Comparison of Multi-criteria Decision Making Methods for Selection of Handmade Carpets

Abstract

ABSTRACT Selecting the best carpet from a set of available alternatives for a particular end-use situation is a difficult job because of association with multiple decision criteria. Three multicriteria decision-making (MCDM) methods, i.e., weighted sum model (WSM), weighted product model (WPM), and technique for order preference by similarity to ideal solution (TOPSIS) were used to evaluate the overall durability value of handmade carpets considering abrasion loss, compression recovery, thickness loss after dynamic loading, and thickness loss after prolonged heavy static loading as decision criteria. The weights of carpet properties (decision criteria) were determined by considering opinion of 10 domain experts. Abrasion loss, compression recovery, thickness loss after dynamic loading and thickness loss after prolonged heavy static loading received weights of 0.366, 0.186, 0.308, and 0.140, respectively. All the three MCDM methods yielded the same ranking for the first position. Moreover, WSM and WPM had complete agreement for the first seven positions. Rank correlation coefficient between WSM-WPM, WSM-TOPSIS, and WPM-TOPSIS are found to be 0.990, 0.943, and 0.973, respectively i.e. very high. Therefore, any one of these MCDM methods can be used for handmade carpet selection problem. Sensitivity analysis proved that the ranking produced by TOPSIS is quite robust in response to the change in weights of decision criteria.