This paper is divided in two parts. The first part presents optimized methods for formalizing and analyzing sensory data provided by different panels on fabric hand evaluation. For achieving this challenge, we first transform each evaluation into fuzzy sets and we submit criteria for assessing the fabric hand evaluation given by experts as well as criteria for computing the distances between different panels and the employed linguistic terms in different evaluation spaces. This sensory data analysis allows us to check expert's and customer's behaviors on fabric hand.Based on this first procedure about formalization and sensory data analysis, we can study the relationships between sensory evaluation given by different experts for a fabric set and the objective data set provided by appropriate measurements. From a set of fabric samples, a database with nearly 10 parameters characterizing the touch handle is built. This way of evaluation is performed by measuring a set of physical parameters on fabrics. Those parameters constitute the input variables of our model. For the textile industry, the major difficulty lies in the fact that the performed measurements on fabrics lead to precise numerical data describing indirectly fabric hand but their relationships with the evaluation given by experts should be exploited. We implement a fuzzy model which predicts the marks reached for a set of linguistic terms.The effectiveness of these methods and criteria is shown through a number of knitted cotton fabrics.