告别经验主义： 在人类算力的极限之外，重建供应链秩序。 在某个普通的早上，南宁某仓的调度大屏跳出来自八个区域仓的300多条配送请求—— 有医院的急救类补货、夜间药店的销量回补、B2B渠道的大批量集采订单，以及由季节性疾病高发触发的突发需求，也有必须关乎生命的、要在两小时送 达ICU的药品需求，每一条请求背后都对应着不同的时效要求、药品合规要求和路线约束。 另一边，仓内的拣货员已经开始处理医院端的大单：一个订单可能包含80种以上的品规，既有高价值药，也有一旦超温就要整批报废的冷链品类。这些药 必须从不同区域的货架拣取，经过核对批次、效期、规格，再以合规方式打包。如果流程中断或错拣，订单需要整单复核。 过去，调度员需要在路线、车型、载重和时效之间来回推演。一遇到跨仓调拨或需求波动——比如某区域突然感冒高发，导致当地仓库缺货——所有路线 都必须重新排。因此，一个人要处理上万SKU与各仓之间的动态关系，往往一排车就要一小时以上。 这是广西柳药集团的日常缩影。 这家区域龙头医药流通企业成立于上世纪50年代，是广西最早的国有医药公司之一。经过数十年的区域扩张与品类延展，柳药已从传统医药流通企业，成 长为覆盖医院、连锁药店 ...

Core Viewpoint - The article highlights the transformation of the pharmaceutical supply chain through AI integration, emphasizing the shift from traditional experience-based methods to data-driven, intelligent decision-making systems [2][11][36]. Group 1: Company Overview - Liuyao Group, established in the 1950s, has evolved from a traditional pharmaceutical distributor to a comprehensive health service group, covering hospitals, retail pharmacies, and B2B clients [2]. - The complexity of Liuyao's supply chain is amplified by the need to manage over ten thousand SKUs, multiple warehouses, and stringent compliance and time constraints [2][4]. Group 2: Supply Chain Challenges - Liuyao faces a "triple constraint" in its supply chain, balancing time, compliance, and cost, where improving one aspect can exacerbate the others [4][5]. - The pharmaceutical industry is under pressure to enhance efficiency and reduce costs due to increasing regulatory demands and market competition [7][10]. Group 3: AI Integration and Transformation - Liuyao has partnered with Huawei Cloud to leverage AI for restructuring its supply chain decision-making processes, focusing on data governance, demand forecasting, and intelligent scheduling [2][11]. - The implementation of a data lake has unified fragmented data, enabling real-time visibility and optimization of supply chain operations [15][21]. Group 4: AI Tools and Their Impact - The Pangu predictive model has improved demand forecasting accuracy to over 89%, directly impacting inventory management and reducing stockout risks [16][21]. - The Tianchou AI solver optimizes complex decision-making scenarios, significantly reducing decision-making time and lowering costs by approximately 20% [21][20]. Group 5: Industry Trends and Future Directions - The article notes a global trend where over 50% of large multinational companies are expected to adopt AI and advanced analytics for supply chain management by 2027 [8]. - In China, over 60% of large enterprises are projected to implement AI and intelligent scheduling systems in their supply chains within the next three years, driven by national policies promoting digital transformation [10][11]. Group 6: Conclusion on Supply Chain Evolution - The shift from experience-based systems to computational systems in supply chains is seen as a critical evolution, enabling companies to predict demand, optimize resources, and enhance operational efficiency [26][36]. - Liuyao's experience serves as a model for the industry, demonstrating that intelligent supply chains can become a new growth engine rather than merely a cost center [36].