Due to its harmful effect on both human health and environment, soot emission is considered as one of the most important diesel engine pollutants. In the last decades, the industrial engine manufacturers have been able to strongly reduce its engine-out value by many different techniques, in order to respect the stricter emission norms. Simulation modeling has played and continues to play a key role for this purpose in the engine control system development. In this context, this paper proposes a new soot emission model for a direct injection diesel engine.This soot model is based on a zero-dimensional semi-physical approach coupled with a crank-angle resolved combustion model and a thermodynamic calculation of the burned gas products temperature. Furthermore, a multi linear regression model has been used to estimate the soot emissions as function of significant physical combustion parameters. The proposed correlation consists of a nominal value of soot and a correction factor that takes in consideration the formation and oxidation processes. Both development and validation phases were achieved on Renault 1.6 liter Euro 5 diesel engine over a large range of operating points. The comparison of simulation results with experiments shows very good correlation coefficients relative to literature soot models. Such model has the advantage of being easily adaptable to different diesel engines and able to be directly integrated into control strategies and industrial applications like MVEM models. The model was performed as part of a simulator project which aims to simulate diesel engine performance and emissions using 0D/1D approaches.