• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>Moratuwa Engineering Research Conference >Rationalizing police patrol beats using heuristic-based clustering
【24h】

Rationalizing police patrol beats using heuristic-based clustering

机译：利用基于启发式的聚类合理化警察巡逻节拍

获取原文
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

The division of police patrol districts affects patrol performance, such as average response time and workload variation. However, the possible sample space is large and the corresponding graph-partitioning problem is NP-complete. Moreover, the resulting patrol beats must be contiguous and compact. We propose a heuristic based, clustering method to divide a given police district into optimal patrol beats based on crime and census data. Use of past crime data, their severity, and census data results in more compact shapes with lower crime response time and equitable workload. Moreover, it enables defining patrol beats for different seasons and time shifts. Furthermore, we considered the actual road distance than the traditional Euclidean distance in responding to crimes. We demonstrated the utility of the proposed method using a real-world crime and census dataset. For the given dataset, maximum response time for Calls For Service (CFS) was 35.2 seconds, which is the time taken to travel to any point in the patrol beat from the optimum positioning of police patrol car. Compactness was measured using Isoperimetric Quotient values for each patrol beat, and the average compactness was 0.7 indicating good compactness. Gini Coefficient was 0.036, which indicates balanced workload distribution among patrol beats.
机译：警察巡逻区的司影响巡逻性能，例如平均响应时间和工作量变化。但是，可能的样本空间大，相应的图形分区问题是NP-Tressim。此外，所得到的巡逻节拍必须连续且紧凑。我们提出了一种基于启发式的，聚类方法，将给定的警察区划分为基于犯罪和人口普查数据的最佳巡逻节拍。使用过去的犯罪数据，严重程度和人口普查数据导致更紧凑的形状，犯罪响应时间较低和公平的工作量。此外，它使得能够为不同的季节和时间转移定义巡逻节拍。此外，我们认为实际的道路距离比传统的欧几里德距离响应罪行。我们展示了使用真实犯罪和人口普查数据集的提出方法的效用。对于给定的数据集，服务（CFS）的最大响应时间为35.2秒，这是从警察巡逻车的最佳定位到巡逻节拍中的任何一点所需的时间。每个巡逻节拍的异常商值测量紧凑性，平均紧凑率为0.7表示良好的紧凑性。 Gini系数为0.036，这表明巡逻节拍之间的平衡工作量分布。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号