Establishing generic problem structures and repeatable solution approaches for healthcare delivery problems

设计长期可行(即可持续)的系统是许多行业面临的问题。接收的产品多于发货的库存系统耗尽了空间。呼叫中心接听电话的速度快于结束电话的速度，等待时间变得无限长。接受比积极治疗更多的新患者的医疗实践将经历拥堵和/或加班。使这些系统复杂化的是不确定性，因此将容量设置为与需求相等是不够的，并导致需要将空闲容量作为缓冲。此缓冲区对于可持续性和性能至关重要，但通常设置得较低，以便最大限度地减少“未使用”的容量。其中，确定这种缓冲的大小和设计巧妙的方式(例如，共享)以较少的资金过活是设计可持续系统的重大挑战。 这项研究计划将专注于设计可持续的医疗保健提供系统。具体地说，我们将解决许多医院普遍存在的病人流量、拥堵和容量规划问题。为了扩大运筹学对这些问题的影响，我们将制定全面的问题定义和结构，使解决方法适用于多种环境。从研究其他服务行业的成功经验来看，我们认为基于排队的方法在开发此类方法方面是有前途的。排队论特别适合于描述和设计一般环境下的系统。这样的结果通常跨越应用领域，可以提供一般性的见解并勾勒出具体的行动方针。这在电信业以及最近在呼叫中心的研究和设计中都可以看到。尽管医疗保健具有高度的随机性和日益增长的需求，但使用排队论来支持加拿大的医疗保健系统并不常见。医疗保健的服务特点是独一无二的，需要开发新的排队模型和方法。从这项研究中，对具体干预措施的一般见解和建议将有助于设计可持续的医疗保健系统。

Designing systems that are viable over the long term (i.e. sustainable) is a problem faced in many industries. Inventory systems which receive more products than they ship run out of space. Call centres receiving calls faster than they end calls see waiting times grow unbounded. Healthcare practices that accept more new patients than they can actively treat will experience congestion and/or overtime. Complicating these systems is uncertainty, thus setting capacity equal to demand is insufficient and results in the need to incorporate idle capacity as a buffer. This buffer is crucial for sustainability and performance but is often set low in order to minimize "unused" capacity. Among others, determining the size of this buffer and designing clever ways (e.g. pooling) to get by with less are substantial challenges to designing sustainable systems. This research program will focus on designing sustainable healthcare delivery systems. Specifically we will solve patient flow, congestion and capacity planning problems that are common in many hospitals. To broaden the impact of operations research on these problems, we will develop comprehensive problem definitions and structures such that the solutions methodologies are applicable in multiple settings. From studying successes in other service industries, we believe that a queueing based approach is promising for developing such methodologies. Queueing theory is particularly apt at characterizing and designing systems for general settings. Such results typically span application areas and can provide general insights and outline specific courses of action. This has been seen in the telecommunications industry and more recently in the study and design of call centres. Despite the highly stochastic nature of healthcare and the growing need, the use of queueing theory to support the Canadian healthcare system is not commonplace. The service characteristics of healthcare are unique and will require the development of new queueing models and methodology. From this research, general insights and recommendations for specific interventions will contribute to the design of a sustainable healthcare system.