M. R. Jelokhani (Presenting Author)
M.Sc. Student, Dept. of GIS Eng.
Mail id: [email protected]
A. Golnarkar
M.Sc. Student, Dept. of GIS Eng.
Mail id: [email protected]
Ali A. Alesheikh
Associate Professor, Dept. of GIS Eng.
Mail id: [email protected]
Abstract:
GIS has been widely recognized to be an instrumental tool to overcome a wide spectrum of difficulties in water and wastewater utilities management. This paper is the consequence of a project that was conducted to provide a powerful tool for water utility disaster management. Due to the remarkable development of GPS and its related technologies, it has been possible to collect real-time data from the current position of a moving object whose location changes over time. In our developed system, firstly, the position of a local repairman is determined at the specific time intervals and then the network analyst unit as a LBS (Location Based Service) will find the optimal path by considering the various situations from the repairman to the damaged part of the utility network. Then a well-organized process will find the best way to manage the damaged area. For instance, the algorithm finds the most appropriate valve to be closed in order to prevent from waste of water and other possible damages. Implementation of this project is on the basis of an object-oriented data model for the representation and analysis of water utility network. The model not only includes an essential set of water object classes and properties, but also contains the rules and the relationships that define the object behaviors which are according to the reality.
Keywords:Water utility network, Disaster management, GPS, LBS, GIS
1.Introduction
When the water utility disaster such as a water main break or a blockage in the water pipe occurs such that water escapes, time is important to consider since a short waste of time can lead to significant damages in the disaster environment. In such cases, an extremely important challenge is the lack of quick emergency response of a local repairman to the damaged part of the water utility network. Second, major challenge, is the lack of most appropriate approach to manage the damaged site. For example, he cannot detect suitable valves which should be considered to be shut off in order to prevent water losing. The main motivations behind this study are to make use of the powerful Geographic Information System (GIS) capabilities and also to integrate it effectively with Global Positioning System (GPS) to develop a location-based service (LBS). To satisfy the first challenge, LBS provides the determining of optimal path on the basis of different criteria from the current position of a moving local repairman whose location changes over the time to the damaged parts of the utility network which have been frequently reported during the day by service station. To solve the second challenge, an object-oriented data model – the geodatabase data model is used. The geodatabase data model is capable of representing natural behaviors and relationships of features that enables users to implement the majority of custom behaviors without writing any code (Zeiler. 2000). As discussed later, the geometric network data model in the geodatabase can be used to model water utility network and to provide users with a rich set of network-based analysis.
2. Location based services (LBS)
Due to the rapid development in mobile communication technologies, the usage of mobile devices such as cell phones, PDAs, GPS receivers has increased significantly so that nowadays, users are more and more equipped with them. One of the most powerful approaches for personalization of mobile services is based on current location of a moving object such as people, cars, etc. LBSs are information services accessible with mobile devices through the mobile network and utilizing the ability to make use of the location of the mobile device (Virrantaus et al., 2001). LBS provides personalized services to the users on demand, based on their current position. For example, this could include information on nearest hospitals. LBS provides the possibility to answer the questions “Where am I?”, “What is near by?” or “How can I go to?”(Steiniger et al., 2005), “where is the optimal route between a given departure and destination point?”, “What is here around me?” etc. The relation of GIS and LBS can be viewed such that GIS is a complement component for LBS so that GIS provides not only the various tools that enables capture, modeling, manipulation, retrieval, analysis and presentation of spatially referenced data but also it is widely recognized that GIS provides powerful spatial analysis functions (LBS-services). This paper examines the algorithm in which firstly, the point feature as the current position of moving local repairman determined by GPS should be projected on right route. And then the cost model of route network is proposed to consider the criteria more than length and time for finding optimum path from a moving repairman’s vehicle to the damaged area of the utility network at the specific time interval.
2.1. The optimal path finding service
In order to enable quick emergency response and effective reduction of the risk resulted from water utility disaster; it is remarkably instrumental to select the optimal route based on different situation between moving repairman and disaster site. LBS encompasses a wide range of services with the common goal of utilizing current location to provide the information relating to that location. Path finding has been considered to be one of the most important services that LBS provides. This service is aimed to determine and represent the optimal route at the specific time interval (e.g. 5 min) with new origins that enables the repairman to see optimal path from its current position to disaster site (Figure 1). And also this provides the possibility to make queries like “where will I arrive in within 5 minutes?”, “what time will I arrive in there?”, “How long dose it take to arrive in disaster site?”, etc. GIS functionality can be utilized to perform advanced analyses in the optimal path finding to provide detailed navigating directions between two locations. Actually, this service provides an on-screen guidance for drivers. If the driver misses a turn, the system displays a warning message and finds an alternative best route based on the current location of the vehicle (El-Rabbany . 2002).
Figure 1: The optimal path at the specific interval (Kwan, Lee., 2003)
2.1.1. The algorithm for optimal path finding service
To implement this service, we proposed the algorithm which encompasses the stages of route finding process for moving repairman. This algorithm consists of several steps which are respectively illustrated (Figure 2). Within this algorithm, first the destination (disaster site) must be specified beforehand by the repairman. Second, it is necessary that the kind of mobile positioning device such as GPS receiver should be installed on the repairman’s car for collecting its current position that is changing continuously. And then this GPS-determined current position of repairman (x , y) is transmitted to the pocket PC containing in its database digital street map and information such as street names, directions and other related information. Third, since, the accuracy of GPS-determined location is in low level, therefore, even by considering high-precision positioning, this results in the problem that within digital street map, the GPS-determined repairman’s vehicle location (x , y) is not superimposed exactly on the road on which repairman’s vehicle is moving. Thus, it is necessary to superimpose this GPS-determined point on corresponding street which this process is illustrated in later steps. Now it is needed to create buffer (R) around GPS-determined point at the specified radius (r) so that the buffer distance depends on positioning accuracy (about 20m). Fourth, after that buffer operation is complete, the intersection of buffered area with street layer will be done to identify which streets are intersected with buffered area. Fifth, the number (N) of intersected streets is counted and then it is provided to check whether N is equal to 1 or more than 1. Sixth, for the case of N=1, the GPS-determined point is projected on the found street. Otherwise, the system will ask the user to choose from selected streets the street on which the vehicle is moving and then GPS-determined point is projected on corresponding street. Seventh, it is now possible to find easily optimum path based on cost model, in which three main criteria including length, traffic volume and speed limit are considered. Finally, if repairman is arrived in disaster site, then procedure is terminated. Otherwise, procedure after the specific time interval (e.g. 5 min) is iterated.
Figure 2: Algorithm for optimal route finding service
2.1.2. Cost model
Path finding based on only the minimal length of roads network as a cost model causes the results to be inefficient for GIS users. To overcome this problem, the method which stated by Sadeghi (2003) to determine a cost model for road network, was considered to be useful in order to combine different criteria with specific weights in optimal path finding. Three main criteria including length, traffic volume and speed limit were used (Figure 3). To combine these criteria, the Analytic Hierarchy Process (AHP) technique was applied in order to provide the costs for streets in the optimal path finding based on Dijkstra Algorithm (DA). After the paired comparison of criteria the following general formula for cost model is achieved:
Where f is the primary cost model of the streets excluding the length criterion, n is the quantity of effective criteria in cost model of the roads and Xi shows the effective criteria to cost model and K is a constant coefficient determined by experts of cost modeling (Sadeghi. 2003). It should be noted that the length criterion due to its particular importance was not considered in the paired comparison but since it has reverse relationship with F (i.e., the value of F is increased when the length is decreased) , it was separately used as the denominator of the equation above. Finally, F is:
Where L is the length of each street, T is traffic volume of each street and S is the maximum speed for each street. In this formula, T and S are available through following tables (table 1, table 2). The relative preference of one street to another is resulted from the quantitative result of F in the above formula. In other words, the street is selected which has the highest value.
Figure 3:AHP diagram of main criteria
Table 1.The level of service for traffic criteria (Sadeghi. 2003)
Table 2.The level of speed limit for speed limit criteria
3. Managing disaster site using utility network analysis
Once the local repairman is arrived in disaster site, he needs the most appropriate way to accurately evaluate and manage the disaster site. GIS has been widely recognized to be an instrumental tool in order to overcome a wide spectrum of difficulties in water and wastewater utilities management. GIS provides utility network analyst that can be used as an efficient means in order to provide the ways for managing the disaster site. In this project, the object-oriented geographic data model- the geodatabase data model was used. In the geodatabase, network is modeled as a geometric network, which is composed of features. Geometric networks, like topologies, must be created from a set of feature classes in the same feature dataset. Geometric networks allow turning simple point and line features into network edge and junction features that can be used for network analysis (Perencsik et al., 1999). A rich set of powerful utility network analyses can be performed on geometric networks that efficiently help the decision making process for managing the disaster site. One of the most crucial analyses for utility network has been found to be tracing analysis. It is performed to understand which part of network conditionally connected to a chosen node on the network, known as the trace origin. For a node or line to be conditionally connected, it means that a path exists from the node/line to the trace origin, and that the connecting path fulfills the conditions such as direction of the path (De by et al., 2000). When working with networks, tracing involves connectivity applying the specific algorithm to find trace path. Network tracing can be classified as trace upstream, trace downstream according to direction of trace path. Trace upstream finds all network elements that lie upstream of a given point in the network. In other words, the tracing conditions were set to trace all the way upstream. Trace downstream finds all network elements that lie downstream of a given point in the network that can be said, the tracing condition were set to trace all the way downstream(De by et al., 2000). In our project, we used utility network tracing to answer two questions “which valves should be shut off to isolate the break in the water main?” and “which customers will be affected during repair work?”. During emergency situations such as a water main break, first answer is aimed to enable the repairman to quickly know which valves should be closed. This rapid identifying of these minimum number of valves to be closed, reduces costly water losses, repair time and also it is intended to shut off the minimum ones which causes fewer customers to be affected by closing the determined valves. Second answer has been widely used to demonstrate which customers are affected of this break so that they do not access any water during repair work that is useful to notify these customers to find alternatives. As shown in the following figure, trace upstream and trace downstream analysis on geometric network were performed to provide respectively first and second answers mentioned earlier for managing the disaster site (Figure 4).
Figure 4. The trace upstream and trace downstream for managing the disaster site
4. Conclusion
The integration of GPS and GIS provides a cost-effective, efficient tool for water utility disaster management. To provide the quick emergency response to disaster site, using LBS with the purpose of finding the optimal path from a moving object to disaster site based on cost model will be significantly efficient. One of the most obvious advantages of this service is in which cost model combines different criteria with specific weights in order to provide the costs for streets to be used in the optimal path finding process. Since within digital street map, the GPS-determined vehicle location is not superimposed exactly on the road on which vehicle is moving, the proposed algorithm superimposes GPS-determined vehicle location on corresponding street. GIS provides utility network analyst that can be used as an efficient means in order to provide the ways for managing the disaster site. Geodatabase data model is used in which geometric network models utility network that can be used for network-based analyses which enables repairmans to manage disaster site.
6. References
- De by, R., Knippers, R., Sun, Y., Ellis, M ., Kraak M ., Weir, M .,Georgiadou, Y., Radwan , M ., Westen, C., Kainz, W., Sides, E., 2000. “Principles of Geographic Information Systems” , ITC Educational Textbook Series , Enschede, The Netherlands, 230 Pages.
- El-Rabbany, A., 2002. “Introduction to GPS: the global positioning system” , Artech House Publishers.
- Kwan, M., Lee, J., 2003. “Emergency response after 9/11: the potential of real-time 3D GIS for quick emergency response in micro-spatial environments”, Computers, Environment and Urban Systems , 29 , 93-113.
- Perencsik, A ., Woo, S ., Booth, B., Crosier, S., Clark, J., MacDonald, A., 2000. “Arc GIS : Building a Geodatabase” , Esri.
- Sadeghi, A., 2002. “Defining the Cost Model of Iran Roads Network in GIS” , GIS Msc. Thesis , 117 pages.
- Steiniger, S., Neun, M., and Edwardes, A., 2005. “Foundations of Location Based Services”, Lesson 1, Lecture Notes on LBS, V. 1.0.
- Virrantaus, K., Markkula, J., Garmash, A., and Terziyan, Y.V., 2001. “Developing GIS-Supported Location-Based Services”, In: Proc. of WGIS’2001 – First International Workshop on Web Geographical Information Systems , Kyoto, Japan , 423-432.
- Zeiler, M., 2000. “Modeling Our World: The ESRI Guide to Geodatabase Design” , ESRI Press.
