Collaborative Research: Risk Management and Control of Dynamic Rate Queues for Healthcare and Transportation Services

这一合作研究项目解决了两个关键的国家挑战：(1)降低医疗保健成本。(2)可持续交通。这两个问题都是通过利用排队或排队的通用建模框架来解决的，这些排队或排队对一天中的时间、一周中的某一天、季节性或年度影响很敏感。在医疗保健领域，可以开发疗养院床位规划问题的解决方案。由于婴儿潮一代人口老龄化，这是一个重要的问题。此外，通过寻找最有效的时间配备清洁人员，可以降低医院床位清洁成本。寻找停车位的司机会产生大量的污染和拥堵。在可持续交通的推动下，在大数据集的推动下，可以开发出减少停车造成的污染和拥堵的方法。这项研究还包括对研究生的培训和纳入代表性不足的少数群体。非裔美国人数学科学研究人员会议(CAARMS)成立于1995年，目的是培养和促进非洲裔美国人研究人员在数学科学领域的发展。两位PI都在CAARMS上做了演讲和教程，与非裔美国数学家和有抱负的学生社区分享他们的一些最新研究成果。这个项目开发了基于一类丰富的时间依赖随机网络模型的动态性能模型和动态优化求解技术，称为马尔可夫服务网络。这种建模和优化框架捕捉了系统的随机动态，而这种随机动态是在静态、长期、稳定状态行为中找不到的。使用这一框架，可以开发出应对医疗保健和可持续交通领域关键挑战的解决方案。这些解决方案解决了诸如急诊室费用稳定等问题，以处理无法预见的医疗紧急情况的费用支付。婴儿潮一代的老龄化人口是这项养老院床位规划工作的推动力。通过寻找最有效的时间配备清洁人员，可以最大限度地减少医院床位清洁成本。平均而言，在中央商务区，30%的司机正在寻找停车位。这导致空气污染和拥堵急剧增加。使用来自大规模停车定价实验的大数据集，可以设定停车价格，以将污染降至最低。这种排队分析也可用于电动汽车充电站的动态优先策略。

This collaborative research project addresses two critical national challenges: (1) cost reduction in healthcare. and (2) sustainable transportation. Both of these problems are addressed by leveraging the common modeling framework of waiting lines or queues that are sensitive to time of day, day of week, seasonal, or annual effects. In the healthcare domain, a solution to the nursing home bed planning problem can be developed. This is an important issue due to the aging baby boomer demographic. Also hospital bed cleaning costs can be lowered by finding the most effective time to have cleaning staff available. Drivers looking for parking generate a large amount of pollution and congestion. Methods, motivated by sustainable transportation and enabled by large data sets, to reduce the amount of pollution and congestion caused by parking can be developed. This research also includes the training of graduate students and the inclusion of underrepresented minorities. The Conference for African American Researchers in the Mathematical Sciences (CAARMS) came into existence in 1995 to nurture and promote African American researchers in the mathematical sciences. Both PIs have given lectures and tutorials at CAARMS to share with the community of African-American mathematicians and aspiring students some of their latest research work.This project develops dynamic performance models and dynamic optimization solution techniques based on a rich class of time dependent stochastic network models called Markovian service networks. This modeling and optimization framework captures the stochastic dynamics of the system that cannot be found in the static, long term, steady state behavior. Using this framework, solutions to critical challenges in both healthcare and sustainable transportation can be developed. These solutions address problems such as emergency room cost stabilization to handle the payments from the costs of unforeseen medical emergencies. The aging baby boomer demographic is the impetus for this work on nursing home bed planning. Hospital bed cleaning costs can be minimized by finding the most effective time to have cleaning staff available. On the average, 30 percent of the drivers in central business districts are searching for a parking space. This leads to a dramatic increase in air pollution and congestion. Using a big data set from a large scale parking pricing experiment, parking prices can be set to minimize pollution. This queueing analysis can also be used for a dynamic priority policy for electric car charging stations.