Managing traffic congestion with GIS

Managing traffic congestion with GIS

SHARE

Amudapuram Mohan Rao, Senior Scientist, Central Road Research Institute
S. Velmurugan, Scientist F and Head, Traffic Engineering and Safety Division, Central Road Research Institute
Arpita Chakraborty, Former Post Graduate Student, NIIT University

Various factors influence the speed of vehicles on the road. Mapping out these factors can help in the assessment and management of traffic congestion. In this article, GIS has been used to identify various roadside friction points that impact vehicle speed on some of the urban arterials in Delhi

Over the years, GIS has emerged as one of the most efficient technological tools in the field of transportation engineering. In this article, the capabilities of GIS have been used to identify various roadside friction points that impact the speed of the vehicle on some of the urban arterials in Delhi. In this regard, the speed of every vehicle is almost impossible to track on a roadway using the conventional methods; and hence there is a need to deploy modern technology like GPS for tracing the speed of vehicles. Consequently, the average speed is deduced based on the sample of vehicles traversing over the defined trap length spread over a period of time or area.

There are various factors that influence the speed of vehicles on the road, such as width of road, structure of the road, construction work on roads (e.g. work undertaken for Metro Rail construction), various land uses that attract motorized / pedestrian traffic bound to hospitals, institutional, commercial area etc. Mapping out these factors using GIS can help in the assessment and management of traffic congestion.

In this article, an attempt has been made to assess the influence of roadside friction location on varying widths of carriageways in Delhi region. In this regard, quantification was done for each friction point and its influence on the traffic speed. This study observed that the impacts of the pedestrian crossing roads and parking of vehicles on the roads would have a negative influence on speed varying from 19% to 64% whereas the bus stops located without the proper provision of bus bays would reduce the speed of the vehicle to the tune of 24% to 43%.

Introduction

Geographic Information System (GIS), over the years, has emerged as one of the efficient technological tools in the field of transportation engineering. It has shown great applications in a number of fields including transportation. Increasingly, urban and transportation planners and professionals are finding that the integration of traditional transportation research methods with the added value of GIS capabilities including integration of geographical spatial-analysis and cartography, provides a robust platform for both traditional and innovative transportation and traffic activities. The various advantages of GIS make it an attractive option to be used to face the emerging traffic problems. The advantage of GIS can be attributed to its capability to cope with the large volume of data with geographic spatial characteristics. GIS has a large database storage capacity, which can integrate data from disparate sources. While working with traffic speed, integrating spatial and non-spatial data from different sources becomes a prime concern. Moreover, along with great data integration capabilities, it is also a great visualization tool as it produces relevant maps assisting in decision making process.
In this article, the capabilities of GIS have been extensively used to identify various roadside friction points that impact the speed of the vehicle on some of the urban arterials in Delhi. There are various influencing factors that affect the speed of vehicles on the road, such as width of road, structure of the road, construction work on roads (e.g. work undertaken for Metro Rail construction); various land uses that attract motorized / pedestrian traffic bound to hospitals, institutional, commercial area etc. Mapping out these factors using GIS capabilities can help in the assessment and management of traffic congestion.

Objectives of the Study
The present study was done with the following objective:
• To identify the roadside friction locations on varying widths of urban, arterials and sub-arterials in Delhi region.
• To predict influence of the friction points on the vehicular speed on urban roads.
Literature Review
Greibbe, et. al.,(1999) emphasised the advantage of Global Positioning System (GPS) to map the relevant traffic parameters. This can easily be done in a GIS environment. This makes possible to make a pre-evaluation of various measures related to the local targets and which usually proves to be conclusive.
Shaopei Chen et. al., (2006) studied the problems related to the transportation aspect of a Guangzhou city in China. They presented an ideology of constructing an integrated, efficient, effective and comprehensive urban transportation information system (UTIS) for managing, monitoring and planning purposes so that its development can coincide with city’s developmental pace. For instance, integrating GIS and traffic information can be used to monitor traffic accidents, flows or congestions that have occurred in the past, this providing useful insight for transportation planning studies.

Joseph Owusu et. al., (2006) highlights the use of GIS to create an extensive database containing all the traffic information such as speed data, road geometry, traffic flow, friction points etc. The data obtained is then processed and mapped. The information obtained can subsequently be used to create a database in GIS to help in decision making for any planning process including a speed management programme.

Kalaga Rao and Mohan Rao (2009) studied the application of GPS for traffic data such as travel time and traffic speed and they validated the GPS data by conventional methods and statistically validated the results of these parameters and found that the GPS data can be used for traffic studies without compromising the accuracy of the data.
Anitha selva sofia et. al., (2013) talks about traffic congestion, which is a condition on road networks that occurs by slower, and increased vehicular queuing. To study the effect of the Transportation System Management (TSM) measures, one needs to have a clear view of the flow patterns, location as well as existing road network. GIS can be effectively used to analyse the problems associated with transportation.

Study area
At 1749 km of road length per 100 km², Delhi has one of the highest road densities in India. Major roadways include the Ring Road and the Outer Ring Road, which had a high traffic density. Total road length of Delhi is about 32,500 km including 388 km of National Highways. Owing to improper development of rail based modes in Delhi, the city is heavily dependent on road based modes of transportation (93 per cent of the total trips performed in the city are made using road based transport systems). As a result of this, the road length within the city has undergone a growth of 4.53 per cent per annum, from a mere 8380 km in 1981 to as high as 20,487 km by 1990, which at present (2013) stands at a total of 32,487 km, the highest in the country. In the present study, five locations were selected in south Delhi area, the locations are shown in Figure 1. The five locations are listed below.

FIGURE 1: Map indicating the Road Segments on Delhi Map

1. Ashoka Road
2. Lodhi Road
3. Delhi Cantonment
4. Munirka To Vasant Kunj
5. IIT Delhi To Mehrauli

Friction points
Friction factors are defined as all those actions related to the activities taking place by the side of the road and sometimes within the travelled way (like bus stops, unauthorized parking), which interfere with the traffic flow on the travelled way. They include but not limited to pedestrians, bicycles, non-motorized vehicles, parked and stopping vehicles, bus stops, petrol pumps on the side roads etc. These factors are normally very frequent in densely populated areas in the developing economies. In this study, initially friction point locations were identified on the selected road corridors and subsequently the influence of these factors on traffic performance measures were assessed. Google Earth was used to demarcate the identified road side friction points on the study corridors. The friction points observed on the study corridors are presented in Table 1.

TABLE 1: The study location and the extracted friction points

Methodology
The methodology deployed in this study is presented in Figure 2 along with a brief explanation.

FIGURE 2: Evaluation Framework for Friction Point Influence Assessment of friction points

Database preparation
A database is an organized collection of data. Databases are created to operate large quantities of information by inputting, storing, retrieving, and managing that information. Once a database is ready, it becomes very efficient to add new data fields, update existing data fields, or simply remove of delete pre-existing data. Hence preparation of an authentic database is a major concern for all application. All the relevant data was made compatible in the GIS environment.

Spatial data preparation
Spatial data is the data or information that identifies the geographic location like features and boundaries on the map. Spatial data is usually stored as coordinates and topology that can be mapped. Spatial data is often accessed, manipulated or analyzed through GIS. For the spatial data is collected in this study using the following sources:

  • Performance Box Data
  • Google Earth images

Performance box data
Performance Box is a high performance 10Hz GPS engine, which measures speed, position, acceleration and heading ten times a second. Performance Box is equipped with an MMC/SD flash memory card socket. This allows 10 Hz logging of time, distance, speed, position, g-force, lap times, and split times. Data logged to the MMC flash card can be analysed in detail using the PC software provided. For convenience, Performance Box can be connected to the USB port of a PC compatible computer to download information stored on the memory card.
Google Earth Images
Google Earth helps locate the selected study locations, and also becomes a part of the spatial database preparation mechanism as it is in relation to the geographic locations on the surface of this earth. The Google earth images were collected and recorded for further analysis of these sites. Google Earth imagery of the five selected locations is shown in Figure 3.

FIGURE 3:Google Earth Imagery of the Selected Corridors

Attribute data preparation
Attribute Data: Non-spatial data has no specific location in space. It can however, have a geographic component and can be linked to a geographic location. The data on traffic speed was collected using the Performance Box wherein the probe vehicle fitted with GPS was deployed. A typical one run data collected early morning on Ashoka Road is shown in Table 2.

TABLE 2: Typical Probe Vehicle GPS data on Ashoka Road

The speed of the probe vehicle collected on all the study corridors (summary of the average speed) is presented in Table 3 which can be easily incorporated onto the GIS environment

TABLE 3: Summary of Probe Vehicle Data of all Sections

Validation of data
After the preparation of the entire database, validation of the data was undertaken. During the process of data validation, care was exercised to check for the authenticity and reliability so as to bring in the completeness of the analysis. The checking of the authenticity by using imagery from Google earth is called field verification/ verification of data. During the course of data validation, the GPS coordinates of the Delhi map readily available were plotted. To cross check whether the GPS points collected during field survey were in conformity with the ground conditions, we imported these imageries onto the Delhi map.

Data analysis

Once the data was collected, stored into a database and then verified and authenticated, the next stage involves the analysis of the dataset. The analysis done for this study is described in the following section.

Speed profile variation

Stream speed properties: The speed of travel on the study corridors were collected by probe vehicle, multiple runs data was collected on each corridor by covering the morning, afternoon and evening peak hours and non peak hours. The average speed of vehicles on each corridor is presented in the Table 4.

TABLE 4: Average Speed from Different Runs on the Study Corridors

The difference in the average speed collected during the lean period of traffic flow and congested flow can be observed in Table 3 and Table 4. The free speed (Table 3) is collected during the following times: Lodhi Road at 5.18 AM, Ashoka Road at 5.35 AM, Delhi Cantonment at 6.53 AM, IIT Delhi at 6.00 AM and Munirka at 6.24 AM. The above speed data was collected during the early hours of the day termed as free speed (at this time of the day the driver has a freedom to choose the speed as he desires). As the time of the day increases the influence of the friction points on the roads will emerge coupled the traffic in the traffic.

Speed differentials between free speed and stream speed: In this study, it was found there is a huge difference between free speed and stream speed of the sections. The speed difference may be due to variation in traffic characteristics and friction points available on the study corridors. Table 5 shows the difference between free speed and stream speed on various sections.

Table 5: Percentage Variation of Free and Stream Speeds

From the above table, it can be inferred that the stream speed reduced more than half on all the sections, the average of 65% reductions is observed on stream speeds.

Temporal variation of speed

The probe vehicle data collected in different runs were grouped in different time slots. Four different time slots were chosen to cover the variations. The time slots are as follows;
1. Period 1 : 7 am – 9 am
2. Period 2 : 9 am – 12 am
3. Period 3 : 4 pm – 6 pm
4. Period 4 : 6 pm – 8 pm
During this time slot, the speed variation profile for all the five segment was plotted on the same graph. A typical temporal variation observed on Delhi Cantonment corridor is presented in Figure 4.

FIGURE 4: Temporal variation of segment speed observed on Delhi Cantonment corridor

From Figure 4, it can be noted that the speed is varying with respect to time and it also can be observed that the speed is low at all times at the distance between 1961 m to 2087 m, which implied that there is some influence due to friction points in the form of bus stop. Similar such data was analysed for all the study corridors and the same is presented in Table 6.

TABLE 6: Summary of speed (km/hr.) of all runs in different time slots

Identification of influence of friction points

Different runs of the vehicle made with Performance Box were plotted on a single graph using the Performance box data to understand the variation in speed. The speed profile plot was plotted by depicting the absolute distance on the x- axis and variable speed on the y- axis as shown in Figure 5.

FIGURE 5: Variation of the corridor speed due to Friction Points

Figure 5 (a) speed data collected on Munirka to Vasant Kunj segment of length of 1.15 km, where there are no friction points exists on this segment, it can be clearly seen that, there is no dip in the speed profile data in all time intervals. Hence this segment can be termed as ideal segment. The data collected other segments, there exists some friction points and Figure 5 (b) shows the dips in the speed data, the reduction in the speed is due to friction points. Figure 5 (b) shows the speed profile data collected on IIT to Mehrauli segment, it can be seen from the graph the speeds even reduced up to zero this is due to the friction due to bus stop with no bus bay on this segment. Similarly, influence of different types of friction points are quantified and presented in Table 7.

TABLE 7: Summary of Speed Reduction on various corridors due to Frictions Points

Conclusion

The following conclusions have been drawn from this study:

  • The average speed, after considering all the runs, using the performance box data were as follows: Lodhi Road 18.09 km/hr, Ashoka Road 19.36 km/hr, IIT to Mehrauli 15.75 km/hr, Delhi Cantonment area is 17.96 km/hr and the maximum speed was recorded for the Munirka segment (friction point free segment ), at 22.38 km/hr. All the road mentioned are arterials roads. The speeds on these roads are expected more or less same, but due to the friction points prevailing on the corridors, it is influencing speed reduction. The speed reduction observed on Lodhi Road was 52.13%, Ashoka Road 59.71%, IIT to Mehrauli 77.19% Delhi Cantonment area 36% when compared with the Munirka to Vasant Kunj section which is devoid of roadside friction.
  • It is observed in this study that, all the friction points will not be active for entire day. The activation times are different for each point. The jay walking resorted by the pedestrians to cross the road is observed to be more pronounced during the morning and evening peak hour traffic period whereas the bus stop friction is almost active for entire day.
  • The influence of the friction points on the traffic speed shows that the influence of the bus stops is up to 93.96% (on the IIT Delhi segment of the Mehrauli Road).
  • This study observed that the impacts of the pedestrian crossing roads and parking of vehicles on the roads would have a negative influence on speed varying from 19% to 64% whereas the.bus stops located without the proper provision of bus bays would reduce the speed of the vehicle to the tune of 24% to 43%.

Acknowledgements
The authors are grateful to the field staff for data collection support. The authors wish to thank Dr. S Gangopadhyay, Director, Central Road Research Institute, New Delhi, India for extending his motivation, guidance, suggestions and kind approval to publish this paper.

References

  • Anitha Selva Sofia Sd., Nithyaa.R, Prince Arulraj.G (2013) “Minimizing the Traffic Congestion Using GIS” IJREAT International Journal of Research in Engineering & Advanced Technology, Volume 1, Issue 1, March, 2013
  • Shaopei Chen, Cyril Ray, Jianjun Tan, Christophe Claramunt (2006) ” Analysis of an Integrated Transportation GIS for the City of Guangzhou, China” https://cyril.ray.free.fr/public/documents/analysis%20of%20transport%20of%20gz.pdf (accessed on 31-05-2014)
  • Joseph Owusu, Francis Afukaar and B.E.K. Prah, “Urban Traffic Speed Management: The Use of GPS/GIS” Conference proceeding, Shaping the Change XXIII FIG Congress Munich, Germany, October 8-13, 2006
  • Greibbe et al, 1999. “Speed Management in Urban Areas, A framework for the planning and evaluation process”, Report no. 168, pg 9-13.
  • Masatu L.M. Chiguma, (2007) “Analysis of side friction impacts on urban roads” Doctoral thesis KTH, School of Architecture and the Built Environment (ABE), Transport and Economics, Traffic and Logistics https://www.diva-portal.org/smash/record.jsf?pid=diva2:11686 (accessed on 31 May 2014)
  • Kalaga Ramachandra Rao and Mohan Rao (2009) “Application of GPS for Traffic studies” in Journal of Urban Transport Volume-8 No.1, December 2009