By: Ng’ang’a Titus, Kirira Dominic, Wango Tim, Mbui John

The purpose of the project was to avail route guided interactive maps to motorists and hence deploy spatially enabled location-based services as a platform for improved services in case of emergencies such as running out of petrol. This was done by showing how the analytical and visualization capabilities of GIS can enhance decision making in transportation through mapping of petrol stations locations. Street-level geocoding and optimized routing gives in-vehicle navigation capabilities to assist motorists in finding service locations easily. Optimal routes generation reduces the distance of travel and hence fuel consumption. The interactive map allows the motorists to find the nearest petrol station. Searching is by the way of incidences. The incidence is shown by a computer guided Global Positioning System. The motorist clicks on the current position/incidence and then chooses nearest petrol stations.

The queries performed on the multimode road network dataset are evident that mapping of petrol stations and network analysis can lead to informed decision making. Similar decisions could be achieved through the same concept (such as locating the nearest hospital by ambulances).

Multimode Transport Network
Networks are the interconnected features that are used for transportation and include highways, avenues and city streets. Networks are an important part of everyday lives and analysis of these networks improves the movement of people, goods, services and the flow of resources (Nancy E, 2003). Networks give the means for the movement of people, the flow of resources and energy and the communication of information (Haggett & Chorley, 1969). Network analysis in GIS provides good decision support for users interested in finding the nearest facility (Pahlavani P. & Samadzadegan F., 2006).When the linear features are joined together to form a single transportation network they are regarded as a multimode infrastructure.

Multimodal networks allow organizations in both the public and private sectors to better perform transportation planning analysis and accessibility modeling (http://www.esri.com/software/arcgis). In this project, highways, avenues and streets have been integrated to form the multimode transport network.

Shortest path in networks
A path between two vertices that minimizes a pre-defined metric such as the total number of steps, total distance or time, is called a shortest path. Determination of shortest paths is often described as shortest path analysis (De Smith et al., 2009). To determine the best way one needs at a minimum an origin and a destination. (Jochen A., 2007). The problem of identifying the shortest path along a road network is a fundamental problem in network analysis, ranging from route guidance in a navigation system to solving spatial allocation problems (Zeng W. & Church R., 2009).

In order to make the output more meaningful, the highlighted route is also described with regard to details like the road to start, the roads to traversed, turns to left or right and distance of travel along each road (http://www.gisdevelopment.net/application/Utility/transport).

GIS and GIS-T

Traditional approaches are no longer adequate for analyzing network flows and conducting minimum cost routing.  GIS provides effective decision support through its database management capabilities, graphical user interfaces and cartographic visualization (Yi-Hwa W. et al., 2001).

GIS-T refers to the principles and applications of applying geographic information technologies to transportation problems (Miller H. & Shaw S., 2001). Bus companies can find the best routes by integrating GIS technology with GPS. Bus drivers can use GPS to find locations. Integration of GIS technology with GPS allows trucking companies to reach locations quickly (Hossein B., 2003). A country’s transportation system represents development stage of that country (Mukti A. et al., 2005).

Finding an efficient route is a difficult problem for many drivers. Car Navigation Systems are sometimes offered as a special feature on new cars. These systems are capable of performing some of the tasks traditionally performed by driver, such as determining the best route to the destination. (Nazari S. et al., 2008).

Transportation infrastructure represents one of the largest and most critical investments made in any nation, at any stage of development. The movement of people and goods either domestically or internationally is vital to every aspect of that economy. GIS can be used to determine the location of an event or asset and its relationship or proximity to another event or asset. Information on bus routes, current location, subway stop location, emergency situations and locations, track condition, demographic changes, and employment centers are all factors that can be used to improve transit performance (Syed A., 2004).

GIS and Networks
GIS contain data related to location points, lines (commonly roadway links and corridors), and polygons. Analysis tools that are part of GIS software packages can be used to relate these data. The use of GIS to manage data can simplify the analysis of transport systems and can enhance the decision-making process (http://www.worldbank.org/transport/roads).

The computation of shortest paths is often a central task because shortest path distances are often needed as input for "higher level" models in many transportation analysis problems such as facility location, network flows, vehicle routing and product delivery (http://www.publish.uwo.ca/-jmalczew).

Because of the spatial nature of most transportation data, transportation professionals find GIS to be a powerful tool to construct and analyze transportation networks and to conduct impact assessment of transportation facilities (Zhong-Ren P. & Edward A., 1998).

Displaying the road network on a computer monitor is a very effective and efficient tool in observing the relationship between the spatial and physical attributes of roadway facilities. The ability of GIS to produce coloured maps has provided a visual dimension for travel demand analysis (Mezyad A, 2001). The application of GIS to a diverse range of problems in Transportation engineering is now well established. It is a powerful tool for the analysis of both spatial and non-spatial data and for solving important problems of networking (Mukti A. et al., 2005).

GIS can be used as an effective tool in Managing and Planning transportation (Siddeswar P., 2003). GIS is successfully used for route planning and analysis, bus dispatch and emergency response, automatic vehicle location and tracking, paratransit scheduling and routing, bus stop and facility inventory, accident reporting and analysis, demographic analysis and route restructuring, and transportation planning and modeling, among others (http://www.mvcommission.org/GIS_for_transportation).

GIS has been recognized for many years now as an invaluable tool for managing, planning, evaluating, and maintaining transportation systems. As the gateway to economic development and, subsequently, a healthy economy, transportation infrastructure represents one of the largest and most critical investments made in any nation, at whatever stage of development. Similarly, for many firms in the transportation industry, profitability and a strong competitive position depend on a safe and reliable system. Roads are the main arteries of a modern society’s infrastructure, contributing heavily to the distribution of goods and persons. GIS provides many helpful applications for ensuring a smooth transportation flow. Customer satisfaction, competitive position, timely response, effective deployment, and profitability are all positively affected (GISDATA Group, 2009).

Methodology
Area of study
The Nairobi Central Business District (Fig 1) takes a rectangular shape, around the Uhuru Highway, Haille Selassie Avenue, Moi Avenue and University Way. It includes many of Nairobi's important buildings, including the City Hall and Parliament Building. However, some of the surrounding areas were included in order to support easier analysis and interpretation.

Data Collection and Preparation
The handheld Garmin GPSmap 60cx GPS receiver with an accuracy of between 3-5 meters was used to pick the locations of the petrol stations. The data consisted of grid coordinates referenced by the UTM WGS 84 Zone 37 0 S projection. Once the location of a petrol station was observed, it was stored within the GPS and downloaded later. Geocoding of the petrol stations was done in Arcview.

The study involved fieldwork in which the petrol stations were visited and their geographic coordinates picked using hand held GPS. Attribute data were obtained from the petrol stations attendants. The directions of the one way roads/streets were obtained from google maps.

The primary data collected required editing. ArcGIS 9.2 suite application ArcMap was used for editing the data.

Data Capture
In this study, the handheld Garmin GPS map 60cx GPS receiver was used to pick the locations of the petrol stations and had an accuracy of between 3-5 meters. The map of Nairobi district analogue was scanned into soft copy so as to result into a portion of it, the CBD. The georeferenced CBD map was created. Before scanning, the document was well prepared to ensure that line widths are resolvable and unwanted data was cropped out.

All the physical locations of the CBD, including roads (linear feature) were digitized. The end result was a digitized map of Nairobi CBD that contained roads network and buildings. Digitizing introduced errors eg undershoots and overshoots. Linear features for network analysis required thorough editing to close gaps and disseminate orphaned junctions. Spatial network editing involved error correction such as closing gaps. Broken road network lines were connected using line snapping by specifying gap threshold value to connect lines that matched the snapping criteria. Line smoothing was done using the generalization technique to remove artifacts caused by image scanning. The line editor was used to add missing lines manually. It was also used for line merging and splitting, node editing, and line labeling functions.


Query Analysis
A key component of ArcGIS 9.2 Spatial Analyst is the ability to perform queries. The query functionality gives the analyst the ability to leverage existing data and to make more informed decisions (ESRI White Paper, 2001).

The key to network representation is to represent nodes, arcs and network topology efficiently. Once the nodes, arcs, and network topology are efficiently represented, other data and information associated with nodes, arcs, stops and turns can be represented as attributes either associated with nodes or arcs (Benjamin F., 2008).

When a geometric network is created, ArcGIS 9.2 also creates a corresponding logical network, which is used to represent and model connectivity relationships between features. The logical network is the connectivity graph used for route analysis. In the project, the route analysis operation was finding the shortest route from an incidence in terms of distance (Impedance).

The logical network allows ArcGIS 9.2 to quickly discover and model the connectivity relationships between connected edges and junctions in a geometric network during editing and analysis. This allows for fast network tracing and facilitates the generation of on-the-fly connectivity while editing. When edges and junctions are edited or updated in the geometric network the corresponding logical network is automatically updated and maintained as well (Benjamin F., 2008).

Finding the nearest Petrol Station(s)
Finding the nearest facility is a multimode infrastructure network dataset query type of analysis. A motorist in a hired GPS fitted vehicle may need to locate the nearest petrol stations. The GPS fitted vehicle shows the right position (Incidence) of the motorist. The motorist enters the incidence within a click of the mouse and enters the number of petrol stations he/she would like to view.

Results and analysis
The main objective of the study was to demonstrate how GIS application could lead to decision making. In order to achieve that objective the facilities considered were petrol stations within the NCBD.  A road network dataset was generated. The analysis of the results was done through spatial queries into the road dataset. The analysis involved identifying the closest facility from a location, tracing the best route to the facility and step-by-step directions along the identified route.

A map showing conspicuous and independent petrol stations within the NCBD was generated.  A motorist/tourist at an incidence outside Hilton hotel was considered and a route to the nearest petrol station was generated through running network analysis (Fig 2). Fig 3 shows the directions from Hilton hotel to Shell, Latema Road, the closet petrol station.

Conclusions
The results of the study were based on the data collected and the analysis undertaken. The analysis mainly involved multimode infrastructure network dataset query for motorists. The results showed a successful completion of data manipulation. The graphical output is in form of maps indicating the route to be traversed along with the distances and directions to be traversed along each road segment.

The queries performed on the multimode road network dataset were satisfactory and demonstrated how mapping of petrol stations and network analysis can lead to informed decision making. The study illustrated the application of GIS in finding the optimal route between the given origin and destination.

The study demonstrated that similar decisions could be achieved through the same concept, e.g ambulances locating the nearest hospital in case of emergencies. This concept as well applies to planes that would need to take an emergency landing on the nearest airport/airstrip during times of emergencies thereby avoiding crash landing.

The results of application of similar projects case studies depict numerous benefits including minimizing travel time, minimizing driving distances, reducing fuel consumption, providing driving directions to new drivers, estimating drive time and increasing the number of trips.