Novel Modeling and Analytical Approaches for Two-stage Assembly and Related Lot Streaming Problems

该奖项通过调查支持产品定制的装配操作中生产调度的改进来提高国家的竞争力。 增强的定制要求生产各种产品和小批量。因此，制造商必须对这种需求做出反应，同时保持或降低制造成本并保持在全球市场上的竞争力。本项目通过两阶段装配系统研究批量流的使用，该系统捕获了大规模定制中的基本操作控制问题，大规模定制是一种重要的生产模式，已出现在工业中，以实现更低的成本，更短的周期时间和更大的产品种类。两阶段装配促进了推迟最终装配以区分产品直到最后一步的想法。这种方法也直接适用于其他两个密切相关的制造领域，即敏捷制造和混合流水车间。该项目将结合建模，解决方案策略和计算方法，以确定这些生产场景中的最佳批量和时间表。 该项目将教育和培训学生，使他们能够解决对美国经济至关重要的制造业中出现的复杂问题。研究结果将被纳入研究生和本科生的课程，从而使其他学生也受益。 两阶段装配批量流问题（TSA-LSP）具有独特的结构，需要专门的方法，目前缺乏的文献，利用其特殊的结构，并提出有效的解决方案。该项目有四个主要目标，即（1）制定一个包含其特点的数学模型;（2）全面的多面体分析，增强模型，并开发一种新的基于分解的方法，以确定最佳/接近最佳的子批次数量、子批次大小和排序决定;（3）提出了一种新的装配批量流问题（ALSP）的求解方法，该问题作为TSA-LSP中的子结构，包含装卸和完工时间成本;（4）将批量流的应用推广到新的装配系统领域，包括陆军可持续准备模型调度问题（ASR-MSP），并通过发现新的见解和开发有用的算法，为批量流的知识库做出贡献。ASR-MSP构成了TSA-LSP的一个重要应用，并将作为一个案例研究来验证所开发的方法。 从ALSP调查中获得的见解本身就很重要，也有助于增强TSA-LSP解决其他具体问题的解决方法。该奖项反映了NSF的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

This award enhances the Nation's competitiveness by investigating improvements to production scheduling in assembly operations that support product customization. Enhanced customization demands production of a variety of products and in smaller batches. Therefore, it is imperative for manufacturers to respond to such a demand while maintaining or reducing manufacturing costs and remaining competitive in the global market place. This project investigates the use of lot streaming via a two-stage assembly system that captures the essential operational control issues in mass customization, an important production mode that has emerged for industry to achieve lower cost, shorter cycle times, and a greater product variety. Two-stage assembly promotes the idea of postponing final assembly to differentiate between products until the last step. This approach is also directly applicable to two other closely-related manufacturing areas, namely, agile manufacturing and hybrid flow shops. This project will incorporate modeling, solution strategies, and computational approaches to determine optimal lot sizes and schedules in these production scenarios. The project will educate and train students to be able to tackle complex problems arising in the manufacturing critical to the U.S. economy. The results of the research will be integrated into the courses at the graduate and undergraduate level, thereby benefitting the other students as well. The two-stage assembly lot streaming problem (TSA-LSP) has a unique structure that requires specialized methodology presently lacking in the literature that exploits its special structure and renders effective solution approaches. This project has four main thrusts, namely (1) formulation of a mathematical model incorporating its special features; (2) a comprehensive polyhedral analysis, enhancement of the model, and development of a new decomposition-based methodology to determine optimal/near-optimal number of sublots, sublot sizes and sequencing decisions; (3) a novel approach for the solution of an assembly lot streaming problem (ALSP) with handling and makespan costs that arise as a substructure within the TSA-LSP; (4) advancing the use of lot streaming to a new domain of assembly systems including the Army Sustainable Readiness Model Scheduling Problem (ASR-MSP), and contributing to the knowledge-base of lot streaming by discovering new insights and developing useful algorithms. The ASR-MSP constitutes an important application of the TSA-LSP and will serve as a case study to validate the methods developed. Insights gained from an investigation of the ALSP are important on their own right as well as serving to enhance the solution methodology for the TSA-LSP for solution of other concrete instances.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.