METHOLOGY FOR CITY LEVEL URBAN ROAD NETWORK CONNECTIVITY ANALYSIS USING GEOGRAPHICAL INFORAMTION SYSTEMS (GIS)

In most of the urban cities in India, the existing road network structure is unable to meet the increasing travel demand. In this context, there is a need to utilize the existing road network in a more efficient manner. To effectively utilize the existing road network, the structural parameters like connectivity, accessibility, hierarchy and morphology should be analyzed and evaluated. This study aimed to analyze and evaluate the road network structure of Hyderabad city, Telangana state, India in terms of its connectivity. Different places in Hyderabad city were considered to quantify the connectivity measures such as alpha index, beta index, gamma index, eta index, Cyclomatic number and Aggregate transportation score. Five kilometers radius buffers were drawn from the center of each location to compute the connectivity measures. The results of the study may act as a guide to the transportation planning authorities to understand the level of connectivity at each location in the city and implement better planning practices to improve the level of connectivity in the city. Geographical Information Systems (GIS) is a platform used for better decision making in urban transportation planning. The present study also tried to prove the efficiency of GIS in analyzing the connectivity based performance of the transport network structure in the study area.


INTRODUCTION
In the urban framework, the transportation structure is very important and helpful in the economic development of a place. The performance of a transport structure represents the quality of engineer's adherence to that place. A well organized and well-connected transport network is very substantial for continuous progress of an area. To enhance the performance and quality of the transport structure, now-a-days, specifically integrated system like Geographical Information System (GIS) is being availed.
The essential ingredients for any road network are connectivity, accessibility, hierarchy and morphology. The ease of travel to desired destinations is facilitated in a well-connected street structure. Different modes of transport, different speeds and volumes of vehicles are facilitated in a well-organized hierarchical road structure [1].
A connectivity indicator is definitely needed to evaluate how well the street structure is organized and interconnected. This can be easily quantified by the ratio of vertices to the edges in a structure [2]. The higher values of connectivity indicators represent a satisfactory Network analysis is helpful for identifying ideal ways for accommodating emergency utilities in Varnasi [4]. Transport network indices were quantified using Arc GIS software which resulted a low level of interconnections between roads in Karsagod taluk, Kerala [5]. The application of network analyst tool of Arc GIS is explained as avail for finding solutions to many network problems relevant to transport structural performance with an example of south west Delhi [6]. Transport structural performance based on connectivity is done using some connectivity indices which proved that Aurangabad has satisfactory network facilities [7]. For Dehradun city, network analyst tool of Arc GIS was used to identify solutions to many problems in transport structure [8]. There are three important indicators measuring connectivity like Alpha, Beta and Gamma proposed by Kansky 1963. With the help of these indicators, the transport structure performance based on connectivity of roads can be explained easily [9]. The road network system of Trans-Amadi, Port Harcourt in Nigeria was assessed in terms of Beta index, road density, level of connectivity, the condition of road and road types [10]. The relationship between interconnectedness and land cover changes was investigated for Lop Buri province; Thailand using graph theory based network indices [11]. The performance of the transport network structure in terms of its connectivity was analyzed using structural measures in Cooch Behar District; West Bengal, India [12]. An attempt to develop GIS based rural road database was done so that planners, decision makers, researchers and other different level authorities in the rural road sector will be benefited from the final output. The database was developed based on the structural analysis of the network in terms of connectivity, average Pavement Condition Index values of the link, shortest path analysis [13].

AIM AND OBJECTIVES OF THE STUDY
This study is aimed to analyze the efficiency of the road network structure in Hyderabad city, capital of Telangana state, India in terms of its connectivity, while the specific objectives are:  To compute different connectivity indices for each location at five kilometer radius.


To evaluate the level of road network connectivity for each location based on Aggregate Transportation Score.  To establish locations with the less connected, moderately connected and well-connected network.

METHODOLOGICAL FRAMEWORK
The flow chart in Figure 1 presents the methodology used in the study to achieve the above mentioned objectives: The number of nodes and links for each location at five kilometers buffers were computed using Arc GIS 10.1 software. The nodes and links along with the 5 kilometer radius buffer for the locations Abids and Ameerpet are shown in Figures 3(a) and 3(b). Similarly, nodes and links of all the remaining locations were identified and computed using Arc GIS 10.1 software. The road network connectivity in the city is analyzed using measures of connectivity like the total number of nodes, the total number of links, alpha index, beta index (Link -Node ratio), gamma index, eta index, Cyclomatic number, Aggregate transportation score. The measures of connectivity applied in this study are described in Table 1.

RESULTS AND DISCUSSIONS
The measures of connectivity like alpha index, beta index, gamma index, eta index, Cyclomatic number (CN) and Aggregate Transportation Score (ATS) are computed for thirty locations at 5 kilometers buffer radius and the results are presented in Table 2. Alpha index gives the percentage of maximum desirable circuits in the network and its value ranges from 0 to 1. Zero indicates minimum connectivity and one indicates maximum connectivity. The alpha index value for thirty locations ranges from 0.08 to 0.74. The mean value of alpha index which is 0.36 indicates a very low level of circuitry in the city network.

Tab. 2-Descriptive Statistics of connectivity measures
Beta index value also ranges from 0 to 1. Zero indicates minimum connectivity and one indicates maximum connectivity. Beta index value greater than 1 indicates a greater complexity of the network connectivity. The value of beta index in all the locations is greater than 1; this shows the complex nature of the road network in the city. Ewing (1996) suggests that the beta index value of 1.68 (mean value of beta index) is a good target for planning purposes.
Gamma index is normally expressed as a percentage of connectivity, so the mean value of gamma index which is 0.57 indicates that the network in the study area is 57% connected.
Higher Cyclomatic number and higher aggregate transportation score values result in higher degree of connectivity and efficiency.

Evaluation of locations based on the Aggregate Transportation Score (ATS)
Aggregate Transportation Score (ATS) is the addition of alpha index, beta index, gamma index and Cyclomatic number. So, ATS can be able to interpret the overall connectivity of a region or location. Higher aggregate transportation score value results in higher degree of connectivity and efficiency. Based on the ATS values obtained, this study tried to categorize and evaluate the thirty locations selected for connectivity analysis. The road network in different locations is categorized as less connected network, moderately connected network and well-connected network. The level of connectivity according to the ATS values is categorized in Table 3. The locations selected for road network connectivity analysis are evaluated based on their ATS values and level of connectivity in each location is presented in Table 4. Based on the results presented in Table 4, the locations Jeedimetla, Nadergul, Sikenderguda, Bhuvanagiri, Ghatkesar, Bhanur, Moinabad, Ramachandrapuram, Shivampet, Sangareddy, Wargal, Domarapochampally, Sitharampuram, Kothur, Golkunda khurd, Thummaloor, Ibhrahimpatnam have a less connected network. The transportation planning authorities may concentrate more on these locations to improve the connectivity of these places which increases the overall connectivity of the Hyderabad city.

CONCLUSION
Connectivity is one of the most important structural parameters to be analyzed and evaluated for an urban road network. This study shows some conclusions based on a partial analysis of the transportation problem in Hyderabad and shows a possibility of using GIS. This study majorly focused on evaluating the structural connectivity of the road network. This study attempted to analyze the connectivity at a micro scale using buffers of 5 kilometers radius at each location. This is helpful in understanding the network connectivity in each location of the city. This study described the road network connectivity with the avail of graph theory based network indices with respect to nodes and links. The study also tried to propose a methodology for the evaluation of different locations in terms of road network connectivity based on the Aggregate Transportation Score. The level of connectivity for each location is categorized as less connected, moderately connected and well-connected network. The study revealed that many locations of Hyderabad city have a less connected network. Therefore, the performance of the road network in terms of its connectivity should be improved by the transportation planning authorities. The results of the study are also useful for the urban transportation planners to understand the situation of connectivity at different locations in the city. The enhancement or improvement of road network has to be done for the locations where there is less-connected and moderately-connected network. There is a need to implement better transportation planning practices in Hyderabad city to increase the connectivity of the road network which increases the accessibility and reduces the travel time. But connectivity is certainly not just about a simple network of streets and intersections. So, the future studies may also include assessing the width of the streets, the number of people living in the locality, etc., to look into accessibility as a wider problem. The integration of GIS with urban road network connectivity analysis gave better results and is helpful in better decision making. This kind of study may be done for all the urban cities so that the planners or decision makers may estimate the level of connectivity and take necessary steps.