数字孪生赋能的决策支持以及食品供应链韧性。研究构建了一个集成框架，其中需求感知产生概率预测，并通过产能、保质期、卫生、换产、劳动力、库存和服务约束转化为生产决策。证据表明，当促销、天气、日历事件、区域差异和截尾需求得到适当表征时，机器学习和深度学习方法能够提高预测精度，而优化模型可将这些信号转化为可行的调度方案，从而减少换产、浪费、缺货和不稳定的资源使用[1-8]。本综述还指出了一个持续存在的实施缺口：预测研究往往止步于预测精度，而调度研究则常常假设需求输入是固定且可靠的。所提出的框架通过将预测不确定性、滚动时域调度、执行反馈和食品安全结果联系起来，弥合了这一缺口。对于沙特制造商而言，这种集成的价值不仅在于降低成本，还在于改善产品可得性、新鲜度、本地化加工能力、冲击恢复能力以及“2030愿景”下的循证管理。本文最后提出了分阶段实施路线图以及可解释性、跨厂区调度、事件感知预测、数据治理和韧性规划方面的研究优先事项。",null,"Iconic Research and Engineering Journals","2026-09-25T00:00:00Z","论文",10,false,60,{"impact":17,"substance":18,"depth":19,"authority":20,"freshness":21,"relevant":22,"comment":23},12,18,16,6,8,1,"综述性论文，提出AI需求预测与生产排程整合框架，对智慧农业与粮食安全有参考价值，但属境外研究且期刊权威性一般。",[25],{"name":10,"url":6},[27,28,29,30,31],"智慧农业","农业人工智能","粮食安全","供应链韧性","生产调度",{"code":33,"slug":34,"name":35,"path":36,"parent_code":37},"A8","yield-food-security","单产提升与粮食安全","\u002Ftopic\u002Ffield-crops\u002Fyield-food-security","A",[39,45],{"code":40,"slug":41,"name":42,"path":43,"parent_code":44},"E1","food-processing","加工与食品制造","\u002Ftopic\u002Ffood-supply\u002Ffood-processing","E",{"code":46,"slug":47,"name":48,"path":49,"parent_code":50},"I2","agricultural-ai","农业AI与大模型","\u002Ftopic\u002Fdigital-agriculture\u002Fagricultural-ai","I",{"status":52,"locked":14,"version":53,"method":54,"evidence":55},"classified","2026-10-22.2","rules",{"candidates":56,"agricultural_evidence":59},[57,64,72,78,85],{"code":40,"score":58,"supported":59,"title":60,"abstract":62,"tags":63},11,true,[61],"food manufacturing",[61],[],{"code":46,"score":13,"supported":59,"title":65,"abstract":66,"tags":71},[],[67,68,69,70],"深度学习","机器学习","artificial intelligence","decision support",[28],{"code":33,"score":73,"supported":59,"title":74,"abstract":76,"tags":77},9,[75],"food security",[],[29],{"code":79,"score":80,"supported":59,"title":81,"abstract":82,"tags":84},"I1",4,[],[83],"数字孪生",[27],{"code":86,"score":87,"supported":59,"title":88,"abstract":89,"tags":91},"E4",3,[],[90],"食品安全",[],[93,94],"沙特阿拉伯 食品制造 需求预测","Vision 2030 粮食安全","沙特阿拉伯食品制造需求预测-3613","10.64388\u002Firev10i3-1723457",{"doi":96,"openalex_id":98,"authors":99,"venue":10,"cited_by_count":102,"oa_url":103,"card":104,"direction":110,"ingested_from":111},"W7214295236",[100],{"name":101,"orcid":9},"Atif Maqbool Syed",0,"https:\u002F\u002Fwww.irejournals.com\u002Fformatedpaper\u002F1723457.pdf",{"tldr":105,"method":106,"finding":107,"direction":108,"opportunity":109},"综述2020-2025年AI需求预测与生产调度研究，提出沙特食品制造整合框架以保障Vision 20","文献综述2020-2025年AI预测、易腐品规划、有限产能调度与数字孪生研究。","预测研究止于精度、调度研究假设需求固定，整合框架可弥合实施鸿沟并提升粮食安全。","农业人工智能与决策模型","可研究事件感知预测与滚动调度耦合、跨工厂协同及可解释AI在食品供应链韧性中的应用。","智慧农业 \u002F 农业物联网","openalex","2026-09-27T23:30:14.516185Z",{"total":20,"page":22,"page_size":20,"items":114},[115,189,267,393,459,519],{"id":116,"title":117,"url":118,"summary":119,"summary_zh":120,"content":9,"source_name":121,"source_url":118,"published_at":122,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":123,"score_detail":124,"sources":129,"tags":131,"primary_topic":134,"secondary_topics":135,"classification":137,"search_phrases":167,"slug":170,"view_count":171,"doi":172,"paper":173,"created_at":188},1614,"Artificial intelligence adoption and food security in Saudi Arabia: the critical role of water-use efficiency management","https:\u002F\u002Fdoi.org\u002F10.3389\u002Ffsufs.2026.1955058","Introduction Ensuring food security and increasing water-use efficiency have become strategic priorities for Saudi Arabia due to water scarcity, climate change, and rising food demand. Artificial intelligence (AI) advances significant opportunities to improve agricultural productivity and resource management; however, empirical evidence on its dynamic relationship with food security and water-use efficiency remains limited. Methods This study investigates the dynamic relationships among Information and Communication Technology (ICT) goods imports as an indirect proxy for AI adoption, food security, and water-use efficiency for various purposes in Saudi Arabia, utilizing annual data from 2000 to 2023. The analysis employs a Vector Autoregression (VAR) framework alongside Granger causality tests, impulse response functions (IRFs), forecast error variance decomposition (FEVD), and historical decomposition (HD). Results The results reveal that ICT goods imports, as an indirect proxy for AI adoption, significantly influence food security and industrial water-use efficiency; likewise, agricultural and service-sector water-use efficiency, in turn, promote AI adoption. The IRF results indicate that positive AI shocks improve food security and industrial water-use efficiency in the short run. However, the FEVD results show that the agricultural and service sectors will become increasingly significant factors in long-term food security. The HD outcomes support the robustness of the study results. Conclusion The study concludes that integrating AI technologies with sustainable water management can boost food security and Structural enhance resource efficiency. These outcomes provide valuable policy insights for advancing Saudi Vision 2030 through AI-driven agricultural innovation and sustainable water resource management. The study contributes to the literature by integrating ICT goods imports, as an indirect proxy for AI adoption, food security, and sector-specific water-use efficiency into a dynamic framework, providing new evidence on the potential role of digital technology diffusion in enhancing food security through improved water-use efficiency.","引言：由于水资源短缺、气候变化及粮食需求不断增长，保障粮食安全和提高用水效率已成为沙特阿拉伯的战略优先事项。人工智能（AI）的进步为提升农业生产力和资源管理提供了重大机遇；然而，关于其与粮食安全和用水效率之间动态关系的实证证据仍然有限。方法：本研究利用2000年至2023年的年度数据，探讨沙特阿拉伯信息和通信技术（ICT）产品进口（作为AI采用的间接代理指标）、粮食安全及各部门用水效率之间的动态关系。分析采用向量自回归（VAR）框架，并结合格兰杰因果关系检验、脉冲响应函数（IRFs）、预测误差方差分解（FEVD）和历史分解（HD）。结果：结果表明，ICT产品进口作为AI采用的间接代理指标，显著影响粮食安全和工业用水效率；同样，农业和服务业的用水效率反过来也促进了AI的采用。脉冲响应函数结果显示，正向的AI冲击在短期内改善了粮食安全和工业用水效率。然而，预测误差方差分解结果表明，农业和服务业将成为长期粮食安全中日益重要的因素。历史分解结果支持了研究结果的稳健性。结论：研究得出结论，将AI技术与可持续水资源管理相结合，可以提升粮食安全并增强资源效率。这些成果为通过AI驱动的农业创新和可持续水资源管理推进沙特2030愿景提供了有价值的政策启示。本研究通过将ICT产品进口（作为AI采用的间接代理指标）、粮食安全和分部门用水效率整合到一个动态框架中，为文献做出了贡献，为数字技术扩散通过改善用水效率增强粮食安全的潜在作用提供了新的证据。","Frontiers in Sustainable Food Systems","2026-09-03T00:00:00Z",76,{"impact":18,"substance":125,"depth":18,"authority":126,"freshness":127,"relevant":22,"comment":128},20,13,7,"研究AI采用与水资源效率对沙特粮食安全的影响，方法严谨，结论具政策参考价值。",[130],{"name":121,"url":118},[27,28,29,132,133],"水资源管理","沙特",{"code":33,"slug":34,"name":35,"path":36,"parent_code":37},[136],{"code":46,"slug":47,"name":48,"path":49,"parent_code":50},{"status":52,"locked":14,"version":53,"method":54,"evidence":138},{"candidates":139,"agricultural_evidence":59},[140,145,149,156,163],{"code":33,"score":141,"supported":59,"title":142,"abstract":143,"tags":144},15,[75],[29,75],[29],{"code":46,"score":17,"supported":59,"title":146,"abstract":147,"tags":148},[69],[69],[28],{"code":150,"score":127,"supported":59,"title":151,"abstract":152,"tags":155},"H3",[],[153,154],"水资源","water resource",[132],{"code":157,"score":20,"supported":59,"title":158,"abstract":159,"tags":162},"H7",[],[160,161],"气候变化","climate change",[],{"code":79,"score":22,"supported":14,"title":164,"abstract":165,"tags":166},[],[],[27],[168,169],"农业人工智能 水资源管理 智慧农业 粮食安全","农业人工智能 水资源管理","农业人工智能水资源管理智慧农业粮食安全-1614",2,"10.3389\u002Ffsufs.2026.1955058",{"doi":172,"openalex_id":174,"authors":175,"venue":121,"cited_by_count":102,"oa_url":182,"card":183,"direction":108,"ingested_from":111},"W7207821862",[176,178,180],{"name":177,"orcid":9},"Raga M. Elzaki",{"name":179,"orcid":9},"Omar A. Alkhateeb",{"name":181,"orcid":9},"Abdulelah A. Alasfour","https:\u002F\u002Fwww.frontiersin.org\u002Fjournals\u002Fsustainable-food-systems\u002Farticles\u002F10.3389\u002Ffsufs.2026.1955058\u002Fpdf",{"tldr":184,"method":185,"finding":186,"direction":108,"opportunity":187},"研究沙特AI采用与粮食安全、水资源效率的动态关系，发现AI提升短期粮食安全与工业用水效率。","使用2000-2023年数据，VAR模型、Granger因果、脉冲响应、方差分解","AI采用显著影响粮食安全与工业用水效率，农业和服务业用水效率促进AI采用，长期农业和服务业对粮食安全","可探索AI与农业用水效率的因果机制，或扩展到其他干旱地区，考虑气候变量和实际AI技术指标。","2026-09-04T23:30:08.321918Z",{"id":190,"title":191,"url":192,"summary":193,"summary_zh":9,"content":194,"source_name":195,"source_url":9,"published_at":196,"category":197,"cover_url":9,"hotness":13,"is_selected":14,"score":198,"score_detail":199,"sources":203,"tags":205,"primary_topic":210,"secondary_topics":211,"classification":218,"search_phrases":262,"slug":265,"view_count":21,"doi":9,"paper":9,"created_at":266},3100,"秋实满田 智赋丰年——中国移动(成都)产业研究院\"谷丰登\"农业大模型赋能山西大同玉米单产提升试点亩产 2481 斤较非试点增产 15.5%","https:\u002F\u002Fiot.youth.cn\u002Fyw\u002F202609\u002Ft20260921_16881000.htm","9-21 中国青年网物联网频道报道中国移动(成都)产业研究院秋收赋能：以\"九天大模型\"为底座自主研发\"谷丰登\"农业大模型、构建覆盖土壤分析\u002F品种优选\u002F精准水肥药方案\u002F产量预估全生命周期服务体系；联合中储粮打造行业首个储粮场景大模型\"稷元\"已落地 3 个中央储备粮库，粮食储存损失降至 0.2% 以下、化学药剂年使用量减少 1-2 次；牵头制定《农业大模型高质量数据集》《农业种植大模型应用测评规范》两项团体标准。山西大同玉米单产提升标杆项目依托空天地一体化监测+北斗高精定位+农业大模型实现玉米耕种管收全流程无人化作业，试点田亩产 2481 斤较非试点增产 15.5%，被农业农村部评为\"全国智慧农业建设典型案例\"。","九月的田野，稻菽飘香，又是一年丰收时。\n\n随着“十五五”规划的全面实施，农业农村现代化建设蹄疾步稳。传统“看天吃饭”的农业生产模式正在被改写，数字技术驱动下，丰收迎来科技赋能、全链提质、数智增效的全新发展阶段。\n\n当前正值金秋丰收的关键时期，中国移动（成都）产业研究院（以下简称“成研院”）充分发挥科技引领作用，贯通智慧种养、智能储粮、数字治理、产销对接全业务链路，推动数字技术与农业生产融合走深走实，为国家粮食安全与乡村全面振兴注入强劲数字动能。\n\n**科技强农：打造从田间到粮仓的“智慧大脑”**\n\n成研院聚焦AI育种创新、种植生产管理与粮食仓储等关键环节，以中国移动“九天大模型”为底座，自主研发谷丰登农业大模型，构建起全周期、智能化的农业生产服务体系，打造扎根田野的“身边的农业知识专家”。\n\n“藏粮于技”的关键，在于让技术真正下沉到田间地头。成研院打造国内首个种植综合农业大模型，覆盖土壤分析、品种优选、精准水肥药方案、产量预估等全生命周期服务，实现“信息随手握、种养全程导、在线有人帮”。在山西大同玉米单产提升标杆项目中，成研院构建“良田、良种、良机、良法”融合体系，依托空天地一体化监测、北斗高精定位与农业大模型三大核心能力，实现玉米耕、种、管、收全流程无人化作业。当年秋收，试点田亩产达2481斤，较非试点区域增产15.5％，被农业农村部评为“全国智慧农业建设典型案例”。\n\n丰产更要保收，储粮减损是守住丰收成果、保障粮食安全的关键一环。成研院联合中储粮打造行业首个储粮场景大模型“稷元”，内置二十余万条储粮知识单元及千余本专业文献，结合智能语音助手“粮小助”，实时采集、动态监测各类粮情核心指标，智能预判粮情变化趋势，提前预警仓储虫害风险，自动生成仓储调控最优方案，同时采用一体化集成设计，适配粮库现有系统，实现核心业务在库内闭环处理，严守粮食仓储信息安全底线。目前，“稷元”大模型已在中储粮三个中央储备粮库落地，粮食储存损失降至0.2%以下，化学药剂年使用量减少1—2次，大幅提升粮食仓储安全水平，守住丰收成果、筑牢储粮安全防线。\n\n此外，成研院牵头制定《农业大模型高质量数据集》《农业种植大模型应用测评规范》两项团体标准，填补行业规范空白，为大模型规范化、规模化应用筑牢基础。\n\n![Image 1](https:\u002F\u002Fiot.youth.cn\u002Fyw\u002F202609\u002FW020260921554681104915.png)\n\n![Image 2](https:\u002F\u002Fiot.youth.cn\u002Fyw\u002F202609\u002FW020260921554681314335.png)\n\n**数字惠农：激活基层治理民生温度**\n\n成研院以数字技术创新破解基层治理痛点、赋能民生服务、激活乡村发展动能，推动数字技术真正扎根田野、服务农民。\n\n在传承着焦裕禄精神的河南兰考葡萄架乡，成研院打造的数字乡村平台全面落地应用。平台贯通乡、村两级业务全链条，整合民政、社保、农业等多部门数据资源，实现“一次采集、多方复用”。平台提供20余项高频服务“掌上办”，让“数据多跑路、群众少跑腿”，并依托AI智能预警，推动基层治理高效化、智能化。目前，平台已覆盖全乡20个行政村，累计收录人口数据超3万条，助力基层干部事务性工作量降低60%以上，实现网格管理一屏总览、AI智能安防实时监测、积分激励激活村民参与，打通基层治理“最后一公里”。\n\n在甘肃积石山县团结村，智慧乡村大脑聚焦民族地区灾后重建与乡村振兴，集成灾后重建、乡村振兴、民族团结、服务治理、综合监测五大模块，一屏统揽耕地分布、帮扶进展、集体经济、群众诉求等村庄“家底”，为村委科学决策提供精准支撑。同时，村里为50户重点关爱家庭免费安装一键呼叫终端，直连紧急救助与亲情通话，让乡村治理更有精度与温度。\n\n**产业富农：搭建产销双向联通桥梁**\n\n从培育乡村产业带头人到打通农产品出村进城通道，成研院以数字技术为纽带，一头连着田间地头的生产者，一头连着广阔市场的消费者，让好产品卖出好价钱，让乡村产业真正活起来、强起来。\n\n作为农业农村部“头雁”项目的技术支撑方，成研院为“头雁”人才培育系统提供的全方位信息化支撑。目前，“头雁”平台已覆盖全国31省份，累计服务7万余名乡村产业带头人，为乡村培育“造血种子”。\n\n如果说头雁工程是为乡村培育“造血种子”，那么“陆海优品公共服务平台”就是为乡村产业打通“出村进城、走向世界”的黄金通道。平台开设免费电商、直播、品控等培训，助力800多万农民转型“新农人”，推动中国移动对口帮扶的五省（区）八县、成渝地区双城经济圈特色农产品实现精准产销对接。目前，平台已带动超2000万元消费帮扶、实现3亿元农产品出口额，服务全国超600万用户，以数智力量畅通国内国际双循环。\n\n丰收，不再只是田野里的金黄稻浪，更是数据流中奔涌的智慧动能。当技术真正扎根泥土，当数字红利真正惠及农民，丰收便有了更坚实的支撑，乡村便有了更持久的活力。面向未来，成研院将继续以数智之力深耕沃土，让“藏粮于技”落地生根，让“数字乡村”可感可及，在广袤田野上书写属于新时代的丰收故事。","中国青年网","2026-09-21T00:00:00Z","报道",84,{"impact":200,"substance":201,"depth":19,"authority":126,"freshness":73,"relevant":22,"comment":202},24,22,"央媒报道的农业大模型落地标杆案例，含亩产2481斤、增产15.5%、储粮损失降至0.2%等硬数据，产业示范价值突出，值得进入每日精选。",[204],{"name":195,"url":192},[206,27,28,207,208,209],"数字乡村","农业大模型","玉米单产提升","粮食仓储",{"code":33,"slug":34,"name":35,"path":36,"parent_code":37},[212,213],{"code":46,"slug":47,"name":48,"path":49,"parent_code":50},{"code":214,"slug":215,"name":216,"path":217,"parent_code":37},"A3","maize","玉米","\u002Ftopic\u002Ffield-crops\u002Fmaize",{"status":52,"locked":14,"version":53,"method":54,"evidence":219},{"candidates":220,"agricultural_evidence":59},[221,226,231,235,239,245,251,257],{"code":33,"score":126,"supported":59,"title":222,"abstract":224,"tags":225},[223],"单产提升",[223],[208,209],{"code":46,"score":126,"supported":59,"title":227,"abstract":229,"tags":230},[228],"大模型",[228],[28,207],{"code":214,"score":17,"supported":59,"title":232,"abstract":233,"tags":234},[216],[216],[208],{"code":79,"score":80,"supported":59,"title":236,"abstract":237,"tags":238},[],[27],[27],{"code":240,"score":87,"supported":59,"title":241,"abstract":242,"tags":244},"H2",[],[243],"土壤",[],{"code":246,"score":87,"supported":59,"title":247,"abstract":248,"tags":250},"I5",[],[249],"物联网",[],{"code":252,"score":87,"supported":59,"title":253,"abstract":254,"tags":256},"J4",[],[255],"秋收",[],{"code":258,"score":22,"supported":14,"title":259,"abstract":260,"tags":261},"I6",[],[],[206],[263,264],"中国移动 谷丰登 农业大模型","山西大同 玉米 亩产2481斤","中国移动谷丰登农业大模型-3100","2026-09-22T00:05:33.912382Z",{"id":268,"title":269,"url":270,"summary":271,"summary_zh":272,"content":9,"source_name":273,"source_url":270,"published_at":274,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":275,"score_detail":276,"sources":279,"tags":281,"primary_topic":283,"secondary_topics":284,"classification":291,"search_phrases":358,"slug":361,"view_count":22,"doi":362,"paper":363,"created_at":392},1380,"Harnessing Digital Twin (DT) Technology for Food Security and Climate Resilience in Sub-Saharan Africa (SSA)","https:\u002F\u002Fdoi.org\u002F10.1016\u002Fj.atech.2026.102532","Sub-Saharan Africa (SSA) faces chronic food insecurity despite possessing approximately 60% of the world’s uncultivated arable land. Field trials generate evidence but are costly and insufficiently scaled to address accelerating climate and demographic pressures. Digital twin (DT) technology, defined as the continuous, bidirectional virtual replication of physical systems using real-time data, supports monitoring, modelling, and optimisation of agrifood systems. To our knowledge, however, no published synthesis has examined DT agriculture research through the lens of SSA food systems or smallholder farming realities. A PRISMA-compliant systematic review was conducted across bibliographic databases using a pre-defined Boolean search and adapted PICOS eligibility criteria. Forty peer-reviewed studies published between 2019 and 2025 were included. Structured extraction captured bibliometric, geographic, methodological, technological, and technology readiness level (TRL) data, analysed using frequency tabulation and thematic clustering. DT agriculture publications roughly tripled between 2019 and 2023 (from three to ten studies), yet 57.5% of studies remain at TRL 2–3. Europe and East Asia dominate the evidence base; no study was led by an SSA institution, and no study reported SSA-based deployment on priority crops such as maize, cassava, sorghum, or cowpea. Internet of Things (IoT) sensor infrastructure (92.5%) and artificial intelligence\u002Fmachine learning (AI\u002FML) algorithms (52.5%) constitute the dominant technology stack. Five thematic clusters were identified: smart farming infrastructure, precision water and nutrient management, controlled environment agriculture, agrifood supply chains, and livestock and aquaculture systems. Government-funded studies tend to achieve higher TRL scores than academically funded counterparts. Synthesising these findings into a system-level, deployment-oriented analysis, we derive a conceptual SSA-DT framework. It combines offline-first edge-computing architecture, hybrid crop modelling, and multi-stakeholder interfaces, with dietary diversity and climate resilience as the two key performance indicators (KPIs), aligned with IITA’s 2024–2030 Strategy. North–South research partnerships and targeted CGIAR investment in DT technology are identified as priority actions for closing the SSA evidence gap.","撒哈拉以南非洲（SSA）尽管拥有全球约60%的未开垦耕地，却长期面临粮食不安全问题。田间试验虽能提供实证，但成本高昂且规模不足以应对日益加剧的气候与人口压力。数字孪生（DT）技术，定义为利用实时数据对物理系统进行持续、双向的虚拟复制，可支持农业食品系统的监测、建模与优化。然而，据我们所知，尚无已发表的综述从SSA粮食系统或小农生产现实的角度审视DT农业研究。本研究遵循PRISMA规范，在书目数据库中采用预定义的布尔检索和适应性PICOS资格标准进行系统综述，纳入2019至2025年间发表的40篇同行评审研究。结构化提取涵盖文献计量、地理、方法、技术及技术就绪水平（TRL）数据，并采用频次制表和主题聚类进行分析。DT农业出版物在2019至2023年间约增长两倍（从三篇增至十篇），但57.5%的研究仍处于TRL 2–3级。欧洲和东亚主导了证据基础；没有一项研究由SSA机构牵头，也没有研究报告在SSA地区针对玉米、木薯、高粱或豇豆等优先作物进行部署。物联网（IoT）传感器基础设施（92.5%）和人工智能\u002F机器学习（AI\u002FML）算法（52.5%）构成主导技术栈。研究识别出五个主题聚类：智慧农业基础设施、精准水肥管理、受控环境农业、农业食品供应链以及畜牧和水产养殖系统。政府资助的研究往往比学术资助的研究达到更高的TRL评分。将这些发现综合为系统层面、面向部署的分析，我们提出了一个概念性的SSA-DT框架。该框架结合离线优先的边缘计算架构、混合作物建模和多利益相关方接口，以膳食多样性和气候韧性作为两项关键绩效指标（KPI），与国际热带农业研究所（IITA）2024–2030年战略保持一致。南北研究伙伴关系和CGIAR对DT技术的定向投资被确定为缩小SSA证据差距的优先行动。","Smart Agricultural Technology","2026-09-01T00:00:00Z",79,{"impact":18,"substance":201,"depth":277,"authority":126,"freshness":127,"relevant":22,"comment":278},19,"系统综述数字孪生技术在撒哈拉以南非洲的应用，揭示研究空白并提出框架，对全球农业信息化有参考价值。",[280],{"name":273,"url":270},[27,282,29,83],"非洲",{"code":33,"slug":34,"name":35,"path":36,"parent_code":37},[285,290],{"code":157,"slug":286,"name":287,"path":288,"parent_code":289},"climate","气候适应与防灾减灾","\u002Ftopic\u002Fenvironment\u002Fclimate","H",{"code":46,"slug":47,"name":48,"path":49,"parent_code":50},{"status":52,"locked":14,"version":53,"method":54,"evidence":292},{"candidates":293,"agricultural_evidence":59},[294,299,304,308,313,320,324,331,338,345,352],{"code":79,"score":277,"supported":59,"title":295,"abstract":297,"tags":298},[296],"digital twin",[27,83,296],[83,27],{"code":157,"score":58,"supported":59,"title":300,"abstract":302,"tags":303},[301],"climate resilience",[301],[],{"code":33,"score":73,"supported":59,"title":305,"abstract":306,"tags":307},[75],[],[29],{"code":46,"score":73,"supported":59,"title":309,"abstract":310,"tags":312},[],[68,311,69],"machine learning",[],{"code":246,"score":73,"supported":59,"title":314,"abstract":315,"tags":319},[],[249,316,317,318],"边缘计算","internet of things","IoT",[],{"code":214,"score":20,"supported":59,"title":321,"abstract":322,"tags":323},[],[216,215],[],{"code":325,"score":20,"supported":59,"title":326,"abstract":327,"tags":330},"A5",[],[328,329],"高粱","sorghum",[],{"code":332,"score":20,"supported":59,"title":333,"abstract":334,"tags":337},"C",[],[335,336],"畜牧","livestock",[],{"code":339,"score":20,"supported":59,"title":340,"abstract":341,"tags":344},"D",[],[342,343],"水产","aquaculture",[],{"code":346,"score":20,"supported":59,"title":347,"abstract":348,"tags":351},"I7",[],[349,350],"文献计量","bibliometric",[],{"code":353,"score":87,"supported":59,"title":354,"abstract":355,"tags":357},"G2",[],[356],"smallholder",[],[359,360],"数字孪生 智慧农业 粮食安全 非洲","数字孪生 智慧农业","数字孪生智慧农业粮食安全非洲-1380","10.1016\u002Fj.atech.2026.102532",{"doi":362,"openalex_id":364,"authors":365,"venue":273,"cited_by_count":102,"oa_url":385,"card":386,"direction":390,"ingested_from":111},"W7204898011",[366,369,371,374,376,379,381,383],{"name":367,"orcid":368},"Henri E. Z. Tonnang","https:\u002F\u002Forcid.org\u002F0000-0002-9424-9186",{"name":370,"orcid":9},"Francis Chianu",{"name":372,"orcid":373},"Siyabusa Mkuhlani","https:\u002F\u002Forcid.org\u002F0000-0001-7872-2036",{"name":375,"orcid":9},"Francis Muthoni",{"name":377,"orcid":378},"John Choptiany","https:\u002F\u002Forcid.org\u002F0009-0008-9961-4606",{"name":380,"orcid":9},"Franck B.N. Tonle",{"name":382,"orcid":9},"Bonoukpoe M. Sokame",{"name":384,"orcid":9},"Mercy Lung’aho","https:\u002F\u002Fwww.sciencedirect.com\u002Fscience\u002Farticle\u002Fpii\u002FS2772375526007574\u002Fpdf",{"tldr":387,"method":388,"finding":389,"direction":390,"opportunity":391},"系统综述数字孪生技术在撒哈拉以南非洲农业的应用，提出SSA-DT框架。","PRISMA系统综述，检索2019-2025年40篇研究，频率统计和主题聚类。","DT农业研究增长但TRL低，SSA无主导研究，提出离线优先边缘计算框架。","数字乡村与农业信息化","SSA地区数字孪生农业研究空白，可探索离线优先架构与本地作物模型结合。","2026-09-02T23:30:04.414055Z",{"id":394,"title":395,"url":396,"summary":397,"summary_zh":9,"content":9,"source_name":398,"source_url":396,"published_at":399,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":400,"score_detail":401,"sources":403,"tags":405,"primary_topic":410,"secondary_topics":411,"classification":418,"search_phrases":442,"slug":445,"view_count":87,"doi":446,"paper":447,"created_at":458},2299,"Causal Effects of Climate-Smart Agriculture Adoption on Smallholder Welfare and Food Security in Zambia: Evidence from Endogenous Switching Regression, Double Machine Learning, and Bayesian Estimation","https:\u002F\u002Fdoi.org\u002F10.21203\u002Frs.3.rs-10994635\u002Fv1","Causal Effects of Climate-Smart Agriculture Adoption on Smallholder Welfare and Food Security in Zambia: Evidence from Endogenous Switching Regression, Double Machine Learning, and Bayesian Estimation。Research Square","Research Square","2026-09-11T00:00:00Z",67,{"impact":17,"substance":125,"depth":18,"authority":21,"freshness":73,"relevant":22,"comment":402},"方法组合新颖、结论有实证价值，但研究对象为赞比亚小农户，对国内三农实践的直接参考有限，适合作为国际经验类精选。",[404],{"name":398,"url":396},[29,406,407,408,409],"因果推断","气候智慧农业","小农户生计","赞比亚",{"code":33,"slug":34,"name":35,"path":36,"parent_code":37},[412,417],{"code":353,"slug":413,"name":414,"path":415,"parent_code":416},"farm-services","社会化服务与规模经营","\u002Ftopic\u002Frural-services\u002Ffarm-services","G",{"code":46,"slug":47,"name":48,"path":49,"parent_code":50},{"status":52,"locked":14,"version":53,"method":54,"evidence":419},{"candidates":420,"agricultural_evidence":59},[421,425,429,434,438],{"code":33,"score":17,"supported":59,"title":422,"abstract":423,"tags":424},[75],[75],[29],{"code":353,"score":58,"supported":59,"title":426,"abstract":427,"tags":428},[356],[356],[],{"code":79,"score":58,"supported":59,"title":430,"abstract":432,"tags":433},[431],"smart agriculture",[431],[],{"code":46,"score":58,"supported":59,"title":435,"abstract":436,"tags":437},[311],[311],[],{"code":157,"score":22,"supported":14,"title":439,"abstract":440,"tags":441},[],[],[407],[443,444],"气候智慧农业 小农户生计 因果推断 粮食安全","气候智慧农业 小农户生计","气候智慧农业小农户生计因果推断粮食安全-2299","10.21203\u002Frs.3.rs-10994635\u002Fv1",{"doi":446,"openalex_id":448,"authors":449,"venue":398,"cited_by_count":102,"oa_url":396,"card":9,"direction":110,"ingested_from":111},"W7212246566",[450,453,455],{"name":451,"orcid":452},"Evans Kudzai Chinembiri","https:\u002F\u002Forcid.org\u002F0009-0003-7304-0214",{"name":454,"orcid":9},"Munyaradzi Mutenje",{"name":456,"orcid":457},"Nkele D. Malahlela","https:\u002F\u002Forcid.org\u002F0009-0001-1653-7553","2026-09-13T23:30:09.568432Z",{"id":460,"title":461,"url":462,"summary":463,"summary_zh":464,"content":9,"source_name":465,"source_url":462,"published_at":466,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":467,"score_detail":468,"sources":470,"tags":472,"primary_topic":475,"secondary_topics":476,"classification":477,"search_phrases":502,"slug":505,"view_count":102,"doi":506,"paper":507,"created_at":518},5588,"Certified Temporal Fusion of Artificial Intelligence Climate Analytics for Crop Failure Early Warning Under Sparse Agricultural Sensors: Error Bounds and Famine Relief Targeting","https:\u002F\u002Fdoi.org\u002F10.5281\u002Fzenodo.23545489","Sparse agricultural sensor networks across sub-Saharan Africa leave crop-failure early-warning systems with too little ground observation to support reliable famine relief targeting. The framework treats sensor sparsity as a missing-data problem with a known observation operator and propagates climate-analytics uncertainty through a temporal Kalman-Buys filter whose gain is constrained by a certified risk envelope. Using NASA POWER monthly reanalysis for Accra, Ghana (2000-2023, 312 observations), the analysis establishes a mean temperature of 26.521 C and a warming trend of 0.1622 C per decade, which calibrates the climate-analytics prior. The derived bounds show that relief-targeting error scales with the square root of sensor density and with the Lipschitz constant of the crop-response function, yielding an explicit sensor-placement rule for a target misclassification rate. The contribution is a certified, assumption-explicit bridge between climate analytics and famine relief targeting that states the conditions under which early warning is trustworthy.","撒哈拉以南非洲稀疏的农业传感器网络使得作物歉收预警系统缺乏足够的地面观测数据，难以支撑可靠的饥荒救援目标定位。该框架将传感器稀疏性视为具有已知观测算子的缺失数据问题，并通过时间Kalman-Buys滤波器传播气候分析不确定性，该滤波器的增益受经认证的风险包络约束。利用NASA POWER对加纳阿克拉的月度再分析数据（2000—2023年，312个观测值），分析确定了平均气温为26.521°C、每十年0.1622°C的增暖趋势，据此校准了气候分析先验。推导出的界限表明，救援目标定位误差随传感器密度的平方根以及作物响应函数的Lipschitz常数缩放，从而为目标误分类率给出了明确的传感器布设规则。该研究的贡献在于构建了一个经认证的、假设明确的气候分析与饥荒救援目标定位之间的桥梁，阐明了早期预警可信赖的条件。","Zenodo (CERN European Organization for Nuclear Research)","2026-10-21T00:00:00Z",74,{"impact":17,"substance":201,"depth":18,"authority":17,"freshness":13,"relevant":22,"comment":469},"该研究提出一种带误差界认证的时序融合方法，用于稀疏传感器条件下的作物歉收预警与饥荒救援瞄准，方法新颖、结论明确，但案例局限于加纳阿克拉，产业影响有限，适合作为技术前沿条目收录。",[471],{"name":465,"url":462},[27,28,29,473,474],"遥感","气候分析",{"code":46,"slug":47,"name":48,"path":49,"parent_code":50},[],{"status":52,"locked":14,"version":53,"method":54,"evidence":478},{"candidates":479,"agricultural_evidence":59},[480,484,489,493,497],{"code":46,"score":73,"supported":59,"title":481,"abstract":482,"tags":483},[69],[],[28],{"code":246,"score":87,"supported":59,"title":485,"abstract":486,"tags":488},[],[487],"农业传感",[],{"code":33,"score":22,"supported":14,"title":490,"abstract":491,"tags":492},[],[],[29],{"code":79,"score":22,"supported":14,"title":494,"abstract":495,"tags":496},[],[],[27],{"code":498,"score":22,"supported":14,"title":499,"abstract":500,"tags":501},"I3",[],[],[473],[503,504],"NASA POWER 加纳 气候分析","作物歉收 早期预警 传感器","NASAPOWER加纳气候分析-5588","10.5281\u002Fzenodo.23545489",{"doi":506,"openalex_id":508,"authors":509,"venue":465,"cited_by_count":102,"oa_url":462,"card":512,"direction":108,"ingested_from":111},"W7223111766",[510],{"name":511,"orcid":9},"Fatima-Zahra El Khattabi",{"tldr":513,"method":514,"finding":515,"direction":516,"opportunity":517},"提出带误差界的时间融合框架，在稀疏传感器下用气候分析预警作物歉收。","卡尔曼-布西滤波+NASA POWER再分析数据，风险包络约束增益。","救济瞄准误差随传感器密度平方根和作物响应Lipschitz常数缩放，给出布点规则。","农业遥感与作物表型","可将该认证误差界框架扩展到多作物多区域，结合卫星遥感弥补地面传感器稀疏。","2026-10-22T11:21:02.682497Z",{"id":520,"title":521,"url":522,"summary":523,"summary_zh":524,"content":9,"source_name":525,"source_url":522,"published_at":526,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":527,"score_detail":528,"sources":531,"tags":533,"primary_topic":535,"secondary_topics":536,"classification":537,"search_phrases":578,"slug":581,"view_count":87,"doi":582,"paper":583,"created_at":604},4321,"Improving early season crop type mapping using machine learning algorithms with combined SAR and optical time series","https:\u002F\u002Fdoi.org\u002F10.1007\u002Fs43621-026-04839-6","Timely and reliable crop type mapping is essential for sustainable agricultural management and food security. This study investigates the potential of integrating multi-temporal Synthetic Aperture Radar (SAR) data from Sentinel-1 and optical imagery from Sentinel-2 for early-season crop classification in Udham Singh Nagar district, Uttarakhand, India. Leveraging the Google Earth Engine (GEE) cloud platform, two ensemble machine learning algorithms, Random Forest (RF) and Gradient Boosted Trees (GBT), were implemented and evaluated for their ability to classify major land cover and crop types using combined SAR-optical datasets acquired during November–December 2022. A total of 1141 ground truth samples were collected through field surveys. The integration of structural (SAR) and spectral (optical) information substantially enhanced classification performance compared with single-sensor approaches. Among the models, GBT achieved the highest accuracy, with an overall accuracy of 93% and a Kappa coefficient of 0.90. Wheat, forest, and water bodies were mapped with high precision (F1 ≥ 0.90), whereas mustard and sugarcane showed lower accuracy due to spectral similarity and limited training data. Feature importance analysis identified late-season SAR (VV, VH) and optical (red-edge and SWIR) bands as the most influential variables. The results demonstrate that the synergy between SAR and optical time-series data, coupled with ensemble machine learning, provides a robust and scalable framework for near real-time crop monitoring. This approach offers valuable insights for precision agriculture and supports data-driven policy planning in India’s agricultural sector.","及时可靠的作物类型制图对可持续农业管理和粮食安全至关重要。本研究探讨了整合Sentinel-1多时相合成孔径雷达（SAR）数据与Sentinel-2光学影像在印度北阿坎德邦乌达姆辛格纳加尔地区进行早季作物分类的潜力。依托Google Earth Engine（GEE）云平台，研究采用随机森林（RF）和梯度提升树（GBT）两种集成机器学习算法，利用2022年11月至12月获取的SAR-光学联合数据集，对主要土地覆盖和作物类型进行分类，并评估其分类能力。通过实地调查共采集1141个地面真实样本。与单一传感器方法相比，结构信息（SAR）与光谱信息（光学）的融合显著提升了分类性能。在各模型中，GBT取得了最高精度，总体精度达93%，Kappa系数为0.90。小麦、森林和水体的制图精度较高（F1 ≥ 0.90），而芥菜和甘蔗因光谱相似及训练数据有限，精度较低。特征重要性分析表明，晚季SAR（VV、VH）和光学（红边和短波红外）波段是最具影响力的变量。结果表明，SAR与光学时间序列数据的协同结合集成机器学习，为近实时作物监测提供了稳健且可扩展的框架。该方法为精准农业提供了有价值的见解，并支持印度农业领域的数据驱动政策规划。","Discover Sustainability","2026-10-05T00:00:00Z",77,{"impact":19,"substance":529,"depth":18,"authority":126,"freshness":73,"relevant":22,"comment":530},21,"该研究将SAR与光学时序数据结合集成机器学习，实现93%早期作物分类精度，方法可扩展、结论可靠，对农业遥感监测有实质参考价值。",[532],{"name":525,"url":522},[27,28,29,473,534],"作物分类",{"code":46,"slug":47,"name":48,"path":49,"parent_code":50},[],{"status":52,"locked":14,"version":53,"method":54,"evidence":538},{"candidates":539,"agricultural_evidence":59},[540,544,548,554,561,568,574],{"code":46,"score":141,"supported":59,"title":541,"abstract":542,"tags":543},[311],[68,311],[28],{"code":33,"score":127,"supported":59,"title":545,"abstract":546,"tags":547},[],[29,75],[29],{"code":79,"score":127,"supported":59,"title":549,"abstract":550,"tags":553},[],[551,552],"精准农业","precision agriculture",[27],{"code":555,"score":20,"supported":59,"title":556,"abstract":557,"tags":560},"A2",[],[558,559],"小麦","wheat",[],{"code":562,"score":20,"supported":59,"title":563,"abstract":564,"tags":567},"A6",[],[565,566],"甘蔗","sugarcane",[],{"code":569,"score":87,"supported":59,"title":570,"abstract":571,"tags":573},"H8",[],[572],"可持续农业",[],{"code":498,"score":171,"supported":14,"title":575,"abstract":576,"tags":577},[],[],[534,473],[579,580],"Sentinel-1 Sentinel-2 作物分类","Google Earth Engine 机器学习 作物制图","Sentinel-1Sentinel-2作物分类-4321","10.1007\u002Fs43621-026-04839-6",{"doi":582,"openalex_id":584,"authors":585,"venue":525,"cited_by_count":102,"oa_url":522,"card":599,"direction":108,"ingested_from":111},"W7219792303",[586,589,592,594,597],{"name":587,"orcid":588},"Rajeev Ranjan Kumar","https:\u002F\u002Forcid.org\u002F0000-0002-2670-6189",{"name":590,"orcid":591},"Jayanta Das","https:\u002F\u002Forcid.org\u002F0000-0003-0995-9114",{"name":593,"orcid":9},"Jadhav Rajaram Anna",{"name":595,"orcid":596},"Abhishek Danodia","https:\u002F\u002Forcid.org\u002F0000-0002-5724-8790",{"name":598,"orcid":9},"Dipanwita Haldar",{"tldr":600,"method":601,"finding":602,"direction":516,"opportunity":603},"融合Sentinel-1 SAR与Sentinel-2光学时序，用集成机器学习实现印度早季作物类型高","GEE平台，RF与GBT集成学习，1141个地面样本，SAR-光学多时相数据。","GBT精度最高（总体93%，Kappa 0.90），SAR与光学结合显著优于单传感器。","可探索更早物候期、小样本作物（如芥菜、甘蔗）的SAR-光学特征优化与迁移学习。","2026-10-06T23:30:38.288174Z"]