Decision Making and Decision Support
The process of decision making implies the selection of the best course of action(s) in order to achieve a set of pre-defined objectives within certain constraints. Such a choice of action(s) is made by a decision maker on the basis of logical analysis of facts coupled with his knowledge of the decision making environment/context as well as his experience and intuition. Process involves repeated consideration of feasible alternatives with regard to action, their evaluation, comparison and ultimate selection of the best action. Decision making process is thus iterative, integrative and participative.
It is iterative because a set of alternative actions are generated which the decision maker evaluates and insights gained are input to, and used to define, further analysis.
It is interactive because the value judgments, which materially affect the final outcome, are made by decision maker who have expert knowledge that must be integrated with the quantitative data in the models (analysis logic).
Decision making issues in spatial context could be more complex than usual. The process formulating the analysis logic and objectives (decision rules) in spatial context can become so ill structured that it may not be possible to define or model them adequately. A decision support system which is aimed at helping the decision makers in solving complex problems, should therefore assist a decision maker by providing problem solving environment rather than presenting a solution at one go. In a wider sense a decision support system could be termed as any organizational set up consisting of personnel and other resources which work together in order to provide inputs to high levels using which the top brass makes decisions. However, in a restricted sense, we are concerned with a computerized mechanism which tries to emulate human expertise and supports the decision makers specific on aspects
Framework for Spatial Decision Support System (SDSS)
Spatial Decision Support System SDSS) could be viewed as an integrated computer S/W and H/W package realized, in spatial context, using Geographic Information System (GIS) functionality on data base management, analytical modeling and spatio-tabular display together with a frame work for adopting the decision makers expert knowledge. More specifically it has certain characteristics (listed below) which differentiates it from any other S/W system.
- It is designed to solve ill-structured or semi-structured problems i.e. where objectives and/or analysis logic cannot be fully defined or modeled.
- It provides an interface, which is powerful and easy to use.
- It enables the user (decision-maker) to combine models and data in a flexible manner.
- It helps the user (decision-maker) to explore the solution space (the various options) by using the models in the system to generate a series of feasible solutions/alternatives.
- It supports a variety of decision-making styles and can be adapted to provide new capabilities as the needs of user (decision-maker) evolve.
- It provides an interactive and recursive problem solving environment (process) in which user (decision maker) proceeds by multiple passes, making use of his own experience, knowledge and intuition.
GIS could provide only a mechanism for data integration, management, analysis and output generation in spatial environment. To start with it can be used for making queries on the underlying database in a series of what lies here or where lies this mode. A logical query has to be translated into a series of what lies here and/or what lies this questions. In such a situation a user is expected to articulate his logical query in a series of razor sharp questions. In return he receives descriptive answers which are fuzzy as far decision support is concerned. Fuzzy because the task of making inferences, deductions and conclusions is left entirely with the decision-makers. An example Query system and query, For pre-dominantly trial regions in Santrampur Taluka which are the village with no drinking water source. Although such a mechanism provides an unlimited scope for seeking the information about the context (feature space) in various permutation and combinations, it cannot be qualified as SDSS because:
- It provides only descriptive answers.
- It does not have any provision for adaptation of decision-maker's knowledge.
- Objectives and analysis logic remains in the mind of user (decision-maker).
- Analysis logic may or may not always be articulatable in form of a series of what lies and where lies queries
An ideal Decision Support System facilitates answers to What if questions
based on pre-defined objectives and a set of decision rules. It is prescriptive rather than just a descriptive Query System
Components of SDSS
There are five key components in a SDSS, which differ quite considerably from a GIS. These are:
- Data base Management
- Analysis logic (objectives, constraints, decision rules)
- Spatial display
- Report generation
- User's interface.
Whereas a GIS offers the above functionality as loosely coupled set of tools or primitives, in SDSS all these functionalities appear to be a seamless entity.
Data Base Management GIS supports spatial query and display of results in as they exist in the data base form. SDSS supports a mix of query, analytical modeling and display of results geared towards the decision making issues, integrating a variety of data sets. i.e. locational, topologic and thematic. SDSS also permits a user to construct and exploit complex spatial relations between all three types of data.
Analytical Procedures Whereas a GIS offers generic analytical functions as applied to wider spatial applications, SDSS focuses on problem specific in-built analysis models which may be realized using a combination of GIS tools. Analysis models could incorporate objectives, constraints and decision rules. These may in turn be expressed in the form of mathematical equations, expressions (arithmetic/logical) and/or if then-else logic.
Graphic and tabular reports SDSS facilitates generation of customized graphic and tabular outputs, which suit a particular decision making context, more importantly, during the interactive, interactive and participative session. A GIS on the other hand provides generic tools for generating these reports.
User Interface User interface of an SDSS has to comply with the requirement of interactive, iterative and participative involvement of a decision-maker. A decision-maker, who could be an expert in his sphere of activity, need not always be a GIS expert. The system should provide an interface which
- Is easy to use in order to be effective.
- Should communication systems capability Via ICONS.
- Should facilitate selection of parameters, data output etc. easily and without forcing a user to refer documents.
- Should be transparent to facilitate visualization of process represented within the analysis model.