【作者】 张帅；

【导师】 彭玉青；

【作者基本信息】 河北工业大学 ， 计算机应用技术， 2004， 硕士

【摘要】 在我国众多大城市普遍存在交通拥挤问题，造成交通拥挤的首要原因是城市交通基础设施的建设远远落后于城市交通需求的增长。大力发展公共交通是解决城市交通拥挤问题的首选措施。 选择路径是公交信息系统中的关键技术，有很多经典的算法解决最短路径问题，如Dijkstra算法和Floyd算法，但大都不适合于公交查询最短路径。一是由于运算速度慢，二是由于计算结果含有大量无用数据。 当乘客乘车时，他的首要目标是尽快到达目的地，第二目标是花费少。因此本文提出一个新的解决方法。第一目标是换乘车次最少，第二目标是出行路程最短。如果乘客换乘车次增多，因此而产生的不可预知的因素也会增多，另外由于大多数城市采取单一票制，所以换乘车次最少也就保证了花费最少。 本算法基于宽度优先并用蚂蚁算法优化，蚂蚁算法是一种基于真实蚁群的人工智能系统，常用于解决组合优化问题，如旅行商问题(TSP)，在蚂蚁算法中一组蚂蚁团结合作，寻找TSP问题的最优解。合作的方式是利用在寻找最优解时在TSP图的边上留下的信息素，是一种间接的通信方式。 在一个城市中，如果它的公交车运营线路是稳定的，那么从一个车站到达另一个车站的路线也应该是比较稳定的。因此我们可以将每对站点之间的路径全部求出，再加上关于本路径的一些信息如长度、信息素等全部放在数据库中，当一个乘客需要查找时就从此数据库中选择。每条路径包含一项信息素用以表示路径的优先级，选择某条路径的乘客越多，这条路径的优先级就会变得越高，以后被选择的机会也大。但是当一条路径被堵时优先级会降低。因此，这个系统使用时间越长，系统越智能化，越能为乘客提供有效的信息。 最后，本文提出了发展适合我国城市的公交乘客信息系统的设想和建议，解决了公交乘客信息系统的一项关键技术—公交乘客出行路径优化模型，并研究开发了唐山市公交问路系统。更多还原

【Abstract】 Traffic congestion exists widely in China large cities. The first reason is the providing of transportation facilities is much less than the development of transportation demand. To develop public transportation is the first choice in solving the problem of congestion.Choosing route for passengers is the key technology in the bus information system. There are many classic algorithms such as Dijkstra algorithm and Floyd algorithm to solve the problem. But they don’t fit for this problem. The first reason is that it doesn’t have a high speed; the second is that its result has large numbers of unuseful data.When a passenger take a bus, his first aim is to arrive at the destination as soon as possible; the second is to pay as least as possible. So a new algorithm is put forward in my paper. The primary object is to use the least transfer times, the second is to choose the shortest route distance. There will be more factors we can’t forecast if a passenger transfer more bus. Furthermore the cost to take a bus is uniform in many cities, so the less time he transfers bus the less he will pay for it.The algorithm is based on the breadth-first search and optimized by ant algorithm. Ants system is an artificial system based on the behavior of real ant colonies, which is often used to solve combinatorial problems, such as Traveling Salesman Problem (TSP). In ant colony system (ACS), a set of cooperating agents called ants cooperate to find good solutions to TSP. Ants cooperate using an indirect form of communication mediated by pheromone they deposit on the edges of the TSP graph while building solutions.I apply the ant algorithm as follows. In a city, if it has stable bus route, it will has stable route from a stop to another by bus. Then we can calculate all possible routes and store them and its length in a database. A passenger can choose them from the database when he need it. Each route has an pheromone to express the route’s priority. The more passengers choose the route the higher its priority will become,and have more chance to be choosed later. But if a route is obstructed the priority will be decreased. So the longer the system is used the more information it can provide for passengers.At last some suggestions and recommendations have been raised to develop suitable transit passenger information system for Chinese cities. A GIS-based passenger route choice model has been raised in this thesis. It is based on a new shortest path algorithm that uses the least transfer times as primary object and shortest path distance as second object. Application of this model in Tangshan public transportation network show that it is efficient and practical.更多还原