Finding Most Likely Sliding Surfaces Using PSO

PSOslope is a Windows application program with a GUI interface written in C# using the evolutionary algorithm of Particle Swarm Optimization (PSO). It was designed and developed to compute the Factor of Safety (FS) of a soil slope. This paper presents a new version of the program called PSOslope 2, which contains a number of improvements over the previous version, including the support for the Bishop´s Simplified Method (BSM) in addition to the existing Ordinary Method of Slices (OMS). The new program also used more stringent criteria for assessing acceptable sliding surfaces than the old version did. Finally, PSOslope 2 was used to compute the FS of a simple vertical cut in a homogeneous soil. The results showed that PSOslope 2 was able to discover a more likely sliding surface (smooth but non-planar) with a lower value of FS than the theoretical prediction (based on the assumption of planar failure). This demonstrated that the PSO could be used to efficiently optimize the problems involving slope stability that includes natural slopes and excavations. It also showed that the PSO was a promising approach and provided practical solutions for the search of most likely sliding surfaces.