Intelligent bus routing with heterogeneous human mobility patterns

Abstract

Optimal planning for public transportation is one of the keys helping to bring a sustainable development and a better quality of life in urban areas. Compared to private transportation, public transportation uses road space more efficiently and produces fewer accidents and emissions. However, in many cities people prefer to take private transportation other than public transportation due to the inconvenience of public transportation services. In this paper, we focus on the identification and optimization of flawed region pairs with problematic bus routing to improve utilization efficiency of public transportation services, according to people's real demand for public transportation. To this end, we first provide an integrated mobility pattern analysis between the location traces of taxicabs and the mobility records in bus transactions. Based on the mobility patterns, we propose a localized transportation mode choice model, with which we can dynamically predict the bus travel demand for different bus routing by taking into account both bus and taxi travel demands. This model is then used for bus routing optimization which aims to convert as many people from private transportation to public transportation as possible given budget constraints on the bus route modification. We also leverage the model to identify region pairs with flawed bus routes, which are effectively optimized using our approach. To validate the effectiveness of the proposed methods, extensive studies are performed on real-world data collected in Beijing which contains 19 million taxi trips and 10 million bus trips.

Taxi Demand Forecasting: A HEDGE-Based Tessellation Strategy for Improved Accuracy

“…The obtained cluster centroids are used to partition the city into geohashes and Voronoi cells. In the past, road intersections [10], [12] and bus stops [26] are employed to act as tessellation centers. In contrast, we use K-Means cluster centroids to act as the tessellation centers.…”

Reverse $k$ Nearest Neighbor Search over Trajectories

“…Our index supports dynamic updating, where new transitions and routes can be added into the index easily. This is in contrast the previous work [8,10] which needs to train whole dataset from scratch once there are new data inserted.…”

“…Chen et al [8] tried to approximate night time bus route planning by first clustering all points in taxi trajectories to determine "hot spots" which could be bus stops, and then created a bus route graph based on the connectivity between two stops. Based on human mobility patterns, Liu et al [10] proposed a localized transportation choice model, which can predict bus travel demand for different bus routes by taking into account both bus and taxi travel demands. However, the method must scan the static records for all of the customers, which is inefficient in practice, and the model has to be rebuilt whenever the records are updated.…”

“…Then the point is inserted into the corresponding sorted point list S filter .P , and each point is affiliated with a list of route IDs containing it. if e is a point then 10 S filter .add(e, C(e)); // Filtering set 11 else 12 foreach child c of e do 13 minheap.push(c, MinDist(Q, c)); 14 return (S filter , S refine ); Algorithm 3 shows how the filtering works. The filtering of a node is conducted in two steps.…”

Device discovery for D2D communication in in-band cellular networks using sphere decoder like (SDL) algorithm

“…On the other hand, every device randomly chooses RRB resources for discovery signal transmission. We concentrate on a random choice due to the human mobility pattern [19]. There are two scenarios for device discovery that depend on mobility, haphazard walk scenarios, and velocity scenarios in which discovery is computed.…”