With rapid economic growth, city sizes are expanding, and city traffic flow is surging. Thus traffic congestion, traffic accidents, and serious air pollution are becoming more common. Priority has been given to the development of urban public transportation policy by using GPS, GIS, the internet, and communication technology to realize the collection, transmission, storage, and processing of the massive historical and real-time data gathered from bus IC cards, in order to build an intelligent, modern city public transport dispatching platform that will solve social problems such as urban traffic congestion, energy shortages, and air pollution. In addition, when passengers choose a reasonable travel plan, they can lessen not only their travel costs but can also increase the efficiency of public transport vehicles. Therefore, the choice of a reasonable bus route is of great importance to the daily operation and management of urban traffic.In recent years both in China and abroad, urban public transport has been the subject of extensive research and discussion. However, there are still many complaints about crowded buses, the length of waiting time, and the unpunctuality of the buses. How to effectively and efficiently recommend comfortable bus routes to bus passengers is a challenging and complex task. To address this critical challenge, we must consider factors reflecting the passengers' demands such as waiting time, crowded time, and driving time. To recommend comfortable bus routes for bus passengers, we suggest using multi-objective programming with various constraints and have developed a genetic algorithm to search for solutions. As a result, a bus route according to the differing requirements of passengers can be recommended.The rest of the paper is organized as follows. In Section 2, we review related work on the data processing of bus IC cards and personalized information recommendation services. In Section 3, we present a multiobjective program with various constraints to recommend comfortable bus routes for bus passengers and use a genetic algorithm to search for solutions. In Section 4, we discuss our implementation and empirical analysis of the proposed method. Finally, we conclude the paper and point out future research directions. With the generation of massive data from bus IC cards, how to effectively and efficiently recommend comfortable bus routes to bus passengers is a challenging and complex task. In this paper, waiting time, crowded time, and driving time between different bus stations on different bus routes at different times of the day are calculated from bus IC cards data history. Then, a multiobjective program with various constraints is suggested to recommend comfortable bus routes for bus passengers, and a genetic algorithm is developed to search for expected solutions. The proposed method is implemented using bus IC cards data from Chongqing, China and will be a promising tool for bus passengers when choosing comfortable bus routes.
PROCEEDINGS PAPER