This paper recalls the main principles of the method of analysis of dam abutments proposed by Pierre Londe [6] in 1973, based on the limit equilibrium of rock wedges determined by preferential surfaces of failure such as faults or joints. The initial method was mainly based on graphical representation of forces (weight of rock, thrust applied by the dam, and resultants of uplift due to water pressure on weak planes) and, therefore, allows parametric analysis of stability as a function of the most uncertain ones, which are uplift forces. It has been observed that these principles, widely referred to all over the world, are actually largely deviated, first by referring to cohesion which, according to Londe, should not be considered safe and, second, by using misleading safety factors. On the other hand, progress in office tools now allows engineers to easily perform parametric studies with all types of graphical presentations of results as a function of uncertain variables. This paper proposes the adequate methodology for comprehensive implementation of the initial method on any type of software, brings to light some common mistakes, and proposes the full set of equations needed.
