SBIR Phase II: Optimizing emergency department nurse scheduling via a novel operational intelligence platform

这项小企业创新研究（SBIR）第二阶段项目的更广泛影响/商业潜力是改善急诊临床护理。急诊部门面临着复杂的护理利用问题，部分原因是建立在错误假设基础上的调度系统。日益严重的护理短缺和倦怠问题必须得到解决，因为疲劳现象十分普遍。由于医院面临着巨大的收入损失，因此没有机会低效地使用护士。护士轮班必须以一种优化可用资产和平衡复杂权衡的方式进行智能排序。提出的解决方案使用人工智能（AI）和运筹学来预测适当规模的临床资源。该平台将打包为基于云的解决方案，减少护理人员流失，减少患者等待时间，降低医疗保健交付成本，并提高收入。复杂的数学方法超越了日历和Excel电子表格中可用的方法，特别是该项目基于约束的算法、模拟方法和机器学习，将成为优化急诊科运营绩效的转折点。这将改善卫生系统的病人护理、运营和财务结果，同时减少工作量、浪费、成本和护士职业倦怠。本项目通过提高可扩展性和可用性来解决多阶段急诊科护士人员配置问题。该方案基于运筹学、复杂的数据科学和排队理论模型，扩展了一套多目标优化算法，构建了一个端到端的决策支持平台。该平台将为急诊科护士人员配置提供纵向决策支持：护士经理将每天使用它来灵活决策，每月使用它来进行日程规划和分配，每年使用它来进行预算编制。目标包括：(1)数据科学：细化和自动化预测，适应优化模型，生成和维护输入文件，自动化输出文件；(2)前端：创建UI组件，创建业务层/API，连接业务层和UI，开发集成和组件测试；(3)后端：实现容器编排系统，升级安全，实现数据摄取协议，支持开发操作；(4)验证推荐并创建客户ROI计算器。该奖项反映了美国国家科学基金会的法定使命，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 Small Business Innovation Research (SBIR) Phase II project is to improve emergency clinical care. Emergency Departments are facing complex nursing utilization problems, fueled in part by scheduling systems built on flawed assumptions. Escalating nursing shortages and burnout must be addressed as fatigue is rampant. With hospitals facing significant revenue losses, there is no room for inefficient use of nurses. Nurse shifts must be intelligently sequenced in a way that optimizes available assets and balances complex sets of tradeoffs. The proposed solution uses artificial intelligence (AI) and operations research to predictively right-size clinical resources. The platform will be packaged as a cloud-based solution that reduces nursing staff churn, decreases patient wait times, reduces healthcare delivery costs, and improves revenue. Sophisticated mathematical approaches beyond what is available in the calendar and an Excel spreadsheet – specifically this project’s constraint-based algorithms, simulation methods and machine learning – will be the turning point in optimizing emergency department operational performance. This will improve the health system’s patient care, operational, and financial outcomes while reducing effort, waste, cost, and nurse burnout. This proposed project addresses the multi-stage emergency department nurse staffing problem by increasing scalability and usability. The proposed solution scales a set of multi-objective optimization algorithms based on operations research, sophisticated data science and queueing theory models to create an end-to-end decision support platform. The platform will provide longitudinal decision support for emergency department nurse staffing: nurse managers will use it daily for flexing decisions, monthly for schedule planning and assignment, and annually for budgeting. Objectives include: (1) Data science: refine and automate the forecasting, adapt the optimization model, generate and maintain input files, automate output files; (2) Front-end: create UI components, create service layer/API, wire service layer and UI, develop integration and component testing; (3) Back-end: implement container orchestration system, upgrade security, implement data ingest protocols, support development operations; and (4) Validate recommendation and create client ROI calculator.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.