The purpose of this paper is to introduce a grey-box model of three-way catalytic converter, which is capable of estimating the oxygen storage level to aid the fuel-to-air ratio control in spark ignited engines. As it is well-known, the prime parameter that drives the transient dynamics in current three-way catalytic converter is their capability to store a certain amount of oxygen, then allowing to oxidize some pollutant species such as carbon monoxide or hydrocarbons even at rich conditions during short periods of time. Since oxygen storage level is considered a good indicator of the catalyst state but it cannot be directly measured, a model based real-time capable estimation like the one proposed in this paper could be valuable. The model accounts for oxygen storing as well as oxidation and reduction of the main species involved, taking as inputs fuel-to-air equivalence ratio, air mass flow, temperature and gas composition at three-way catalyst inlet. From these inputs, oxygen storage level and brick temperature are calculated as model states, which finally provide the gas composition downstream of the catalyst as output. In addition, a simplified model of narrowband [Formula: see text] sensor is included, it provides a voltage from gas composition at the outlet of the catalyst and allows to assess the model behaviour by comparison with the on-board [Formula: see text] sensor measurements. Finally, the validation of the model performance by means of experimental test as well as different practical cases, where the benefits of oxygen storage level estimation plays a key role, are introduced.