Taxis play an important role in offering comfortable
and flexible service within Singapore’s public transport system.
Due to the inherent randomness in taxi service system, many taxi
companies still rely on the drivers’ experience to seek passengers.
Today, Singapore's five taxi companies now use some form of
wireless and GPS (Global Position System) satellite to track taxis
traveling in urban area. GPS-equipped taxis can be viewed as
ubiquitous mobile sensors which enable us to collect large
amounts of location traces of individuals or objects. In this paper,
we first investigate the characteristics of travel behavior of urban
population. Next, a taxi business intelligence system is proposed
to explore the massive transportation data based on spatialtemporal
data mining techniques. Furthermore, various taxi
business models are created to make comprehensive analysis on
taxi business problems. Finally, the value of the taxi business
intelligence system is demonstrated by applying it to some realworld
scenarios. Results show that this system can significantly
improve the quality of taxi services.