SPATIAL DECISION SUPPORT SYSTEM VIA THE WEB Arif Cagdas AYDINOGLU 1, Tahsin YOMRALIOGLU 2 Departmant of Geodesy & Photogrammetry Engineering, GISLab Karadeniz Technical University, Trabzon, TURKEY
[email protected],
[email protected] Keywords: Geographical Information Systems, Spatial Decision Support Systems, Internet/Web/Online GIS Summary Geographic Information Systems (GIS) have become an effective tool for decision support. Spatial Decision Support Systems (SDSS) are decision support systems where spatial properties of the data to be analyzed play a major role in decision making. Maps and geographic features can be used to show decision related information and relationship between objects to solve important problems. A combination of technological improvements regarding GIS and web leads to the modern management support environments. Data can be maintained and be updated in a centralized location or distributed locations. Many users can reach this information in a common platform as a web browser. With the example of Trabzon city of Black Sea Region, city planning, land use, topography, cadastral, agricultural, and forestry data were collected. Spatial Data Infrastructure was built for Trabzon city. The capture of information for this city is essential for supporting operational and corporate decision making with the role of web technologies. The information about a regional spatial data infrastructure’s planning, management, and implementation is given. 1. Introduction Many factors have led GIS to be accepted as a decision making tool. Environment management for decision making and future planning is current topic in global and regional works all over the World. Constructing Spatial Data Infrastructure in wide extend areas provides many opportunities for controlling natural resources and environmental changes. At this point, the using of developing information technologies such as GIS, Remote Sensing (RS), and the Internet is considerably important. The use of Internet has changed the way organizations use geographic information. GIS has begun to appear on the World Wide Web ranging from simple demonstrations to more comprehensive internet GIS and spatial decision support system. A web based service named as internet GIS is described as a network-centric GIS tool using the Internet to access and transmit data and the analysis tools to enhance the visualization and integration of spatial data. This service makes relevant data more accessible to the politicians, the public, and municipal administration. Capturing information for a city with web technologies is essential to support operational and corporate decision making process. In this study, the importance of GIS as a decision support system over the web is explained. Building a regional database was examined for Turkey as a prototype. Trabzon City has been selected as a pilot area. This study is an initiative to establish a nation-wide Geographic Information System. Requirements for Internet GIS application are tested. Building a web service is traced. 2. Decision Support Systems (DSS) and development of the DSS concept Information systems can significantly support managerial decision making and help in the intelligence stage by providing information about different conditions. Decision support systems (DSS) are a major category of management information systems and computer-based technology solutions that provide interactive information support to help complex decision making and problem solving. [5, pp.318-332] Classic DSS tool design is comprised of components for (i) sophisticated database management capabilities with access to internal and external data, information, and knowledge, (ii) powerful modeling functions accessed by a model management system, and (iii) powerful, yet simple user interface, designs that enable interactive queries, reporting, and graphing functions. Much research and practical design effort has been conducted in each of these domains.
GEOMATICA 5 "Cartography, Telematics And Navigation"- Barcelona, Spain, February 11th-14th, 2003
Since the early 1970s, DSS have evolved enormously. DSS once utilized more limited database, modeling, and user interface functionality, but technological innovations have enabled far more powerful DSS functionality. [1] The evolution of information technology infrastructures parallel the three eras of growth in the computer industry. These are the data processing (DP) era, the microcomputer era, and the network era [2] A DSS was defined as a computer system dealing with a problem where at least some stages are structured, semistructured or unstructured. Because decision makers must also deal with unstructured portion of a DSS problem, flexible query language and modeling environment are needed for decision support. [3] DSS once supported individual decision-makers. Beginning in 1985, group decision support systems (GDSS) evolved to provide brainstorming, idea evaluation, and communication facilities to support team problem solving. This resulted in appearing artificial intelligence and expert systems. [4] Beginning in the early 1990s, four powerful tools emerged for building DSS. The first new tool for decision support was the data warehouse. The two new tools that emerged following the introduction of data warehouses were on-line analytical processing (OLAP) and data mining. The fourth new tool set is the technology associated with the World Wide Web which is the most permanent technology in the beginning of 21st century. 3. Spatial Decision Support Systems (SDSS) and GIS A SDSS is a DSS in which the spatial dimension of the data is fundamental to the analysis of decisions. Spatial decision support relies heavily on maps, which form the backbone upon which plans and policies are defined. Problems can roughly be classified into siting ( i.e., WHERE to place some given object e.g., a dam, a house, a park) and spatial allocation (i.e., for a predefined location, WHAT is the best object among a class of objects to place there e.g., a crop or building type). In the first case, the main issue is determining the location, whereas in spatial allocation the unknown is the object itself. Some problems may require combination of both characteristics (e.g., in routing or urban expansion planning) . Environmental planning, furthermore, involves studies of risk/impact assessment, and contingency planning, which combine WHAT and WHERE to WHEN and HOW. [14] Spatial decision making has traditionally been associated with the use of GIS. Muller [13] identifies SDSS as a growth area in the application of GIS technology. A system to support these decisions must be capable of effectively handling this spatial data. The display and manipulation of spatial information on a computer is usually achieved using GIS. It provides mechanism to store, analyze, manipulate and visualize georeferenced data. GIS products are increasingly available on inexpensive personal computers. These products offer significant GIS functionality at a much lower cost than traditional workstation based software. While some of these PC based GIS systems are less powerful than their workstation based equivalents, the subset of functionality offered is often quite suitable. With the increasingly use of GIS, the possibility exists of developing a SDSS to be used on relatively inexpensive equipment with which potential users can familiar. [6] The growth of SDSS applications is also made possible by the growing availability of spatial data for popular GIS software, thereby reducing the costs of data collection. [12] 4. The Web environment as a decision making tool In the 21st century, the Internet, the web technology is the center of activity in developing DSS and can be expected to result in organizational environments that will be increasingly more global, complex, and connected. The advent of the Web has enabled inter-organizational decision support systems, and has given rise to numerous new applications of existing technology as well as many new decision support technologies themselves. They are referring to a computerized system that delivers decision support information of decision support tools to a manager of business analyst using a Web browser. The web environment is emerging as a very important DSS development and delivery platform. The only text and images could be browsed over the Internet some years ago. HTTP (Hyper Text Transfer Protocol) and CGI (Common Gateway Interface) programming technologies initiated preliminary applications.The primary Web tools are Web servers using Hypertext Transfer Protocol (HTTP) containing Web pages created with Hypertext Markup Language (HTML) and JavaScript accessed by client machines running client software known as browsers. This environment has started to develop in 1990 by Tim Berners-Lee. The first Internet GIS applications have started to develop in 1994. While internet technologies were improving, web based applications have continued to develop. New technologies like Java, ActiveX provide alot of new opportunities for Internet GIS. A variety of programs which different vendors have created are available nowadays.
GEOMATICA 5 "Cartography, Telematics And Navigation"- Barcelona, Spain, February 11th-14th, 2003
General working principles have to be explained to understand application components. Client and server communicate across a computer network either on Internet or Intranet by means of a HTTP. It lies in creating software systems that are platform independent and run on open TCP/IP (Transmission Control Protocol/ Internet Protocol) based networks. A web browser such as Netscape Navigator or Internet Explorer is a usual client program. The browser program sends a request to the server and finally a file is transmitted from the server to the client by means of URL (Uniform Resource Locator). Different strategies having advantages and disadvantages appeared for employing Internet applications. These are server side, client side strategies or hybrid strategy including the common features of other strategies. Server-side strategies allow users (client) requests to a Web browser. The server processes the requests and returns data or a solution to the client. Client side strategies allow the users to perform some data manipulation and analysis locally on their own machine. Hybrid strategies combining server and client processes optimize performance and meet special user needs. [9] Internet has many technical advantages for GIS or SDSS. Data can be maintained and be updated in a centralized location or be integrated with many sources on broad spectrum platforms. A web-based map can be used both privately and publicly. Access can be regulated and redundancy eliminated. The end-user software is an easy and cheap web browser, not an expensive and complex GIS program. The web browser provides a much more dynamic map tool than a static map display. System may support various the end users. When people save and publish these data on the Internet, other people can access and browse these data simultaneously. Maps can be copied easily on the Internet and be downloaded for a more detailed use. [8, 15] Over the past 20 years, one of the most significant trends has been the evolution from individual stand-alone computers to the highly interconnected network environment of today. Initially, computers within firms were connected via local area networks (LAN s), allowing teams and workgroups to share decision-making information more easily. Then, firms began to connect their networks in wide area networks to facilitate sharing of information across organizational boundaries. Finally, the Internet and Web created an environment with almost ubiquitous access to a world of information. [1] At the same time, many organizational decisions migrated from individual decisions to ones made by small teams to complex decisions made by large diverse groups of individuals within a firm or even from multiple firms. 5. Internet GIS Development Strategy with the example of Trabzon The city of Trabzon is situated between 39o 7’ 30’’ - 40o 30’ east-longitudes and 40o 30’ - 41o 7’ north-latitudes in the middle of East-Blacksea Region of Turkey. (Image.1) Trabzon City has been selected as a pilot area for building a regional database. Building spatial data infrastructure for Trabzon City provides a national wide approach. Because there are three base local administration systems which are province, county, and village in Turkey, provinces constitute the basic administrative parts of the country. Province system is accepted as a base in view of central management, eliminating regional development differences, and providing a national wide balanced improvement. This administration forming integrity in its constitution is a small model of national administration. This system, therefore, can be used as a spatial decision support system for Turkey.
Image.1: The location of Trabzon in turkey 5.1 Database Design The first step for this process is determination of user requirements. Database design is executed, depending on these requirements. (Image.2) A planning process is executed about what kind and quality of data should be involved in such a system. A regional wide Spatial Data Infrastructure is built because it reflects all country structure including a variety of data.
GEOMATICA 5 "Cartography, Telematics And Navigation"- Barcelona, Spain, February 11th-14th, 2003
GEOGRAPHICAL DATA LAYERS
Forest
Geology
Climate
Arrival
Settlement
Soil
Streams
Forest Cadastral Map
Lithological Units
Meteorology Stations
Highway
Province and County Borders
Agricultural Fields
Lake, Pond, Dem-water ...
Forest Cover Type Map Forest Sit Map Forest Admin. Borders National Park Forest Protection Areas
Fault
Rain Dispersion Snow, Rain, Moisture ...
State Highway
Admin. District, Village Borders Administrative Centers
Sub-soil Group Erosion Map Main Soil Groups Slope - Depth Combination Vegetal Production Areas Vegetation Cover Actual Landslide Areas
Water Pollution Water Pollution Underground Water Hydrogeology Thermal Water Fresh-water Resources
Fold
Village-way
Engeneering Geology
Railway
Plateaus
Decomposition
Maritime Lines
Tourism Centers Industrial Areas
Base Layers Geodetic Referance Topography Map Index
Digital Elevation Model Slope Map
Potential Eresion Map
Aspect Map
D 885 E7
Potential Landslide Settlement Suittability
Tourism Map
RCDB
Road Map Avalanche Map
Regional Geographic Database
Geomorfology Potential Land Use Classification Map
Land Use
Image. 2: Cartographic design of Regional Spatial Database [7]
GEOMATICA 5 "Cartography, Telematics And Navigation"- Barcelona, Spain, February 11th-14th, 2003
5.2 Data Collecting and Processing In producing region wide geographical data, it was took advantage of high data gathering capacity of the Remote Sensing techniques, and in processing gathered data, it was gained from high power querying and analyzing facilities of GIS. The point that should firstly be paid attention is to classify this data in accordance with required standards, accuracy and scale in acquisition process of the data sets. In Turkey, not forming a standard between public foundation in respect of authority and responsibility in producing spatial data up to now, lots of data types are acquired by different institutions with different quality, scale, accuracy and standards at the same time. Therefore, the proper information for this study that comes from different sources was integrated with each other. By taking advantage of the features of Landsat ETM+ images used most in region wide works and acquiring data related to natural resources, missing or lacking data were gathered Land cover types were obtained by classifying Landsat ETM+ images in ErMapper software. Supervised classification technique was used for this process. Area proportion for each class was calculated. A 3D model of this area in TIN data model was formed and converted to GRID data model. Digital Elevation Model (DEM) obtained with resolution of the image by re-sampling was overlapped with image. Administrative and demographical structures were represented with the using of graphic and non-graphic data in database. Various queries and analysis that provide many opportunities for investment and services can be done for different purposes with these databases. 5.3 System Design ESRI ArcIMS Architecture has been engineered to serve geographic data and services on the Internet for Trabzon City. Microsoft IIS is used as a web server. ArcIMS has a multitier architecture consisting of presentation, business logic, and data storing tiers. The business logic tier including server components are used to process requests, create and run MapServices, and manage the site. The data storage tier includes the sources of data. Communication between the tiers is handled through ArcXML a specified XML language. Client viewer processes the data on the client machine performing many tasks without further interaction with the server. Basicly, ArcIMS work can be phased over six steps. (Image.3) Firstly, The client over the Internet sends a request to the web site. Secondly, the Web server receives the request and passes the connector. The connectors provide a communication pathway between a Web Server and The Application Server. Thirdly, the request handed from the connector providing communication for request and response to the Application Server. The Application Server handles the load distribution of incoming requests. Fourthly, The Application Server distributes the requests to appropriate The Spatial Server. The Spatial Server processes requests for maps and related information. When a request is received, Spatial Server performs functions such as creating map image files, streaming map features, feature extraction, geocoding, querying. Fifthly, The Spatial Server creates response. Finally, the response returns through the reverse order of the initial request. [8] Internet Web Browser Presentation Tier Web Server Microsoft IIS on Windows2000
Servlet Connector Application Server
ARCIMS ArcXML Spatial Server Business Logic Tier
Data Storage Tier
TRABZON Data Sources
Image.3: ESRI ArcIMS Architecture
GEOMATICA 5 "Cartography, Telematics And Navigation"- Barcelona, Spain, February 11th-14th, 2003
Active X Connector is used to develop specific applications, depending on user needs. Microsoft ASP (Active Server Pages) as a COM application is used for ActiveX Connector. (Image.4) This implementation allows for processing on the server side. The client makes method calls to the Connector Object Model API. The ActiveX Connector receives this information and translates the request to ArcXML. The ArcXML request and response is handled by the ActiveX Connector, and an HTML page is generated on the fly using ASP. The advantage to this scenario is that the client viewer does not need to generate a request or parse the response, making the client much thinner.
Image.4: Using the ArcIMS ActiveX Connector [10] 5.4 Testing the system This web site can be reached from anywhere all over the world. A user may have only a web browser such as Internet Explorer to benefit GIS opportunities. The users include planners and administrative personnel, organizations, politicians, educational institutions, public, industry/companies. [11] The web page includes a template including the map layers display, a legend column that displays an explanatory table of the symbols appearing on the map. Users can use the traditional GIS functions and search links created by ASP. When a client sends a URL request to the server that hosts esrimap.dll. Esrimap.dll determines which map service communicates to respond the request. Each argument and value pair is referred to a query parameter. Arguments transmit the information to the application which function will be executed by the GIS application. The system transmits an HTML page (Image.5) to the user for the request.
Image.5: A user window over the web browser
GEOMATICA 5 "Cartography, Telematics And Navigation"- Barcelona, Spain, February 11th-14th, 2003
Following functions can be implemented with function buttons; • Display a map with multiple map layers, such as roads, streams and boundaries. • Pan, zoom, queries throughout the map • Draw descriptive lines. • Identify and find the features on the map by pointing at them, • Search for features • Change layer properties • Select features with some criterions. • Query database and find the location on the map by ASP • Print the map with required display (Image.6)
Image.6: Print preview display on the web browser 6. Conclusions The developed system provides a spatial data infrastructure environment for the combination of multiple land information datasets. Users can reach conventional GIS, query, analysis functions. With the helping of this system, decision making and future planning can be possible if the data accurately and up to date. Internet GIS provides easy access to the Internet for the users with low requirements and extends the SDSS capabilities to a very large number of users in corporate decision making process. This method can also incorporate up-to-date and real-time information and make overlapping inter-disciplinary information possible But, internet GIS except for desktop or workstation GIS do not provide adequate analysis tools to solve spatial problems for SDSS. Multiple criteria decision making techniques can allow the users to select for a satisfactory solution. Therefore, telecommunication structure of today is not suitable for advanced analysis with the helping of SDSS. If technique problems are overcame in the near future, it seems likely that mobile tools, mobile e-services, and wireless Internet protocols will mark the next major set of developments in SDSS. As a result of this, greater collaboration functions will be enabled and lead to ubiquitous access to information and decision support tools. References [1]. Shim, J.P., Warkentin, M., 2002. Courtney, J.F., Power, D. J., Sharda R, Carlsson C., Past. present and future of decision support technology, Decision Support Systems 33, pp.111 –126.
GEOMATICA 5 "Cartography, Telematics And Navigation"- Barcelona, Spain, February 11th-14th, 2003
[2]. Nolan, R.L., Croson, D.C., 1995. Creative Destruction, Harvard Business School Press, Boston, MA, [3]. Gorry, G.A., Morton, S., 1971. A framework for management information systems, Sloan Management Review 13 (1), pp.50– 70. [4]. Courtney, J. F. Paradice. D.B., 1993. Studies in managerial problem formulation systems, Decision Support Systems 9, pp.413 –423. [5]. O’Brien, J.A., 1990. Management Information Systems: a managerial end user perspective, Richard D. Irwin, Inc., USA [6]. Yomralioglu, T., 2000. Cografi Bilgi Sistemleri: Temel Kavramlar ve Uygulamalar, Istanbul, Turkey [7]. Reis, S., Inan, H.I., Yomralioglu, T., 2002. Designing A Regional Geographic Database And Its Application, International Symposium on GIS, Istanbul, Turkey [8]. Aydinoglu, A.Ç., Yomralioglu, T., 2002. Web based Campus Information System, International Symposium on GIS, Istanbul, Turkey [9]. Foote, K.E., Anthony, P.K., 1998. WebGIS, NCGIA Core Curriculum in GIScience, Available: http://www.ncgia.ucsb.edu/giscc/units/u133/u133.html, USA [10]. Environmental Systems Research Institute, Inc., 2002. ArcIMS 4 Architecture and Functionality, White Paper, California, USA [11]. Garagon, A., Selcuk, T., Ozener, H., Yılmaz, O., Toz, G., Gurkan, O., 2002. Interactive Earthquake Information on the Internet, International Symposium on GIS, Istanbul, Turkey [12]. Keenan, P.B., 1998. Spatial decision support systems for vehicle routing, Decision Support Systems 22, pp.65 –71. [13]. Muller, J.C., 1993. Latest developments in GIS/LIS, Int. J. Geogr. Info. Syst. 7 (4), pp.293-303 [14]. Seffino, L.A., Medeiros, C.B., Rocha, J.V.R., Yi, B., 1999. WOODS- a spatial decision support system based on workflows, Decision Support Systems 27, pp.105-123. [15]. Tsou,M., 2001. Internet Mapping and Distributed GIServices, GEO596 Lecture Notes, Available: http://map.sddu.eu/geo596, San Diego State University, USA
GEOMATICA 5 "Cartography, Telematics And Navigation"- Barcelona, Spain, February 11th-14th, 2003
