The objective of this study was to model the physical and mechanical properties of 100% cotton slub yarns commonly used in denim and other casual wear. Statistical models were developed using central composite experimental design of the response surface methodology. Yarn’s linear density, slub thickness, slub length and pause length were used as the key input variables while yarn strength, elongation, coefficient of mass variation, imperfections and hairiness were used as response/output variables. It was concluded that yarn strength and elongation increased with increase in linear density and pause length, and decreased with increase in slub thickness and slub length. Yarn mass variation and total imperfections increased with increase in slub thickness and pause length, whereas yarn imperfections and hairiness decreased with increase in slub length. It was further concluded that due to statistically significant square and interaction effects of some of the input variables, only the quadratic model instead of the linear models can adequately represent the relationship between the input and the output variables. These statistical models will be of great importance for the industrial personnel to improve their productivity and reduce sampling.