Metadata education suggestions and materials for:
|
Decision Making / Decision Support |
Learning Material | Preparatory topics | Complementary topics | Vocabulary
Conviction
Motivation
Skills
Knowledge
Maps have a long history of being used to support decision making for navigators, military planners and explorers. Maps and GIS continue to play an important role in modern-day decision-making including retail planning, policy-making, business management and strategic planning, facilities management, and natural resources management.
Because of the power of GIS to integrate different types of data in a spatial context it is a particularly relevant tool for decision-making, providing a framework for structuring data into information needed to make a decision. For instance, deciding whether a parcel of land should be permitted for a particular use: data on existing roads, developed areas, potential hazards, and environmental factors can be structured in a GIS and evaluated with respect to the location of the parcel in question. Metadata plays an important role in structured decisions because it provides documentation not only for the data sources used to arrive at the decision, but it can also provide documentation for the process used to arrive at the decision. This documentation is important in the event that the decision is ever questioned, either from a legal standpoint or from the standpoint of developing a more accurate, efficient, or appropriate approaches to decision-making.
Metadata is also critical in the process of converting raw data into useable information for decision making. As new, user-friendly desktop software makes GIS and related decision support systems technologies more accessible to users, care must be taken that inappropriate decisions are not made because of lack of information about data and its fitness-for-use for different applications or analyses. Decision makers (not just technicians/analysts) need to be able to determine the data's structure so that they can use it for decision making, as well as to evaluate it for its fitness-for-use. Standardized metadata provides a means for both technical and non-technical users to better understand the uses and limitations of different datasets.
Decision support systems (DSS) are designed to deal with ill-structured or semi-structured problems, where criteria for making decisions are difficult to define. For instance, increased pressure for development may require a re-evaluation of the criteria currently used for permit decision-making. Policy makers need additional information on cumulative impacts from past developments in order to define new criteria. Decision-making that is based on multiple criteria is another form of problem that is difficult to structure, since interactions between different objectives means that treating individual objectives in isolation can be problematic. For instance, managing natural resources for sustainability is a multiple-objective problem, with competing objectives (finding a balance between harvesting resources, and conserving them). Decision support systems can provide the means for generating a series of decision alternatives for comparison and evaluation of different objectives. Metadata to document the data used to base decisions on is still important, but in a more complex decision support environment which may incorporate different models, parameters and alternative solutions, metadata to document the process becomes even more important. Ideally, decision support systems should incorporate some method of logging and tracking actions or choices made by the users to provide this type of process documentation.
Many GIS projects are developed for a one-time planning effort or problem solution, and data is subsequently archived or deleted once the project is complete. On the other hand, decision support systems are designed to be used on a regular or semi-regular basis, to provide support for multiple, often inter-related decisions. Since these systems are used over time, they frequently depend on dynamic, changing data from different sources. Keeping data up-to-date and adding new data relevant to specific problems is therefore an important and problematic issue in keeping a decision support system functional and useful. The Interagency Group on Decision Support for Land, Natural Resources and the Environment (1998) identified a list of important features for an "ideal" decision support system. One of the identified features for a DSS is internet connectivity and ability to use distributed databases and models. This might arguably be the future of all decision support systems that rely on dynamic, changing data from different source, and/or rely on input and evaluation from different sources/offices. Collaborative decision-making is becoming a substantive field within decision-making science as it recognized that different experts and interest groups are often involved in a decision-making process. Distributed databases and models promise more efficient means for creating, maintaining, updating and incorporating data for analysis, but only if there is a well-developed infrastructure of communication between the distributed components. This is another area where metadata plays a critical role. If metadata exists in a structured format that the DSS is able to read, it is much easier for users of the system to incorporate the data and models with assurance that the information or analysis tools are appropriate for use (see the topic on Data sources and determining fitness-for-use for more details).
Another important element of a decision support system includes methods to account for and display error and uncertainty (Interagency Group on Decision Support for Land, Natural Resources and Environment 1998). Metadata provides the most straightforward means to provide information about error and uncertainty, to assist users in evaluating data sets for fitness-of-use. Decision support system have the potential to provide more active incorporation of metadata into the decision-making process. Ideally, metadata should not be treated as a separate entity from the data, but should be incorporated within the data and communicated as requested, or even automatically, as the data is being used. For instance, if two or more datasets of different source-scales are selected by the user to be combined in some manner for analysis, the system can be programmed to first check the metadata and then warn the user of different source-scales before they proceed with the analysis.
Standards and interoperability of decision support systems: see the topic on Interoperability/standards
Metadata and Agents
Metadata enables "agents" to operate in distributed heterogeneous environments: by interpreting, filtering, and converting information automatically, agents help decision makers access information for decisions more easily. Agents help users access distributed data objects and GIS components on heterogeneous GIS platforms across the internet by interpreting, filtering, and converting information automatically. An agent is an autonomous computer program that has specific functions and responds to specific events, based on pre-defined knowledge rules or user designated instructions. Using metadata, an agent can bridge heterogeneous information systems and translate different data types and models for different GIS tasks.
View the full abstract: Tsou, M.H. and B.P. Buttenfield, 2000. Agent-based mechanims for distributing geographic information services on the Internet. GIScience 2000: the First International Conference on Geographic Information Science.
View the full abstract:: Sengupta, R. 2000. A distributed intelligent geographical modeling environment for spatial decision support. Ph.D. thesis, Southern Illinois University at Carbondale. 203 pp.
Interagency Group on Decision Support for Land, Natural Resources and the Environment, 1998. A Strategy for the Development and Application of Decision Support Systems for Natural Resources and the Environment, a report for the Aurora Partnership. http://www.aurorapartnership.org/strategy.html