The paper deals with the development of an empirical model to compute the return flow due to the movement of navigation traffics in restricted waterway when the speed of the navigation traffic is under sub-critical range. The velocity profile of the return flow is modelled, where the functional dependence with respect to ambient flow, principal dimensions of the vessel, wetted cross-sectional area of the river, width and average depth of the river, and the lateral distance of the vessel from the bank are implicitly considered. This model has been validated with the observed data collected from the four specific sites, such as Kampsville, Apple River Island, Goose Island and Clarks Ferry of Illinois River during barge-tow movement. The present model shows significant improvement of return flow prediction in comparison to earlier empirical results. This model has also the ability to determine the locus of zero velocity point. The paper deals with the development of an empirical model to compute the return flow due to the movement of navigation traffics in restricted waterway when the speed of the navigation traffic is sub-critical range. The velocity profile of the return flow is modelled, where the functional dependence with respect to ambient flow, principal dimensions of the vessel, wetted cross-sectional area of ​​the river, width and average depth of the river, and the lateral distance of the vessel from the bank are implicitly considered. This model has been validated with the observed data collected from the four specific sites, such as Kampsville, Apple River Island, Goose Island and Clarks Ferry of Illinois River during barge-tow movement. The present model shows significant improvement of return flow prediction in comparison to earlier empirical results This model has also the ability to determine the locus of zero velocity point.