Using qualitative reasoning to evaluate performance: An application in the retail sector

This paper offers a new method inspired by classic importance-performance analysis (IPA) that provides a global index of importance versus performance for firms together with a new version of the IPA diagram. The index compares two rankings of the same set of features regarding importance and performance, taking into account under-performing features. The marginal contribution of each feature to the proposed global index defines a set of iso-curves that represents an improvement in the IPA diagram. The defined index, together with the new version of the diagram, enables, by means of qualitative reasoning techniques, the assessment of a firm’s overall performance and therefore enhance decision making in the allocation of resources. The proposed method has been applied to a Taiwanese multi-format retailer and managerial perceptions of performance and importance are compared to assess the firm’s overall performance. Introduction In Importance-performance analysis (IPA), firm features are ranked regarding either their importance and their performance. Differences between importance and performance rankings of features are considered when assessing a firm’s resource allocation. Initial approaches in the late 70s were based on simple and intuitive graphic techniques (Martilla and James 1977). The traditional IPA methodology basically consists of representing ratings of importance and performance for several features on a two-dimensional chart. The resulting importance-performance grid is divided into four quadrants. To interpret the results, Martilla and James give a name to each quadrant to help managers determine the highest and lowest priorities for improvement, as shown in Figure 1 (Martilla and James 1977). We present in this paper a new similarity index to compare the importance and performance rankings of the same set of features. The proposed index is based on induced ordered weighted averaging (IOWA) operators (Yager and Filev 1999), and importance and performance rankings are obtained by means of qualitative assessments of the different features considered in a firm evaluation. These assessments, given by a set of experts, are expressed using orderof-magnitude models, allowing the experts to use different Copyright c © 2016, Association for the Advancement of Artificial Intelligence (www.aaai.org). All rights reserved. levels of precision for each feature. Differences between the importance and performance ordered lists are considered to define the index of similarity. This index, when applied to a firm’s features rankings for both importance and performance, enables a firm’s global performance to be assessed. There are two main differences between our index and existing indexes such as Kendall’s Tau and Spearman’s Rho correlation coefficients. On the one hand, the asymmetry of the features treatment, i.e., it just takes into account underperforming features, and, on the second hand, the specific relation between the weights and the importance, i.e., the more important an under-performing feature, the greater its weight is considered in the similarity index.