RESEARCH ARTICLE | SEPTEMBER 14 2017

Analysis of vehicular traffic flow in the major areas of Kuala

Lumpur utilizing open-traffic 
Saargunawathy Manogaran; Muhammad Ali; Kamaludin Mohamad Yusof; Ramdhan Suhaili

AIP Conf. Proc. 1883, 020013 (2017)

https://doi.org/10.1063/1.5002031

CrossMark

 
View Export
Online Citation

24 February 2024 21:27:19

 Analysis of Vehicular Traffic Flow in the Major Areas of
Kuala Lumpur Utilizing Open-Traffic
Saargunawathy Manogaran 1, a) Muhammad Ali1, b) Kamaludin Mohamad Yusof 1, c),
Ramdhan Suhaili1,d)
1
Advanced Telecommunications Technology Research Group, Faculty of Electrical Engineering, Universiti
Teknologi Malaysia, Skudai, 81310, Johor, Malaysia
a)
Corresponding author: saarguna89@gmail.com
b)
kmuhammadali92@gmail.com
c)
kamalmy@fke.utm.my
d)
ramdhansuhaili@gmail.com

Abstract. Vehicular traffic congestion occurs when a large number of drivers are overcrowded on the road and the traffic
flow does not run smoothly. Traffic congestion causes chaos on the road and interruption to daily activities of users. Time
consumed on road give lots of negative effects on productivity, social behavior, environmental and cost to economy.

24 February 2024 21:27:19

Congestion is worsens and leads to havoc during the emergency such as flood, accidents, road maintenance and etc., where
behavior of traffic flow is always unpredictable and uncontrollable. Real-time and historical traffic data are critical inputs
for most traffic flow analysis applications. Researcher attempt to predict traffic using simulations as there is no exact model
of traffic flow exists due to its high complexity. Open Traffic is an open source platform available for traffic data analysis
linked to Open Street Map (OSM). This research is aimed to study and understand the Open Traffic platform. The real-
time traffic flow pattern in Kuala Lumpur area was successfully been extracted and analyzed using Open Traffic. It was
observed that the congestion occurs on every major road in Kuala Lumpur and most of it owes to the offices and the
economic and commercial centers during rush hours. At some roads the congestion occurs at night due to the tourism
activities.

INTRODUCTION
Many people have proposed and implemented several methodologies to find the real-time traffic flow. Response
team depending on all those methodologies to control traffic flow which at the end is not effective enough. A number
of cameras are placed at the intersection of the density of heavy traffic or roads capturing real-time traffic images.
Road sensors are also used to collect data on the main streets of heavy traffic. Various traffic applications were
developed on Android and iOS platforms such as Waze and Google Map to provide real-time traffic updates to users
[1].
During the emergency, behavior of traffic always changes and will not remain the same. This behavior leads to
havoc movement. Thus, creates complications for the relevant team to carry out their task in shorter time. Traffic
congestion is one of the unavoidable things during festive season and also during phenomena such as flood, accidents
and etc [2].
The study on pattern of road traffic using real-time data is quite challenging since traffic is always unpredictable
and varies. Due to heavy traffic flow especially in Kuala Lumpur area a solution to avoid should be suggested.
Therefore, it has been the motivation in the research to study and to collect the real-time traffic flow and analyze using
Open Traffic software.
The organization of the paper consists of 5 main headings. First main heading presents the introduction of the
work. A detail problem background was illustrated in this section. Second main heading presents the literature review
related to platform. Firstly, a review of traffic monitoring and traffic analysis from other researchers is described. In
third main heading, an overview of the Open Traffic system is given. In the fourth main heading, the result and the

Advances in Electrical and Electronic Engineering: From Theory to Applications

AIP Conf. Proc. 1883, 020013-1–020013-7; doi: 10.1063/1.5002031
Published by AIP Publishing. 978-0-7354-1563-8/$30.00

020013-1
discussion based on the data collected from the Open Traffic system is discussed. Last heading concludes the research
findings.

RELATED WORKS
The work consists of considering various aspects that previous researchers have done on traffic monitoring and
traffic analysis. In the following sub sections, previous works related to this work are reviewed.

Intelligent Transport Information System

Traffic monitoring and analysis is an essential requirement for management of a country to plan for new
development at particular location. Traffic congestion is one of the unavoidable events across the globe. Moazzem
Hossain [3] presented the issues related to Integrated Transport System (ITS) in Malaysia focusing on Klang Valley
region. Malaysia Highway Authority (MHA) and Kuala Lumpur City Hall (DBKL) are responsible for planning and
monitoring of highway network. MHA has 12 different highway control centres with one national traffic management
centre (TMC) hub. Those control centres were connected to about 155 vehicle detection stations, more than 200
cameras for Closed Circuit Television (CCTV) and 80 Variable Message Signs (VMS). The purposes of this devices
are to extract data, video and voice informations.The collected traffic information are stored and then processed before
sends to public user via VMS, Short Message Service (SMS) and Multimedia Messaging Service (MMS) [3]. Kuala
Lumpur Intelligent Transport Information System (ITIS) was commissioned with its TMC in Technology Park, Bukit
Jalil in March 2003. The main purpose of ITIS is to manage wide spectrum of CCTVs, vehicle detectors, VMS boards,
probe vehicles equipped with Global Positioning System (GPS) units, emergency and incident management. ITIS
provides real time areas involved in traffic congestion, arrival time estimation, cause of traffic congestion and an
alternative route the next destination [4]. Unfortunately, the implementation of ITIS in Klang Valley has lack of
collaboration among public and private agencies. Therefore, it is required to develop a common user friendly platform

24 February 2024 21:27:19

which can be used by multi-agencies [3].

Crowdsourcing based Traffic Information System

Crowdsourcing based Traffic Information System (CroTIS) is an server and client based architecture proposed by
Roopa, Anantharaman and Shanta [5] to solve challenges of real-time traffic data. Figure 1 shows the architecture of
CroTIS. It consists of mobile user and CroTIS components.
Those who are registered with real-time traffic server via mobile or tablet are called as mobile users. An user
friendly application installed on the mobile will enables the device connected to CroTIS traffic server and exchanges
real-time traffic informations. Mobile or tablet acts as a sensor and sends location information to the traffic server.
Then, the mobile updates the changes in location frequently [5]. CroTIS server constantly updates traffic informations
to mobile users.

FIGURE 1. Architecture of CROTIS

020013-2
 Subscriber system stores GPS coordinates from mobile user in database called as Crowd Data Source. While
Publisher provides traffic updates to user which received from Traffic Information Controller (TIC). TIC consists of
Navigation controller (NC) and Location Source. NC estimate traffic density with lists of paths given by Location
Manager and return it to Navigation Service Provider (NSP) [5]. Google Maps, Android and Hadoop are used to
implement the CroTIS. The server has been developed by using Python and Hadoop while for client application has
been developed based on Android. Hadoop was used to generate millions of GPS information and the proposed system
was simulated successfully [5].

Geosocial Networks
Nowadays, users for geosocial networks are increasing and touched millions of users such as FourSquare,
Facebook Places, Google Latiude and Waze. Michael Fire [6] presented data mining using geosocial networks for
improving road safety based on collaborative community-based navigation applications. Traffic data such accidents
reports, traffic jams reports, speed traps reports and traffic data were collected from Waze application. The data was
extracted from Waze geosocial network by using a dedicated Web crawler. The Web crawler focused on traffic reports
in Israel. This is because Israel has 1.1 millions of active waze users and this contributes to huge data collection. The
author concludes that Waze helps to identify the intersections and areas which are more likely to suffer from accidents
[6].

OVERVIEW OF THE OPEN TRAFFIC SYSTEM

The implementation of the proposed system requires several open source software. Firstly, Java and Maven will
be installed, following with the Traffic Engine software and Open Traffic application. The work also involves
downloading the Open Street Map of Kuala Lumpur. At later stages of the implementation the configuration file will

24 February 2024 21:27:19

be edited and the Traffic Engine app will be used to extract the real-time data to be analyzed.

Components of the Open Traffic System

The components of the Open Traffic system mainly consists of the traffic engine, traffic data pool and OSM-linked
traffic data set. The Traffic Engine is responsible for the vehicle location translation to OSM-linked speed estimates.
Traffic Engine is located inside a fleet operator. It converts real-time GPS location information to usable traffic
statistics. This block ensures the data fetched from data provider’s network are completely anonymous traffic statistics.
On the other hand, the Traffic Data Pool is a central storage repository for traffic flow monitoring. The pool collects
traffic information from Traffic Engine and provides access to an overall real time snapshot of traffic conditions. It
also stores historical traffic data for analysis purpose [8]. Also, OSM-linked Traffic Data Set builds static
archival and real-time snapshot of traffic data. This will be useful for third-party application developers to integrate
traffic data sets into routing, mapping and analytic applications without any restriction [8]. Along with that, Real Time
Routing API will be build. This pool comes up with multiple interfaces to data conditions including a realtime routing
API. This allows the run time configuration of the routing resources. This helps to generate routing and estimate arrival
time [8].

Data Extraction
Open Traffic collaborate with Malaysia based GrabTaxi, a taxi hailing app company that operates in Malaysia,
Vietnam, Philippines, Indonesia, Singapore and Thailand [8]. They have an unlimited access to the traffic data from
Grab application without any charges. The traffic data was updated for every 6 seconds which was collected from at
least 64000 vehicles. The data is much accurate as it is taken by the help of the GPS installed in each of the cars. The
GPS data is formatted in comma-separated value (CSV or .csv) file. Following are the steps to convert GPS data into
travel times by road segments [8].
(i) Calling the OSM tiles
(ii) Mark the virtual detectors to the OSM tiles
(iii) Estimating travel time based on virtual detector crossings

020013-3
 Open Street Map and Virtual Detectors
Open Street Map (OSM) is an open source map which is available for editing by third party. The Open Traffic
platform depending on OSM features including an average traffic speed calculation. The specific needed OSM tile
can only be called for traffic speed estimation to increase the processing speed and efficiency. GPS coordinates used
from csv data file which is loaded by csv loader [7]. Virtual detectors are nodes assigned on the OSM tiles to estimate
traffic speed. Figure 2 shows the assigned virtual detectors on OSM tiles.

FIGURE 2. Virtual Detectors

24 February 2024 21:27:19

RESULTS AND DISCUSSIONS

Traffic Analysis for Whole Kuala Lumpur

Around 2 million of traffic speed data was exported from Open Traffic software. Graphs are plotted using the
average traffic speed for whole Kuala Lumpur area. The collected traffic data in figure 3 is an average speed for all
districts in Kuala Lumpur. The districts are Bukit Bintang, Titiwangsa, Setiawangsa, Wangsa Maju, Batu, Kepong,
Segambut, Lembah Pantai, Seputeh, Bandar Tun Razak and Cheras. Figure 3 shows in general, that the average traffic
speed is higher during weekdays and smoother during weekend.

FIGURE 3. Traffic Analysis For whole Kuala Lumpur area in October 2016

020013-4
 It is observed that on Saturdays and Sundays, the average speed falls between 45 - 50 km/h for October. While for
November the speed is between 44 - 49 km/h. The minimum speed is detected on 7th October, 2016 and 14th October,
2016. This shows that the overall average speed on every Fridays is slower compared to other days. This is because,
Muslim people performs Friday prayer on every Fridays which results in high traffic during lunch time until 3pm. On
the other hand, people going back to their hometown, students visiting parents and people going out for shopping also
contributes for heavy traffic on Fridays. However, the traffic on road is very light on weekends due to most of the
people are resting at homes and for students as well as for the workers there is no activity. It is observed that the traffic
is heavy on weekdays because most of business operates during that period of time.

Traffic Analysis of Busy Roads

Busy roads in Kuala Lumpur were analyzed for month of October 2016. The roads were selected based on red
colour observation given in the open street map, which depicts that those roads usually are congested. Figure 4 shows
the traffic analysis based on average speed per day for Gombak Lane, Tunku Abdul Rahman Road, Masjid India Road,
Munshi Abdullah Road and Bunus Satu Road. The dotted bar with large dots shows the slowest average speed for the
month of october. The reason for the congestion is the presence of the office buildings surrounding that region. During
rush hours these areas experience congestion. Particularly for Masjid India road, it experiences congestion during
Friday prayers when people attend the mosque in huge numbers. Whereas Tunku Abdul Rahman road is considered
as the bottle neck during rush hours. This is a critical spot during the festivals and owing to the problem of congestion,
the government sometimes shuts down the TAR interchange in order to ease the traffic. On the other hand Munshi
Ibrahim road links the Dangi Wangi road to the jalan Ampang.

24 February 2024 21:27:19

FIGURE 4. Traffic Analysis For Busy Roads in October 2016 : Gombak Lane, Tunku Abdul Rahman Road, Masjid
India Road, Munshi Abdullah Road, Bunus Satu Road

This spot experiences congestion most of the times when the public activities are at their peak.

Traffic Analysis of Major Roads Near the Centre

Busy roads in Kuala Lumpur were analyzed for month of October 2016. The roads were selected based on red
colour observation in the map. Figure 5 shows average traffic speed obtained for major roads such as, Sultan Ismail
Road, Dang Wangi Road, Raja Laut Road, Doraisamy Road,Raja Abdullah Road, Medan Tuanku Road and Dewan
Sultan Sulaiman Road. In Figure 5, it was observed that 30 km/h shown with bar having small squares, is the minimum
and 40 km/h as shown in the dotted bar is the maximum average speed obtained. This is probably because the roads
are surrounded by government and private sector’s office blocks, schools, educational institutes and etc. Therefore,
the road was congested during office hours by user. However, the roads are also congested after office hours because
it is located near to nightclub spots, hotels and restaurants. Due to this reason the congestion is followed from the early
morning to the very late hours at night.

020013-5
 FIGURE 5. Traffic Analysis For Busy Roads in October 2016 : Sultan Ismail Road, Dang Wangi Road, Raja Laut Road,
Doraisamy Road, Raja Abdullah Road, Medan Tuanku Road, Dewan Sultan Sulaiman Road

Traffic Analysis Per Single Road

Table 1 shows the analysis of average traffic speed for each road separately. It was observed that Masjid India

24 February 2024 21:27:19

road has the slowest traffic flow compared to the others. Masjid India road well known for jewelry shopping and it
also located nearby to Sogo shopping complex. Doraisamy road, Medan Tuanku Road and Raja Abdullah are also on
list for heavy traffic flow. This is because Doraisamy road was surrounded by twenty-two 1940s shop houses cafes,
restaurants, shops and live music venues. Raja Laut road is the main road linking to Jalan Sultan Ismail and Jalan
Ipoh. It is surrounded mainly of commercial establishment of high-rise buildings, hotels, shopping complexes,
medium-rise office buildings and terrace shop houses. These roads experience surge of vehicles most of the time in
the day.

TABLE 1. Average speeds of each road separately

020013-6
 An overall insight to the traffic flow in Kuala Lumpur at selected locations is given in this work. The objective of
this work was achieved where Open Traffic software is deemed useful to analyze traffic speed by exporting the data
to csv file. It also shows that analysis done for specific road based on monthly and daily data. This feature is an
advantage for future research to make use the available functions for traffic management.

CONCLUSION
In this work average Traffic Speed was collected at selected places in Kuala Lumpur area for the whole month of
October (from 1st to 31st), 2016. The traffic speeds and arrival time collected from the Open Traffic software through
Grab car. The main objective of this work is to study and understand the Open Traffic platform. Traffic Engine, Traffic
Engine App and Open Street Map used to run the Open Traffic software to analyze the problem caused by congestion.
Collection of the real time traffic flow data using Open Traffic platform and analyzing the data are the other two
objectives for this project. The data is exported in csv format and then analyzed by plotting graphs using excel to
observe the pattern of the traffic. Based on the results obtained, the traffic is heavy during weekdays compared to
weekends. It was proven that Open Traffic software can be used to analyze real time traffic flow. Therefore, it is
recommended that Open Traffic can be used for traffic analysis in future work. Besides this, the data can also be used
by the private or the public sector for the traffic management.

REFERENCES
1. Peter Samuel. Issue in Science and Tech. XV, (1999).
2. Traffic congestion and reliability: Trends and advanced strategies for congestion mitigation. 6, (Federal Highway
Administration, 2005).
3. Hossain, Moazzem. Int. Seminar on Intelligent Transport Systems (ITS) in Road Network Operations. (Kuala
Lumpur, Malaysia. 2006).

24 February 2024 21:27:19

4. Yong, Michael. ”Integrated Transport Information System (ITIS) for Kuala Lumpur: The clear way to go.” 2005.
5. Roopa, T., Anantharaman Narayana Iyer, and Shanta Rangaswamy. Int. Congress on Big Data. (IEEE, 2013).
6. Fire, Michael, et al. 27th Convention of Electrical & Electronics Engineers in Israel. (IEEE, 2012).
7. Holly Krambeck, Natasha Beschorner, et al. World Combining Taxi GPS Data and Open-Source Software for
Evidence-Based Traffic Management and Planning, Open Traffic, Report No: ACS15491, September 2015.
8. Miska, Marc, et al. Proceedings of the Australasian Transport Research Forum. (2011).

020013-7