Page 112 - NGTU_paper_withoutVideo

Modern Geomatics Technologies and Applications

Submission Template for 1st international conference and 2nd national conference on
“Modern Geomatics Technologies and Applications”, in honor of Professor Abbas
Rajabifard

Spatio-Temporal Analysis of Traffic Congestion Based on Data Obtained From Modern
Technologies

Matin Shahri
Department of Geoscience Engineering, Arak University of Technology, Daneshgah St., Arak, Iran
* shahri@arakut.ac.ir

Analysing traffic conditions and presenting appropriate methods for traffic flow management play a critical role in
evaluating the performance of transportation systems. Among different methods of collecting traffic data, approaches
based on new technologies that enable the collection of large volumes of dynamic spatio-temporal data and facilitate the
extraction of trends and patterns have attracted more attention. .In this study, Tehran as the capital of Iran with its special
socio-economic characteristics and the variety of trips that lead to variable traffic situation, has been studied. Using
analysis and traffic image data processing relating to Tehran network, the trend of average changes in traffic congestion
within Congestion Charging Zone (CCZ) Air Pollution Control Zone (APCZ) and out of these areas on working and non-
working days have been investigated. The Moran’s I spatial autocorrelation index, indicates the congested regions in the
study area. Also, Havmoler diagrams showed the trend of spatio-temporal changes of traffic congestion on weekdays and
weekends.

1. Introduction
Nowadays evaluating the traffic conditions through providing appropriate and cost-effective methods plays a key role in
analysing the performance of transportation systems. City developments as well as growing number of vehicles lead to increase
road traffic congestion in urban areas with its negative subsequent effects such as fuel waste, environmental pollution, high rate
or number of accidents or injuries and overall delay in the transport system. Accordingly, any appropriate operation to contribute
to better understand the situation and to eliminate the effects of traffic congestion or reduce its impact have always been the main
concerns of transportation managers. To this end, collecting comprehensive and sufficient spatial and temporal data from the
entire study area using low-cost data collection instruments will be the matter of concern. Among different methods of traffic
data collection, methods based on new technologies and intelligent transportation systems (ITS) have received more attention in
recent years which allow the collection of large volumes of data and facilitate the process of extraction of patterns over space
and time.
In recent years, various methods and algorithms have been proposed to manage and monitor traffic conditions and reduce
its negative impacts. In literature, traffic congestion in urban areas has been studied from different points of view [1-5] In some
studies, traffic congestion has been measured based on some univariate parameters such as travel time and speed [6, 7]while
some, used multiple parameters [8-10] . Several research can also be addressed in which, predicting and estimating traffic
congestion in different time horizons have been the matter of concern [11-13].
When dealing with spatial data such as many traffic parameters (e.g. travel time, speed and congestion), the value of a
variable at a specific location indicates correlation with the same value in proximate locations which is known as spatial
autocorrelation[14]. Several techniques have been suggested to measure the amount of spatial correlation among which, Moran’s
I can be regarded as one of the most frequently-used one that has been applied in various fields such as environmental sciences
[15-18], medicine [19-21] and urban studies [22-25].
On the other hand, most traffic data such as congestion as a dynamic variable not only vary spatially in different parts of
the study area but has variation within different time intervals (peak and non-peak hours) and therefore, spatial- temporal analyses
can give an insight into better understand the relationships. Studies can also be mentioned in which spatial patterns of traffic
congestion [26-28] as well as its heterogeneity have been investigated [29-31].
On the other hand, data source availability has an underlying effect on reliability of the results obtained from the studies
of traffic congestion, as well as, the accuracy of outputs. Different types of methods have been developed over the past decades
to collect traffic information among which, approaches based on new technologies and intelligent transportation systems have
attracted more attention due to collection of large amounts of data over time. Additionally, the descriptive analyses and extracting

1