It has always been an urgent issue for the oil and gas industry to improve oil, gas, and condensate recovery at liquid and gaseous hydrocarbon fields developed with the use of artificial formation pressure maintenance techniques that involve injection of water or water combined with other displacement agents. Therefore, due to the aforesaid issues, permanent attention should still be paid to the practical problem of optimizing the non-stationary hydrodynamic pressure applied to a reservoir by regulating the operating conditions of the production and injection wells, development process optimization in general, and water flooding in particular. The theory of Buckley and Leverett, does not take into account the loss of stability of the displacement front, which provokes a stepwise change and the triple value of water saturation. Traditionally a mathematically simplified approach was proposed-a repeatedly differentiable approximation to eliminate the “jump” in water saturation. Such a simplified solution led to negative consequences well-known from the water flooding practice, recognized by experts as “viscous instability of the displacement front” and “fractal geometry of displacement front”. The core of the issue is an attempt to predict the beginning of the stability loss of the front of oil displacement by water and to prevent its negative consequences on the water flooding process under difficult conditions of interaction of hydro-thermodynamics, capillary, molecular, inertial, and gravitational forces. In this study, catastrophe theory methods applied for the analysis of nonlinear polynomial dynamical systems are used as a novel approach. Namely, a mathematical growth model is developed and an inverse problem is formulated so that the initial coefficients of the system of differential equations for a two-phase flow can be deter mined using this model. A unified control parameter has been selected, which enables one to propose and validate a discriminant criterion for oil and water growth models for monitoring and optimizing.