A Strategy to Alleviate Rush Hour Traffics in Urban Areas Based on School-Trip Commute Information

Abstract

As one of the main social issues in the 21st century, urban traffic congestion has become a severe challenge, especially during rush hours. Investigation indicates relationships between school-trip commutes, i.e., students' trips to and from school, and traffic congestion. Therefore, optimizing traffic in terms of the school-trip commute is considered as an important practice of smart cities. This article proposes a long-coding adapted genetic algorithm for large-scale school district replanning. Moreover, the fitness function takes into account both the congestion coefficient model and the grayscale difference model, which can reduce the traffic congestion as well as balance the traffic load on school-trip routes. Based on the above results, the school-trip route optimization is formulated as the school bus routing problem (SBRP) and the split demand school bus routing problem (SDSBRP). A max-min ant system with a pheromone smoothing mechanism is proposed for the SBRP and the SDSBRP. Considering the large scale of the problem, a high-compactness K-means algorithm is proposed to determine bus stops before school bus routing. Besides, a public bus allocation strategy is proposed to allocate a fleet of public buses to serve school-trip routes with low allocation costs. The proposed algorithms are tested on a realistic and challenging dataset from real world. Experimental research results show the advantage of our proposed strategies in reducing overall rush-hour traffic and optimizing school-trip routes.