In this paper, "finite point method" (FPM) is presented for modeling 2D shallow water flow problem. The method is based on the use of a weighted least-square approximation procedure, incorporating QR factorization and an iterative adjustment of local approximation parameters. The stabilization of the convective term in this present work is derived from the approximate Riemann solver proposed by Roe. The present method is shown to produce competitive accuracy in the comparisons with the analytical solutions and the well-known Galerkin characteristic-based split (CBS) algorithm.