Traffic sequencing with Genetic Algorithm case study: La Paz traffic intersection
No Thumbnail Available
Date
2016-03
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
KNUST
Abstract
Managing traffic flow at complex intersections is a challenging task. The
intersection under study is situated at La Paz on the N1 highway. One can see
wreckages of crushed vehicles at different locations of the highway. Within one
month of its commissioning, the road registered fifteen (15) accidents with huge
loses to life and property. Between February 2012 and December 2013, 489
accidents had been recorded on the road with 350 people sustaining severe
injuries. Two hundred (200) pedestrians had also been knocked down by
speeding cars while 70 pedestrians died on the spot. These accidents occur when
pedestrians attempt to cross the road. Some identified causes of these accidents
are inefficient traffic order sequence, inadequate green time allocation, and non user friendly traffic installations. The breadth of the highway is about 50 metres
with no islands for pedestrians use. There are no overhead footbridge or
underground passage at the intersection for pedestrian use. These causes lots of
chaos at the intersection and leads to road block and heavy traffic during rush
hours. To save lives and properties, and improve traffic flow, the intersection is
investigated to propose new ways of improving pedestrian safety and traffic flow
sequence. Genetic Algorithm approach was used to find an optimal traffic flow
sequence for the intersection. The intersection was simplified into a ’plus’
intersection that has traffic streams moving in fourteen directions. The streams
were encoded and put through the genetic algorithm process. The output was
evaluated for both computational efficiency and accuracy. The program returned
a flow sequence that is very efficient. It gives hope that soon, the chaos and
accidents at the intersection will become
history
Description
A thesis submitted to the Department of Mathematics, Kwame Nkrumah University of Science and Technology in partial fulfilment of the requirement for the degree of MSc Industrial Mathematics