Semiconductor: Demand Planning and Supply Chain Coordination in the Contract Manufacturing Environments

这笔拨款支持开发新的分析工具，用于半导体需求规划和供应链协调，重点是新兴的合同制造平台。本研究的目标是：(1)建立数学模型，描述技术生命周期短的产品的需求驱动因素；(2)创建经济模型，描述合同制造环境中的激励机制；(3)构建计算模型，将理论见解转化为运营决策工具；(4)进行广泛的行业测试，验证分析结果。本研究的一个主要动机是，目前半导体行业的需求和产能管理实践是在具有长技术生命周期和稳定需求模式的垂直整合企业的背景下发展起来的。虽然该行业正朝着日益复杂的供应链合作伙伴关系、加速的技术创新和快速的产品引入周期发展，但这些变化对运营的影响尚未得到很好的理解。这项研究提供了创新的模型，增强了我们对这种动态变化的环境的理解。如果成功，本研究将从两个重要的角度做出贡献：1。通过缩短技术生命周期有效地描述需求。技术生命周期的缩短对所有需求规划系统的基础时间序列预测方法提出了重大挑战。本研究考察了基于领先指标和生命周期增长模型的需求表征方法，而不依赖于时间序列模型所需的大量历史数据。2. 复杂供应链伙伴关系中的有效供需计划。大公司正在远离垂直集成，而他们的产品实现是通过建立在外包和伙伴关系上的供应链结构来实现的。本文提出了一种针对新兴代工平台的供需规划框架；在这种情况下，需求和产能的不确定性都被明确地建模，而供需计划决策与合同、采购和产能配置决策集成在一起。

This grant supports the development of new analytical tools for semiconductor demand planning and supply chain coordination, focusing on the emerging platform of contract manufacturing. The objectives of the research are to (1) develop mathematical models that characterize demand drivers for products with short technology lifecycles, (2) create economic models that characterize incentives in contract manufacturing environments, (3) construct computing models that translate theoretical insights into operational decision tools, and (4) conduct extensive industry testing that validates the analytical developments. A main motivation of this research is that current demand and capacity management practices in the semiconductor industry were developed in the context of vertically integrated enterprises with long technology lifecycles and stable demand patterns. While the industry is moving toward increasingly complex supply chain partnerships, accelerated technology innovations, and rapid product introduction cycles, the operational implications of these changes are not well understood. This research offers innovative models that enhance our understanding of this dynamically changing environment. If successful, this research will make contributions from two important perspectives: 1. Effective Characterization of Demands with Shortening Technology Lifecycles. The shortened technological lifecycle poses major challenges to the time-series forecasting methods fundamental to all demand planning systems. This research examines demand characterization methods based on leading indicators and lifecycle growth models without relying on the sizable historical data required by time-series models. 2. Effective Supply-Demand Planning Within Complex Supply Chain Partnerships. Major corporations are moving away from vertical integration, while their product realization is being carried out through a supply chain structure built on outsourcing and partnerships. This research proposes a supply-demand planning framework designed for the emerging platform of contract manufacturing; in this setting, the uncertainty in both demand and capacity that is prevalent is modeled explicitly, while supply-demand planning decisions are integrated with contracting, sourcing, and capacity configuration decisions.