Biography: Dr. B. K. Mohanty is a Professor in the Decision Science Group at Indian Institute of Management, Lucknow, India. He received his M.Sc in Mathematics, from Berhampur University, Orissa, India and PhD in Operations Research, from IIT Kharagpur India. Prior to joining IIM Lucknow, he served at Tata Research Development and Design Centre, Pune, as a member of Technical Staff, as an Associate Professor at Xavier Institute of Management, Bhubaneswar, India and as Scientist at CFTRI Mysore, India. His research interests include Fuzzy Applications in MCDM, e-commerce and various functional areas of Management. To his credit, he has about 28 research publications in the journals of International repute. To name a few these include, Int. J. Systems Science, Fuzzy Sets and Systems, Expert systems with Applications, Computers and Industrial Engineering, Computers and Operations Research etc.
Speech Title: Product Preferences with Multiple Attributes- A Fuzzy Set Theoretic Approach
Abstract: In both online and traditional markets, a customer makes choices on the products based on the products' multiple attributes. Often these attributes are conflicting, non-commensurable and fuzzy in nature. The concepts of fuzzy sets help us to measure the satisfactions; a consumer has on these attributes. In real life problems, a buyer always chooses a product having maximum satisfaction on its attributes. However, this is rarely possible as the higher satisfaction in one attribute lowers the satisfaction in another attribute. For instance given a product CAR, a price sensitive buyer always looks for a cheapest car and scans over the available cars. That is given a product; our methodology makes a consumer to move from one product to another in an increasing satisfaction (Increased membership value) direction. No doubt, this movement in the path of products will improve the level of satisfaction in one attribute, but it may impede the satisfaction of other attributes. The maximum level of impediment is taken here as the cost to the consumer.
By taking other products, we can have the paths as above. The paths over all the products give us a network of products with nodes as products and arcs as the pointers from one product to another. The shortest distance (least cost) between the products, when a product moves to another, help us to obtain a group of bundles of similar products. The smaller the cost of moving from a product to another, is equivalent to the larger the similarity degree amongst the products. Note that the similarities amongst the products are obtained here with respect to a particular attribute. The degree of similarity is measured using fuzzy membership values amongst the products. Similarly we have the similarity degrees of the products with respect to the other attributes. The degrees of similarity of a product across the attributes are combined and the maximal similarity corresponding to a product is selected as a final product.