An interactive physician scheduling tool

基于我们去年完成的Engage项目EGP 462181-13，我们发现在人员调度方面存在许多挑战，包括数据标准化、数据评估、人员分配和约束理由。在将可用时隙分配给可用医生时存在许多不同的约束。事实是，没有任何一次性的调度可以满足所有的限制。仅靠人工工作，人工调度人员可能需要几天甚至几周的时间来调整相关人员的不同要求，以获得满意的调度方案。构建医生排班辅助工具是满足这一需求的一种很有前途的方法。交互式人事调度助理工具的目标是建立一个系统，该系统考虑具有特殊约束的特定业绩衡量标准，并根据特定参数优化每个互动。 一般调度问题是一个极具挑战性的问题，通常被称为非多项式(NP)困难问题。因此，世界各国的研究人员都在努力研究各种方法来提高调度算法的效率。这也是世界上至今还没有这样实用的调度软件的原因。对于Metricaid Inc.来说，该工具无疑是一个增值产品。面临的挑战是：1)算法本身很耗时；2)一些优化工具太复杂，无法理解和学习应用于现实世界的管理；3)与人类调度员进行有效和用户友好的交互；以及4)人类调度员接受的权衡，以考虑尽可能多的 医生的要求内在地相互冲突。

Based on the Engage Project EGP 462181-13 we completed last year, we find that there are many challenges in personnel scheduling including data standardization, data evaluations, personnel assignment, and constraint justifications. There are many different constraints in assigning available time slots to available physicians. The fact is that no one-time scheduling can meet all the constraints. With only manual work, human schedulers may take days or even weeks to adjust the different requirements from involved persons to acquire a satisfactory scheduling scheme. To build a physician scheduling assistant tool is a promising way to meet this requirement. The objective of an Interactive Personnel Scheduling Assistant Tool is to build a system for which particular measures of performance are considered with special constraints and each interaction is optimized based on specific parameters. General scheduling is a very challenging problem, normally called Non-Polynomial (NP) - hard problem. Therefore, the researchers in the world try hard to investigate variety of methods to improve the efficiency of scheduling algorithms. This is also the reason why there is no such practical scheduling software in the world until now. The proposed tool is a definitely value-adding product for the MetricAid Inc. The challenges are: 1) the algorithm itself is time-consuming; 2) some optimization tools are too complex to understand and learn to be applied to the real-world management; 3) effective and user-friendly interactions with the human schedulers; and 4) trade-offs accepted by human schedulers to consider as many as possible the physicians' requirements that are inherently conflicting with each other.