This paper presents a novel discrete-time linear parameter-varying model for gas exchange dynamics during cycling. This model is intended to be used for enhancing the control design process of electrical assistance in bicycles. The mathematical model relates the oxygen consumption and carbon dioxide production to the developed power at the pedal level. The proposed model is also able to predict the excess of CO 2 while high intensity workout is performed. This has been possible by using a time-varying parameter which is scheduled according to the difference between the masses per unit time of O 2 and CO 2. An illustration of the proposed methodology for identification and validation of the model is also presented in this paper. The model parameters for a given individual, were obtained by using measured data of cycle-ergometer tests from two different cycling scenarios.