PFI (MCA): Price and Lead Time Quotation for Make-to-Order Firms with Contingent Demand and Demand Learning

这个创新伙伴关系-中期职业发展（PFI（MCA））项目的更广泛的影响/商业潜力是加强按订单生产（MTO）、定制服务和医疗保健提供组织的运营和客户服务。 该研究将侧重于开发和实现决策支持工具的商业化，以生成实时、可操作的定价和交货期报价。医疗保健提供组织的成功，其中许多是中小型企业，是复杂的，由于不确定性。 如果一家公司投标过于激进，他们往往会遇到需求过剩，导致服务质量差，并因未履行提案条款而受到处罚。如果一家公司过于谨慎地提出建议，他们就会变得没有竞争力，导致业务损失和产能利用不足。这项研究将建立一个通用的决策支持框架，这将在工业合作伙伴的销售和运营环境的背景下进行验证。该项目直接有助于本科生和研究生的招聘，教育和STEM培训，领导力，创新，并通过参与和推广活动的创业精神。代表性不足的学生将通过利用与充满活力的行业合作伙伴和各种支持计划的合作来招募。 该项目旨在开发决策支持工具，将需求学习和收入管理技术相结合，同时明确考虑“应急需求”的存在。处理潜在的不确定性与偶然的工作，这可能会有很大的不同，在他们的要求，需要强大的方法建立在学习客户行为和随机系统性能的指数可能实现的需求积压。实时报价到达工件存在或有需求的文献中没有得到太多的关注。在运营决策中平衡勘探和开发的优化方法的发展是最近的，并在此背景下提出了额外的挑战。该问题将以数据科学和运筹学技术为基础，以全面的方式处理，以优化系统在一段时间内的性能。将研究各种客户行为函数和特殊情况，以更深入地了解受偶然需求影响的系统的理论特性，并利用它们来解决一般情况。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。

The broader impact/commercial potential of this Partnerships for Innovation – Mid-Career Advancement (PFI(MCA)) project is to enhance the operations and customer service of make-to-order (MTO), customized service, and healthcare delivery organizations. The research will focus on developing and enabling the commercialization of decision support tools to generate real-time, operationally attainable pricing and lead-time quotes. The success of healthcare delivery organizations, many of which are small-to-medium size enterprises, is complex due to uncertainty. If a firm bids too aggressively, they often experience a demand surplus, resulting in poor service and penalties from unfulfilled proposal terms. If a firm proposes too cautiously, they become noncompetitive, resulting in business loss and under-utilization of capacity. This research will build a general decision support framework, which will be validated in the context of the industrial partner’s sales and operations environment. The project contributes directly to undergraduate and graduate recruitment, education and training in STEM, leadership, innovation, and entrepreneurship through participation and outreach activities. Underrepresented students will be recruited by leveraging the pull of working with dynamic industry partners and various supporting programs. This project seeks to develop decision support tools that integrate demand learning and revenue management techniques while explicitly considering the presence of ‘contingent demand.’ Dealing with the underlying uncertainty associated with contingent jobs, which may vary wildly in their requirements, necessitates robust approaches built around learning customer behavior and the stochastic system performance resulting from the exponential number of possible realizations of the demand backlog. The real-time quoting of arriving jobs in the presence of contingent demand has not received much attention in the literature. The development of optimization approaches that balances exploration and exploitation in operational decision making is recent and presents additional challenges in this context. The problem will be approached in a comprehensive fashion building on data science and operations research techniques to optimize system performance over a time horizon. Various customer behavior functions and special cases will be studied to gain a deeper understanding of the theoretical properties of systems subject to contingent demand and leverage them to address general cases. An online simulation platform will be built for practical evaluation and performance comparison of various approaches.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.