Workshop on Decision Analytics for Dynamic Policing; Arlington, Virginia; May 9-10, 2019

该奖项支持一个会议，该会议召集了来自操作工程和刑事司法领域的研究人员，以及来自美国各地的警察部门领导，探索决策分析和运筹学的研究方向，以改进动态的、数据驱动的警务方法。从2008年开始，许多州的警察部门开始采用所谓的“预测性警务”，即使用分析方法预测犯罪将如何、何时、何地以及由谁实施。这些方法依赖于收集的时间、地点和犯罪细节的历史数据，以及肇事者和受害者的特征。虽然一些司法管辖区在实施预测性警务后显示出犯罪总体下降，但其他评估研究表明，其影响可以忽略不计，甚至是有害的。这些方法在三个关键方面受到了批评：(1)它们没有从社会学的角度出发；(2)假设数据包括预测和有效应对所必需的信息；(3)他们忽视了警务系统的动态性。本次会议将汇集适当的研究人员，以及从业人员，以建立一个跨学科的警务挑战的理解，可能是通过决策分析和运筹学解决。警察部门面临的最严峻挑战之一是有效利用资源来解决服务需求，特别是在应对严重犯罪时。警察部门认识到，社区安全综合办法的一部分包括在警察遭遇时保护所有参与者的权利。这次讲习班的重点是确定业务方面，包括数据收集和分析、规划、培训和犯罪现场调查，在这些方面进行进一步的研究可能会改善资源分配。会议的第一天包括预测警务和数据分析案例研究的主题演讲，以及工作组的分组会议。会议的第二天将有一位主讲人就国际警务战略发表演讲，工作组将继续讨论和记录研究需求。与大型警察组织建立了关系的四个研究小组担任讲习班的组织者。将招募更多的与会者小组（业务工程师、刑事司法研究人员、警察部门领导）参加讲习班。这项活动将提供一个独特的论坛，将各种利益攸关方聚集在一起，鼓励对这一重要社会话题的研究。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。

This award supports a conference that convenes researchers from the fields of operations engineering and criminal justice, along with police department leadership from across the US, to explore research directions in decision analytics and operations research to improve dynamic, data-driven policing methods. Beginning in 2008, so-called "predictive policing", which entails using analytical methods to predict how, when, where, and by whom crimes will occur, has been deployed by police departments in many states. These methods rely on historical data collected on time, location, and details of crimes, as well as on characteristics of perpetrators and victims. While some jurisdiction have shown an overall decrease in crime after implementation of predictive policing, other evaluative studies have shown the effects to be negligible or even detrimental. These methods have been criticized on three critical aspects: (1) they are not informed by a sociological perspective; (2) they assume that data include the necessary information for both prediction and effective response; and (3) they ignore the dynamic nature of the policing system. This conference will bring together appropriate researchers, along with practitioners, to build a cross-disciplinary understanding of policing challenges that may be addressable by decision analytics and operations research.One of the most serious challenges facing police departments is the effective use of resources to address calls for service, especially when responding to serious crimes. Police departments recognize that part of a comprehensive approach to community safety involves protecting the rights of all involved during police encounters. The focus of this workshop is on identifying operational aspects, including data collection and analysis, planning, training, and crime scene investigation, where additional research may lead to improved resource allocation. Day 1 of the conference includes keynotes on predictive policing and data analytic case studies, as well as breakout sessions for working groups. Day 2 of the conference will include a keynote speaker addressing international policing strategies, and working groups will continue to discuss and document research needs. Four teams of researchers who have established relationships with large police organizations serve as workshop organizers. Additional teams of attendees (operations engineers, criminal justice researchers, police department leadership) will be recruited to participate in the workshop. This activity will provide a unique forum to bring together a variety of stakeholders to encourage research on this important societal topic.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.