Development of a geographic information system (GIS) to identify potential high collision locations in major cities in Saskatchewan

萨斯喀彻温省最近五年（2004-08年）的碰撞数据表明，每10万辆登记车辆的碰撞增加了9%，每10万人口的碰撞增加了16%。仅2009年一年，SGI就收到了156人死亡、6369人受伤和46429人财产损失的报告。萨斯喀彻温省每10万名有执照司机的死亡率最高，也是加拿大每10万人口的死亡率第二高。为了减少碰撞和伤亡的数量，正在应用《公路安全手册》中概述的程序，以协助安全网筛选过程。该过程涉及安全性能函数（spf）的创建，可用于预测未来在任何特定路段和十字路口的碰撞次数。萨斯喀彻温省的道路安全机构SGI可以在网络筛选过程中使用spf来确定可以安装红灯摄像头等安全改进措施的位置，以主动减少可预防的碰撞数量。HSM的基本安全系数可能有助于确定萨斯喀彻温省城市的高风险地点，但由于安全系数是根据其他司法管辖区的碰撞和交通量数据开发的，因此它们可能无法准确地模拟萨斯喀彻温省的安全水平。另一个限制是spf不能用于筛选没有交通量信息的道路网络。例如，SPFs不能用作萨斯喀彻温省当地街道的网络筛选工具，因为这些街道的交通量数据不可用。拟议研究的目标包括：i)根据萨斯喀彻温省主要城市的道路配置和碰撞严重程度，为城市十字路口和路段制定一系列spf；ii)将使用萨斯喀彻温省碰撞和交通量数据开发的SPFs的性能与HSM的基础SPFs的性能进行比较；iii)测试并选择一种可替代的网络筛选方法，如高比例网络筛选方法，适用于缺乏交通量输入的街道；（四）开发一个地理信息系统，分析空间碰撞模式，并将已确定的具有安全改进潜力的地点可视化。

The five most recent years' collision data from Saskatchewan (2004-08) show that collisions per 100K registered vehicles have increased by 9%, and collisions per 100K population have increased by 16%. In 2009 alone, 156 fatalities, 6,369 injuries and 46,429 property damage only collisions were reported to SGI. Saskatchewan has the highest number of fatalities per 100K licensed drivers, and the second highest number of fatalities per 100K population in Canada. To reduce the number of collisions and casualties, the procedures outlined in the Highway Safety Manual are being applied to aid the safety network screening process. The process involves the creation of Safety Performance Functions (SPFs) which can be used to predict the number of future collisions at any particular segment and intersection. SGI, the road safety authority in Saskatchewan, can use the SPFs in the network screening process to identify locations where safety improvements, such as red light cameras, can be installed to proactively reduce the number of preventable collisions. The HSM's base SPFs may be useful for identifying high risk locations Saskatchewan's cities, but as the SPFs were developed using collision and traffic volume data from other jurisdictions, they may not accurately model the level of safety in Saskatchewan. A further limitation is that SPFs cannot be used to screen a road network that does not have traffic volume information. For instance, SPFs cannot be used as a network screening tool for Saskatchewan's local streets as traffic volume data are not available for these streets. The objectives of the proposed research include: i) Develop a series of SPFs for the urban intersections and road segments by road configuration and collision severity for the major cities of Saskatchewan; ii) Compare the performance of the SPFs developed using Saskatchewan collision and traffic volume data to the performance of the HSM's base SPFs; iii) Test and select an alternative network screening method, such as a high proportion network screening method, that can be applied to streets that lack traffic volume input; and iv) Develop a GIS that analyses spatial collision patterns and visualizes the locations identified as having the potential for safety improvements.