Abstract:
Reducing the road tra c congestion has become an important challenge in recent years; the
researchers have focused on identifying causes and remedies for the tra c congestion. However,
the impact of the locations and activity time of points of interests (POIs) such as shopping
centers, hospitals and schools, on road tra c has not yet been explored. More importantly, in
developing countries, POIs are not established in a planned manner and cause tra c congestion.
In this thesis, we analyze how POI locations and activities a ect the road tra c. Speci cally,
we rst identify the tra c congestion pattern caused by POI activities in our eld study in a
metropolitan city Dhaka of a developing country and develop a model to quantify the spatiotemporal
impact of POI activities on road tra c congestion. In addition, we develop a model
to predict the impact of POI activities on road tra c congestion using fuzzy regressions for
di erent POI categories. We categorize POIs based on predictability, regularity, cause (i.e.,
event or schedule driven) of POI tra c. Our developed models would enable us to select an
appropriate location for future POIs, suggest changing of activity time of POIs, and predict
a fastest path from a source to a destination based on POI impact on surrounding areas and
thereby reduce the tra c congestion. We perform a set of case studies and experiments to
show the e ectiveness of our quanti cation and prediction models using real datasets of POI
activities and road tra c of Dhaka city.
