Flotation is one of the feasible separation methods suggested for recovery of petroleum coke from the tailings of lime calcination furnaces. In this study, analyses of ash content and calorific value of petroleum coke in lime calcination tailings were used to measure its floatability and product quality. In addition, seven most common flotation kinetics models were fitted to the obtained experimental data. Based on the maximum recovery, minimum ash content, and maximum calorific value of the flotation products, optimum dosages for collector (kerosene) and frother (MIBC) were found 30 g/t and 60 g/t, respectively. Regarding the flotation kinetic modeling and the obtained sum of squared errors (SSEs), Agar and Klimpell models were found to have the best and the poorest fits to the experimental data, respectively. Finally, it was concluded that new statistical concepts such as information criteria (IC) and non-linear generalized least squares estimation (NLGLSE) must be applied to the process of model selection owing to consideration of goodness of fit, complexity of a model and model consistency.