Matching Supply and Demand Through Dual-Sourcing

随着企业将其供应链扩展到全球范围，供需匹配变得越来越具有挑战性。许多拥有全球供应网络的跨国公司的一个主要担忧是供应的不确定性，特别是供应中断。另一个大问题是由于较长的交货时间而降低了响应速度。在实践中，双重采购作为一种减轻供应风险和在成本和响应能力之间进行权衡的有效战略，已得到普遍实施。这种做法需要先进的双源模型，同时将供应的不确定性和大的补货提前期，并制定和分析有效的库存战略。这项研究的目的是提供一个复杂的双重采购系统的全面了解，以及提供新的见解和分析工具，以处理这种复杂程度。该研究有可能提供新的工具和方法来有效地管理价值链系统，从而提高公司的竞争优势。此外，该项目还将为研究生和本科生提供教育机会，并有可能通过伊利诺伊大学的研究生教育多元文化工程招聘项目（MERGE）扩大代表性不足的群体和少数民族的参与。如果成功，该项目的结果将对双源库存优化的理论和实践产生重大影响。首先，本研究将发展新的转换技术，将非凸最小化问题转换为适用于复杂双源模型的等价凸最小化问题。其次，本研究将为实际应用中广泛使用的几种算法提供理论支持，设计新的有效算法，并探索在本项目主持下开发的先进方法在其他问题中的潜在应用。

It has become increasingly challenging to match supply and demand as firms have extended their supply chains to a global scale. One major concern for many multi-national firms with global supply networks is supply uncertainty, especially supply disruption. Another big concern is the reduced responsiveness due to longer lead times. Dual-sourcing has been commonly implemented in practice as an effective strategy to mitigate supply risks and provide a tradeoff between cost and responsiveness. This practice calls for advanced dual-sourcing models that incorporate supply uncertainty and large replenishment lead times simultaneously, and the development and analysis of effective inventory strategies. The objective of this research is to provide a comprehensive understanding of complex dual-sourcing systems as well as provide new insights and analytical tools to handle this level of complexity. The research has the potential to provide new tools and methodologies to effectively manage value chain systems and thus improve companies' competitive advantage. In addition, it will provide educational opportunities for graduate and undergraduate students, and has the potential to broaden participation of underrepresented groups and minorities through the Multi-Cultural Engineering Recruitment for Graduate Education Program (MERGE) at the University of Illinois.If successful, the results of this project will have a significant impact on the theory and practice of dual-sourcing inventory optimization. First, the research will develop novel transformation techniques to convert non-convex minimization problems to equivalent convex minimization problems applicable to complex dual-sourcing models. Second, the research will, provide theoretic support of several heuristics widely used in practice, design new efficient algorithms, and explore potential applications of the advanced methodologies developed under the auspices of this project in other problems.