期刊
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
卷 21, 期 10, 页码 4225-4240出版社
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TITS.2019.2939224
关键词
D-day itinerary; genetic algorithm; itinerary planning; multiple itineraries; niching strategy
资金
- National Natural Science Foundation of China [61873095, 61873097, U1701267]
- Guangzhou Science and Technology Planning Project [201904010211]
- Guangdong Natural Science Foundation Research Team Project [2018B030312003]
- Guangdong-Hong Kong Joint Innovation Platform Project [2018B050502006]
Automatic itinerary planning is a crucial and challenging issue in tourism. This paper proposes a novel automatic planning method to suggest multiple itineraries that satisfy the specific demands of tourists. First, a multiple-itinerary planning model is developed, which provides three customized goals for a tourist to choose and supports generating multiple $D$ -day trips. The model makes fewer assumptions than the literature works did, while it provides more flexibility to the tourists. Then, based on the multiple-itinerary planning model, we design a niching genetic evolution approach to accomplish the automatic itinerary planning task. The genetic evolution approach guarantees a high search efficiency, while the niching strategy facilitates maintaining the population diversity. Consequently, the resultant algorithm can finally provide a number of diverse and superior solutions. Experimental results on real-world datasets show that our proposed algorithm not only outperforms state-of-the-art methods in considering different user-specified goals, but it is also capable of generating a set of diverse itineraries for the tourist to select. Additional experiments further verify the scalability of the proposed algorithm in terms of the problem size and the optimization objective.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据