Building a pathway of GenAI-enabled capabilities for supply chain management
本研究基于任务技术适配视角，通过系统文献综述和模糊德尔菲法，构建了一个四阶段能力框架，帮助供应链管理者优先发展GenAI赋能能力以创造持续价值。
The integration of Generative Artificial Intelligence (GenAI) into Supply Chain Management (SCM) has accelerated rapidly. However, limited understanding exists on how GenAI-enabled capabilities should be prioritised to create sustained value. Existing research predominantly describes applications but overlooks the hierarchical structure of underlying capabilities required for effective adoption. In response, this study develops a capability-oriented framework grounded in the Task Technology Fit (TTF) perspective. A Systematic Literature Review identified capabilities, which were refined via Fuzzy Delphi and structured using Interpretive Structural Modeling (ISM) with Fuzzy MICMAC. The resulting framework was corroborated through secondary case analyses of DHL Supply Chain and Walmart, generating empirically derived propositions regarding adoption mechanisms. Findings reveal a four-stage progression from data consolidation to operational intelligence, adaptability, and differentiation. Real-time data integration serves as the enabling factor, supporting intermediate automation capabilities, while adaptability and differentiation emerge as dependent outcomes that represent the strategic value frontier. The study extends TTF by offering a capability-mediated perspective that provides a prescriptive roadmap for prioritising capability building and guiding the design of GenAI-based systems in supply chains.