Recent studies indicate that the public transport bus would
be a better mode of travel for city people if a real-time bus
service can be provided, where quality attributes of bus such
as price, convenience and environmental friendliness can be
persisted at a high level [1, 2]. Travel time is the most widely
used variable in the transit information system because it can
be easily understood by people. From the transport company’s
side, travel time information is useful in routing and scheduling
of buses. Hence, there is an urgent need to develop a real-time
bus arrival time prediction algorithms that can provide more
accurate information under nowadays conditions.
The bus arrival time is the primary information to most
city transport travelers. Excessively long waiting time at bus
stop often discourages the travelers and makes them reluctant
to take bus. Bus arrival time prediction is particularly powerful:
travelers who would otherwise be enduring a potentially
frustrating wait can now adjust their travel plans to minimize
waiting time. Even where buses typically run on time, bus
arrival time prediction can improve travelers’ confidence in
the transit service, allowing users to make reasonable travel
arrangement before a trip.
Nowadays, most bus operating companies have been providing
their station lists and timetables on the web freely
available for the travelers. But, from the bus timetables,
travelers can only get very limited information (e.g., operating
hours, time intervals, etc.), which are typically not timelyupdated. These are far from satisfactory to the bus travelers.
For example, the schedule of a bus may be delayed due to
many unpredictable factors (traffic conditions, bad weather
situation, etc). When traveling with bus, the travelers usually
want to know the accurate arrival time of the bus.