GOALI/Collaborative Research: Optimal Inpatient Discharge Planning under Uncertainty

这一学术与行业联络合作研究资助机会（GOALI）的目标是为改善住院病房的病人流量和有效管理员工工作量提供见解。这项研究将使医院管理者能够提高患者满意度和服务吞吐量，政策制定者能够为医院工作流程制定改进的基准，一线供应商能够做出智能的数据驱动决策。当地和全国的医院将能够更好地调整护理流程，改善患者流量，同时降低成本。患者获得护理的机会和质量将得到改善。该项目的具体产出之一是决策支持系统，该系统预计具有很高的商业化潜力，因此在医疗保健分析市场中产生广泛的研究影响。研究目标将通过开发一套创新模型和解决方案来实现，这些模型和解决方案用于分析和优化与患者流程对齐和护理专业工作量管理相关的住院患者出院决策。这项工作将有助于减少住院病人出院迟到和过夜，减轻上游单位转院的延误，并平衡医院工作人员出院和非出院相关的工作量。这些改进将通过优化上游单位的床位要求与住院单位的床位释放来实现。本文将开发一种新的决策依赖的随机组合优化框架，该框架将住院病人出院优先级和工作量调度决策相结合，并将使用创新的场景分解方法来解决这一问题。我们将利用医院合作伙伴提供的真实数据，对提出的解决方法进行严格的理论推导和全面的计算研究。

The goal of this Grant Opportunity for Academic Liaison with Industry (GOALI) collaborative research award is to offer insights into improving patient flow through inpatient units and efficiently managing staff workload. This research will allow hospital managers to increase patient satisfaction and service throughput, policy makers to develop improved benchmarks for hospital workflows, and front-line providers to make intelligent data-driven decisions. Hospitals locally and nationally will be able to better align care processes, improve patient flow, all while lowering costs. Patients will have improved access to and quality of care. One of the concrete outputs of the project is a decision-support system that is expected to have have high potential for commercialization, and hence yield broad impact of the research, in the healthcare analytics market.The research goals will be met via the development of a set of innovative models and solution approaches for analyzing and optimizing inpatient discharge decisions pertaining to patient flow alignment and care professional workload management. This work will help reduce inpatient discharge lateness and overnight stays, alleviate delays in transfers from upstream units, as well as balance discharge and non-discharge related workload for hospital staff. These improvements will be accomplished by optimally aligning bed-requests from upstream units with bed-releases from the inpatient unit. A novel decision-dependent stochastic combinatorial optimization framework will be developed that integrates inpatient discharge prioritization and workload scheduling decisions, which will be solved using an innovative scenario decomposition method. Rigorous theoretical derivation and comprehensive computational studies will be conducted for the proposed solution approach using real-data from our hospital partner.