Classify M CDM problems according to the meaning of evaluation criteria,the number of decision makers involved in the decision-making process,and the amount of information available to decision makers.2. Discuss the major differences between conventional M CD M and spatialMCDM.3. Define the concepts of M CDM , M ODM , and MADM.4. Define the major elements of spatial multicriteria analysis. Identify the roleof GIS and M CDM techniques in each stage of the analysis.5. Compare the decision matrix, geographical data matrix, and spatial interaction matrix (see Figures 1.1 and 3.1). Identify the similarities and differences.Why is it useful to organize the input data for multicriteria decision analysisin a matrix format?C h a p t e r4Evaluation CriteriaSpatial multicriteria decision making requires an articulation o f the decisionmaker's objectives and an identification o f attributes useful for indicatingthe degree to which these objectives are achieved. An attribute is used tomeasure performance in relation to an objective. The objective and underlyingattributes form a hierarchical structure o f evaluation criteria for a particulardecision problem. In this chapter we establish guidelines that facilitate theselection o f evaluation criteria (objectives and attributes). Both individual criterion and a set o f criteria should possess properties to adequately represent themulticriteria nature o f a decision problem. Each criterion must be comprehensive and measurable. A set o f criteria should be complete, operational, decomposable, nonredundant, and minimal. Once the hierarchical structure o f objectives and attributes is established, each criterion should be represented as amap layer in the GIS database. The set o f criterion maps is a representationo f a particular decision situation or a particular segment o f the real-worldgeographical system. Given the variety o f scales on which a criterion can bemeasured, multicriteria decision analysis requires that the values contained inthe various criterion map layers be transformable to comparable units. Anumber o f approaches can be used to make the criterion map layers comparable.The approaches are categorized into deterministic, probabilistic, and fuzzymethods.4 1 S E L E C T I N G E V A L U A T I O N C R I T E R I A4 .1 .1 Hierarchical Structure of Evaluation CriteriaAs suggested earlier, evaluation criterion is a generic term including both theconcepts of objectives and attributes (see Section 3.2). An objective is a state1 0 3

troduction to Multicriteria Decision AnalysisI n this chapter we focus on multicriteria decision making (MCDM). TheI terms multicriteria decision analysis and multicriteria decision analysis(MCDA) are used interchangeably. Broadly speaking, M CDM problems involve a set o f alternatives that are evaluated on the basis o f conflicting andincommensurate criteria. Criterion is considered a generic term that includesboth the concepts o f attribute and objective. Accordingly, two broad classeso f M CDM can be distinguished: MADM (multiattribute decision making) andM O D M (multiobjective decision making). Both MADM and M O D M problemsare further categorized into single-decision-maker problems and group decisionproblems. These two categories are, in turn, subdivided into deterministic,probabilistic, and fuzzy decisions. Deterministic decision problems assume thatthe required data and information are known with certainty and that there is aknown deterministic relationship between every decision and the correspondingdecision consequence. Probabilistic analysis deals with a decision situationunder uncertainty about the state o f problem’s environment and about therelationships between the decision and its consequences. Whereas probabilisticanalysis treats uncertainty as randomness, it is also appropriate to considerinherent imprecision ofinformation involved in decision making; fuzzy decisionanalysis deals with this type o f uncertainty. Conventional M CDM techniqueshave largely been aspatial in the sense that they assume a spatial homogeneitywithin the study area. This assumption is unrealistic in many decision situationsbecause the evaluation criteria vary across space. Consequently, there is a needfor an explicit representation o f the geographical dimension in MCDM. Thesecond part o f this chapter provides a framework for GIS-based (or spatial)multicriteria decision analysis. The framework integrates the GIS capabilitieso f data acquisition, storage, retrieval, manipulation, and analysis and the capabilities o f M CD M techniques for aggregating the geographical data and thedecision maker’s preferences into unidimensional values o f alternative decisions.8 1

appropriate diagram to explain a municipal service provider the importance of geospatial statistics in decision making

Which technique is used to identify the land managementa.Ecological Information Systemsb.Geological Information Systemsc.Historical Information Systemsd.Geographic Information Systems

Which technique is used to identify the land managementa.Geological Information Systemsb.Ecological Information Systemsc.Historical Information Systemsd.Geographic Information Systems

Which technique is used to identify the land managementa.Ecological Information Systemsb.Geographic Information Systemsc.Geological Information Systemsd.Historical Information Systems