Abstract : Slugging is an undesirable multiphase flow regime occurring on oil production facilities. This manuscript studies the dynamics of this intermittent phenomenon in view of suppressing it by feedback actuation of the outlet valve. We propose control solutions applicable in a broad range of industrial settings. After a quantitative description of the physical characteristics of slugging, we propose a model for two-phase (gas-liquid) flow with distributed parameters reproducing the phenomenon. The model takes the form of a hyperbolic system of transport equations, for which we propose a numerical solving scheme. Besides, we proceed to a stability analysis by constructing a strict Lyapunov function for the mixed initial-boundary value problem. Then, we present a reduced-order model which reproduces the pressure and flow rate oscillations of slugging. After a dynamical analysis of this model, we describe how to calibrate its parameters so that its behavior corresponds to that of a given slugging system. Finally, we propose feedback control laws, based on the analysis of the reduced-order model, in two distinct industrial setups: whether a bottom pressure sensor is available or not. The performances of these solutions are compared with the state-of-the art methods in each situation. The conclusion is that a bottom pressure sensor is not systematically required to stabilize the flow. When one is available, the control law we propose yields better stabilization properties than the solution commonly used in the industry, which should improve the oil recovery process.