Analytical Network process (ANP), is applied here as a decision making technique for the selection of appropriate robots for industrial and automation applications. The core motivation of applying, in particular, the ANP technique is that robot selection is dependent upon a number of attributes and criteria which have strong influences/interdependencies upon each other. The ANP, as a multiple attribute decision making (MADM) technique for robot selection, captures the effects of these cross hierarchical dependencies, and appropriately maps the influences within the clusters and between the various alternatives. Simultaneously, the technique does not include the assumption of independence of higher-level elements from lower level elements and about the independence of the elements within a level. First, a set of attributes, which influence the selection of the robots, are identified. Next, using the various steps of ANP, viz., pair wise comparisons matrices and priority vectors determination and the development of the super-matrix the global weights of the attributes with respect to other attributes are determined. The final alternatives are then rated as per the graduated weights of the respective attributes. Thus, a comprehensive solution towards selection of robots enabling the decision-makers to suitably understand the complex relationships of the relevant qualitative and quantitative attributes in the decision-making is obtained. The technique is also illustrated using detailed analysis for a specific case of decision making between three robot suppliers and selection of appropriate robot from alternatives. In order to get more insight into relationships among various attributes and their effect on decision makers, the sensitivity analysis of the results with respect to determinant level attributes is carried out.
Dynamic Hybrid Multiple Attribute Decision-Making Problem Based on Reference Point Adaptation
