Abstract

In nowadays transportation networks, many multimodal itineraries are possible when traveling from an origin to a destination. Due to the plurality of alternatives, the choice of an option is difficult for the user. In this paper, we propose a novel decision support system to help users in that task. Our method ranks itineraries according to users’ preferences and trip profile based on a fuzzy logic system that transforms the itineraries’ characteristics and user’s preferences into a real number score. Unlike classical systems from Multiple Attribute Decision Making, we take into account the uncertainty both in the evaluation of the attributes and in the relative importance of those attributes. From an itinerary ranking point of view, our fuzzification methods enable to take into account user’s general preferences and preferences in the context of the current trip, hence using the complete alternative set to evaluate a single solution. Finally, the new defuzzification method scores the itineraries, which enables to build a connex transitive order on the solution set. Dependence in alternative evaluation and connex transitive order are scarcely used together for itinerary ranking. The proposed decision support system is generic and could be adapted to many decision criteria, depending on the application. An example, with several relevant ones, is presented through this work. To assess our method, we benchmark it against rankings from real persons obtained by the bias of two surveys. The results show that the rankings generated by the fuzzy logic system are convincing.

摘要

在当今的交通网络中，从起点到终点的旅行可能有许多多式联运路线。由于有多个备选方案，用户难以选择一个选项。在本文中，我们提出了一种新颖的决策支持系统来帮助用户完成该任务。我们的方法基于一个模糊逻辑系统，根据用户的偏好和行程概况对行程进行排名，该系统将行程的特征和用户的偏好转换为实数得分。与多属性决策的经典系统不同，我们在属性评估和这些属性的相对重要性中都考虑了不确定性。从行程排名来看，我们的模糊化方法能够在当前行程的上下文中考虑用户的一般偏好和偏好，因此使用完整的替代集来评估单个解决方案。最后，新的去模糊化方法对路线进行评分，从而能够在解决方案集上建立连接传递顺序。替代评估中的依赖和连接传递顺序很少一起用于行程排序。建议的决策支持系统是通用的，可以根据应用适应许多决策标准。通过这项工作提出了一个例子，其中有几个相关的例子。为了评估我们的方法，我们将其与通过两次调查的偏差获得的真实人物排名进行对比。结果表明，模糊逻辑系统生成的排名是有说服力的。