Traffic control is a rapidly evolving subject, reflective of new developments in electrical sensor technology, and information and communication technology. In Dhaka, a city of about 12 million people, traffic congestion has worsened dramatically over the last 2-3 years, in spite of the introduction of automated traffic lights. Expensive and drastic measures, such as the construction of flyovers, have been implemented to counter traffic congestion. Although various studies have been conducted on this topic, few or none have identified the many advantages of human traffic control. Dhaka and other cities of developing nations, present traffic situations quite unlike that of developed countries. As a result of improper planning and land management, only about 7 % of the city area consists of road space (compared to about 12% of New Delhi, or 20 % required ideally). In 2004, human traffic control started getting replaced with a modern signaling system, with the good intent of bringing discipline in traffic. In spite of optimistic predictions, it was seen that traffic congestion had become much worse. In the field of information and communication technology, the advantages of human decision-making over automated or microprocessor-based decision making is well recognized. It is proposed and shown here that for Dhaka, human traffic control is a better alternative than automated traffic lights. Human traffic control is preferred for Dhaka, and other developing nations, because of the relatively fewer cars, the few major intersections, and the low cost of human traffic-controllers. As signaling intervals are determined by past measurements of traffic, automated traffic control cannot take into account statistical and event-related variations of traffic. A human traffic controller makes the better decision of allowing through a long line of cars, avoiding the transition times during predetermined changing of signals. He/she does not show green to an empty street with cars waiting in another line. The neural-networks of a traffic controller can assess traffic in visual range, and take intelligent and adaptive decisions. Even with the best road sensor technology, accompanied by telecommunication and fast microprocessors, it is a formidable task for an automated traffic system to match the decision-making capacity of a human traffic controller.
[1]
G W Euler.
INTELLIGENT VEHICLE/HIGHWAY SYSTEMS: DEFINITIONS AND APPLICATIONS
,
1990
.
[2]
Patrick van der Smagt,et al.
Introduction to neural networks
,
1995,
The Lancet.
[3]
Simon Haykin,et al.
Neural Networks: A Comprehensive Foundation (3rd Edition)
,
2007
.
[4]
J. Nazuno.
Haykin, Simon. Neural networks: A comprehensive foundation, Prentice Hall, Inc. Segunda Edición, 1999
,
2000
.
[5]
Simon Haykin,et al.
Neural Networks: A Comprehensive Foundation
,
1998
.
[6]
K W Ogden.
Human factors in traffic engineering
,
1990
.
[7]
H. S Tsay.
A MICROCOMPUTER-BASED ON-LINE TRAFFIC CONTROL SYSTEM
,
1989
.