Bertram Schoner

Interpretation of criteria weights in multicriteria decision making

Multicriteria decision making models are characterized by the need to evaluate a finite set of alternatives with respect to multiple criteria. The criteria weights in different aggregation rules have different interpretations and implications which have been misunderstood and neglected by many decision makers and researchers. By analyzing the aggregation rules, identifying partial values, specifying explicit measurement units and explicating direct statements of pairwise comparisons of preferences, we identify several plausible interpretations of criteria weights and their appropriate roles in different multicriteria decision making models. The underlying issues of scale validity, commensurability, criteria importance and rank consistency are examined.

A UNIFIED APPROACH TO AHP WITH LINKING PINS

Abstract Recent work demonstrates the need for paired comparison estimates of criteria in the Analytic Hierarchy Process (AHP) to follow prescribed rules, and for the subsequent analysis to be contingent on these rules. Many different approaches are valid, each reflecting a different rule for criteria comparison. A generalized formula from which valid approaches may be generated is presented, and the connection between these methods and Saatys supermatrix approach is explored.

Magnitude adjustment for AHP benefit/cost ratios

Abstract A feature of the Analytic Hierarchy Process (AHP) that has not been subject to close scrutiny is its use for benefit/cost analysis. If benefit priorities are divided by project costs to yield benefit points per dollar, then the resulting numbers can be used to allocate program budgets. But if benefit priorities and cost priorities are derived from two separate hierarchies, then it is likely that the ratio of benefit priorities to cost priorities produces misleading results. To correct the situation, adjustments must be made to put the numerator and denominator priorities into commensurate terms.

Incorporating negative values into the analytic hierarchy process

This paper describes why and how ratio scaled multiple criteria analysis techniques, such as the Analytic Hierarchy Process (AHP), should allow for a subtraction mechanism whereby negative preference (aversion) is combined with positive preference. The main contribution of the paper is a demonstration of how the current imposition of a strictly positive additive value structure can lead to incorrect preference ratios and even incorrect ranking of alternatives. The proposed preference elicitation and computation method solves these problems in a simple and intuitive manner. Scope and purpose: We focus in this paper on common cases where the AHP is used to evaluate alternatives in light of multiple criteria and where at least some of the considerations involve negative rather than positive effects from the point of view of the decision maker. The main objective is to correct how such negative effects are evaluated within the AHP.

Starting rules for incomplete comparisons in the analytic hierarchy process

Entering pairwise comparisons to the Analytic Hierarchy Process (AHP) can be very tedious if the hierarchy is large. Accordingly, techniques exist to derive priorities from incomplete comparisons. This study investigates the effect of different reference items for the first (n - 1) pairwise comparisons of the Incomplete Analytic Hierarchy Process. It analyzes the accuracy of AHP users in estimating the proportions of five colors in a rectangular picture. The empirical results show that significantly greater initial accuracy is achieved if the items are ranked and the lowest ranked item is used as a common referent for the first (n - 1) comparisons.

Alternative Scales in AHP

The nine point scale within AHP has been criticised on various grounds. These include the fact that the scale is bounded, that semantic anchoring phrases are employed without validation, and that users are not informed that the numerical values generated are actually assumed to be ratio scaled. Nevertheless, applications of the scale to the measurement of weight, light intensity and distances have generated accurate estimates. Furthermore, it appears to be the case that decision makers do not experience difficulty in using the scale.

A Field Experiment Comparing Anchored and Unanchored Criteria Weights in the Analytic Hierarchy Process

FacultyofBusinessAdministration,SimonFraserUniversity,Burnaby,B.C.V5A1S6,CanadaABSTRACTThis study reports the results of a field experiment comparing the predictive validity of two approaches tomulticriteria assessments: the absolute measurement mode of AHP and the absolute measurement mode of linkingpin AHP. The questioning procedures of the two differ only in that the former employs unanchored criteria weightassessments and the latter anchored criteria weight assessments.The decision task required insurance agents to respond to a mailed questionnaire in which they evaluated non-monetary incentives (contests) according to (1) the public recognition received from winning, (2) the criteria forwinning and (3) the nature of the reward. There were four levels for each dimension. A between subjects design wasused, with each subject receiving one of the two methods. In addition, all subjects divided 100 points among fourcontests and these hold-out assignments were employed as a validity check. Linking pin AHP was found to besuperior to conventional AHP in this experiment, lending weight to the argument that the use of unanchored criteriaweights is problematic. & 1997 by John Wiley & Sons, Ltd.