Environmental benefits mining based on data-driven taxi cruising recommendation strategy

Abstract

Provision of effective taxi cruising recommendation strategy is vital to drivers for improving their profit and reducing taxi exhaust emissions. This paper focuses on the problem of mining environmental benefits from the taxi cruising recommendation strategy while maximizing the profit of drivers. With abundant global position system (GPS) trajectories of taxis in Beijing, China, passenger hotspots are identified by density-based spatial clustering of application with noise (DBSCAN) algorithm, including weekdays and weekends. Then, spatio-temporal patterns of passenger hotspots as well as idleness emissions are analyzed and visualized based on the geographic information system (GIS) techniques. Based on spatio-temporal patterns of passenger hotspots and emissions under the idleness state, a profitable taxi cruising recommendation strategy is proposed to quantify and analyze the environmental benefits in terms of empty emissions. The results show that the proposed strategy significantly outperforms the current method by reducing idleness emissions 19.6% and 31.7% on the weekday and weekend, respectively. The results also appear to demonstrate that the proposed strategy is an effective approach in energy conservation and emission reduction.