Progressive Usage of Business and Spatial Intelligence for Decision Support in the Delivery of Educational Services in Developing Countries

Abstract

Managers and administrators in public sector organisations have the responsibility to make various decisions concerning service delivery to the citizens of the country. The provision of education to the children of the country is one fundamental type of service. Decision makers often work within many constraints, some of which are budgetary, political, geographical, and lack of relevant and timely information to support decision making. A well-educated population is a necessity for the economic development and well-being of a population. The provision of effective educational services is therefore crucial for every economy. The purpose of this chapter is to provide a discussion of knowledge discovery technologies and systems that can be implemented to support the gathering, storage, and analysis of data for purposes of supporting decision-making activities in the public sector in general and the public education sector in particular. The technologies discussed are business intelligence (BI), geographical information systems (GIS), and free/libre/open source software. It is argued in this chapter that the combination of these technologies can and should provide operationally effective and cost-effective solutions to the problem of gathering, storage, and analysis of data to support decision makers’ information needs. A case study of an education department which makes use of BI and GIS technologies in their management activities is provided as an example of effective usage of BI and GIS technologies for decision support.

Appendix

Technical details on databases that support spatial analysis for GIS are provided in this appendix. Figure 5.1 shows the relationship between the GIS, spatial database system, and object-relational database system. As depicted in Fig. 5.1, a GIS can be used as a front-end to a spatial database system, so that the GIS accesses spatial data through the spatial database in order to perform the required spatial analysis. An ORDBMS provides the platform to support the spatial database. When the spatial database uses the ORDBMS as the underlying DBMS engine, spatial data is queried using the SQL language. ORDBMSs provide constructs that have been specified in the SQL:2008 standard (Connolly & Begg, 2010). These constructs include built-in data types such as row types and collection types as well as programmer-defined types which are object-oriented programming (OOP) constructs in the form of classes and class hierarchies, objects, methods, and operators. A row type is an aggregate data type with fields of various atomic types, e.g. numeric, date, and character strings. Collection types include arrays, sets, and lists (Connolly & Begg, 2010). The SQL:2008 constructs described above make it easy to implement the necessary computational data structures for spatial data. As a result, the specification provided by the OGC for incorporating two-dimensional geospatial abstract data types (ADTs) into SQL can be easily implemented. These ADTS are based on the computational object model and include operations for specifying topological and spatial analysis operations.

A typical spatial database system today is an ordinary database with additional capabilities and functions that include spatial data types, spatial operators, spatial indexing, and spatial data management functionality which includes data loading and transaction control. Spatial data types are stored as simple features as defined by the Open Geospatial Consortium (OGC, 2013) or as binary large objects (BLOBs). A simple feature is a graphic representation of a physical spatial feature (e.g. road) or conceptual spatial feature (e.g. service area). OGC simple features are defined inside the spatial database and can therefore be processed by the database. The SQL:2008 standard specifies user-defined types (similar to classes in OOP) which can be used to define data types and operations (methods) on these data types. On the other hand, the processing of BLOBs requires additional software components. Spatial operators are functions for performing spatial joins of tables, performing calculations on table data, retrieving data from tables, etc. These functions are invoked through SQL queries. Spatial indexing differs from transactional database indexing since spatial data is typically represented as coordinate pairs in two-dimensional or three-dimensional space.

In the spatial database and GIS environment, data is commonly grouped according to the function it serves in spatial analysis. These groupings were discussed in Sect. 5.4.2. The relationships between the layers are shown in Fig. 5.4. The base map data layers provide the geodetic (survey) control network and topographic base data. Geodetic (survey) control network provides the spatial reference framework for all database data. Topographic base data provides geographical referencing required for collections, analysis, and the display of data in higher layers (application and business). The framework data layers are three related layers for the geographical referencing of human activity on land. The parcel layer provides the framework for land development and administration applications. The facilities layer forms the basis for facilities management in public utilities and resource management. The address layer supports various land and resource applications that require postal addresses. The application data layers comprise of various spatial datasets collected for different database applications in land and resource management. Application datasets use the base map and framework data layers for geographical referencing. The business solutions layers consist of collections of spatial data layers including framework and application data layers plus related spatial and non-spatial data. The data is assembled to support operations and decision making functions of the business units in an organisation (Yeung & Hall, 2007).