A. R. Dasgupta
Distinguished Professor BISAG and
Honorary Advisor, GIS Development
[email protected]
Geospatial information is increasingly becoming a part of the information needs for the daily operations of an enterprise. The enterprise could be a business like a department store, a utility like a power company or a local government entity like a municipality or even a government department like roads and buildings.
In each case the importance of ‘where’ becomes enhanced as enterprises try to maximize their profits or effectiveness in an increasingly competitive and demanding environment. In this milieu GIS plays an important role by ingesting, storing, processing and analyzing geospatial information relevant to the operation of the enterprise. GIS utilization began as a standalone operation in the IT or MIS department of an enterprise but, as the technology matured it migrated to the desktop, and now to hand-held devices. GIS operations have thus become more decentralized, democratized and ubiquitous.
This process has major benefits as it puts the information right in the hands of an end user but it also results in problems like fractured databases, lack of synchronization, data duplication, loss of data and ultimately loss of efficiency and accuracy. Unorganized growth also leads to a proliferation of systems with attendant interoperability and compatibility problems. An Enterprise GIS is the solution to such problems. Such a system is designed to provide an integrated and interoperable environment in which the individual departments and functionaries of an enterprise can create, access, view, and analyze data and information relevant to their tasks. This information could encompass spatial as well as nonspatial data sets. Applications could range from complex spatial models to delivery of services encompassing government, business and citizens.
The first gain from an enterprise GIS is the reduction in data redundancy and the standardization of data acquisition and content. Data needs to be acquired by the focal agencies as per a predefined format agreed upon in advance by the data using agencies. Such data has an associated accuracy, integrity and reliability assured by the focal agency. By reducing redundancy in data collection, costs are saved in terms of collection, timely data availability and repeated reuse of the collected data. Standardization results in better data management and enhanced modelling capabilities. Standardization also ensures a higher degree of semantic interoperability since all users will use standard terms to describe geospatial features.
Limiting standardized data acquisition to focal agencies also ensures data security particularly where the data is confidential to an agency or an individual like financial records and health records. Standardization also frees users from the bother of data collection. They can concentrate on the analytical tasks and resort to data collection only when such data is not readily available. Such application specific data can also be shared for use by other agencies. This scenario has two underlying implications. Firstly, the availability of standardized data sets must be catalogued and the catalogue must be made available across the enterprise. Secondly, data access has to be regulated depending on its level of confidentiality. Thus street can be open access but property tax records need to have controlled access. These features and controls are easy to implement in an Enterprise GIS.
Enterprise GIS will also enable interoperability across applications. Thus applications like citizen services, road maintenance, health care, market research, etc can use geospatial data seamlessly with other business processes like Enterprise Resources Planning systems, SCADA, e-Governance, etc. To enable this Enterprise GIS needs to be in the core of the organization with links to all services and applications. Enterprise GIS thus becomes subsumed in the IT infrastructure of an enterprise.
Architecturally, an Enterprise GIS is a hybrid of tightly and loosely coupled systems. Individual departments can have tightly coupled systems, software and hardware but across departments, field units and public interfaces the coupling has to be loose to allow independence to the users to choose their own systems. Such an approach will also insulate end users from changes which may be required to be made from time to time in individual departments. Such changes may involve system upgrades or workflow changes and these should not necessitate corresponding changes in end user systems. Similarly users may upgrade their systems without fear of losing access to their data and information sources. This type of coupling also enables easy scalability. New departments can be brought on line without disturbing the system for others. Increase in the number of users can be managed through additional systems being brought in to the network. Distributed GIS required every user to become a Jack of All Trades; data collector, GIS operator as well as applications expert.
The end users of an Enterprise GIS need to be trained in the usage of the system, its rules and regulations and features, in order to enable them to use the system for their needs. A plan for establishing an Enterprise GIS must be financially viable and technically sound. The plan should address five aspects:
- definition and design specifications for enterprise GIS;
- description of internal and external databases being managed;
- plan for conceptual applications and database architecture;
- system architecture, including hardware, software, and applications; and
- implementation plan encompassing tasks, methods,and activities, schedule,funding sources and organizational responsibilities.
Most GIS vendors provide the tools necessary to create an Enterprise GIS. However, open source solutions are also attractive and they provide the opportunity to share and learn from others’ experiences.
The cost of an Enterprise GIS is not just the cost of the hardware and software. The initial planning, data standardization, identification of core applications, data acquisition, applications development and final systems deployment will involve costs up front.
The payoff will come from higher data usage, ease of data access and sharing of data and applications over time.
Hence any commitment to the deployment of an Enterprise GIS should be made with a clear understanding of these issues. Finally, the measure of success of an Enterprise GIS is how well it performs its intended service. This could be in terms of ease of data access, variety of applications, number of users and uses, reliability and stability.
