Motor vehicle crashes are a leading cause of death in the U.S. In order to reduce death and serious injury, road and traffic engineers manually evaluate road segments and visualize the safety level of roads. These existing risk maps can be confusing and must be manually interpreted by drivers to find the safest path from a source to a given destination; this can result in ignoring the safety of the routes by drivers. In addition, common navigation systems such as Google Maps and Waze present two or three alternative paths from a source to a given destination based on the travel time and distance. A navigation system is required to take the safety level of the road segments into consideration while suggesting a path. This navigation system needs to acquire knowledge from various sources, a user interface to obtain user preferences, and an inference engine to find the best paths. Such a system can still suggest multiple conflicting paths, such as shortest, fastest and safest paths. This paper presents the addition of a multi-criteria decision-making (MCDM) method, Analytical Hierarchy Process, to a previously designed Safe Route Planner to aid users in choosing the most suitable path among M alternative paths. Different MCDM methods can generate different results while applied to the same problem. There are a few comparative studies to compare the results of different Multi-Criteria Decision-Making (MCDM) methods. Therefore, a particular attention is devoted to comparing the results of five decision-making techniques, namely AHP, Fuzzy AHP, TOPSIS, Fuzzy TOPSIS and PROMETHEE through two real-world case studies. In addition, the comparative studies fail to adequately quantify the results of the MCDM methods; consequently, another aim of this research is to investigate the applicability of Spearman's rank correlation coefficient, Average Overlap and Discounted Cumulative Gain techniques to quantify the results of the MCDM methods.

在美国，机动车撞车是导致死亡的主要原因。为了减少死亡和重伤，道路和交通工程师手动评估了路段并形象化了道路的安全等级。这些现有的风险图可能会造成混淆，驾驶员必须手动对其进行解释，以找到从源头到给定目的地的最安全路径。这可能会导致驾驶员忽视路线的安全性。此外，基于行进时间和距离，诸如Google Maps和Waze之类的常见导航系统还提供了从源到给定目的地的两条或三个替代路径。要求导航系统在建议路径的同时考虑路段的安全等级。该导航系统需要从各种来源获取知识，需要一个用户界面来获取用户偏好，还有一个推理引擎来找到最佳路径。这样的系统仍然可以建议多个冲突的路径，例如最短，最快和最安全的路径。本文介绍了在先前设计的“安全路线计划”中添加的多标准决策（MCDM）方法“层次分析过程”，以帮助用户在M条替代路线中选择最合适的路线。当应用于同一问题时，不同的MCDM方法可以产生不同的结果。有一些比较研究可以比较不同的多标准决策方法（MCDM）的结果。因此，通过两个真实的案例研究，特别关注比较AHP，Fuzzy AHP，TOPSIS，Fuzzy TOPSIS和PROMETHEE这五种决策技术的结果。此外，比较研究未能充分量化MCDM方法的结果；因此，本研究的另一个目的是研究Spearman秩相关系数，平均重叠和折现累积增益技术对MCDM方法的结果进行量化的适用性。