Abstarct: If we know the distances between municipality and offices within a city, we can draw the locations on a 2D map using multi-dimensional scaling Approximately distance matrix can approach to observation matrix. This approach has been traditionally used in a host of different applications. Not only can it be used for finding the location of cities but also for reducing variants dimensions. This approach works baxsed on the observed data, a distance matrix is formed from which the designated configuration can be achieved in a lower dimension. Multidimensional scaling include several methods from analyzing multi variable data that deals with observable Euclidian space. This method provides data on distances between variables that could be either metric or nonmetric. Such distances should not necessarily be Euclidian. Bayesian multidimensional scaling is one of the most recent of multidimensional scaling that at this thesis the method which retrieves the configuration in a lower dimension from a Bayesian perspective. In this thesis examinedis Bayesian multidimensional scaling method with prior distribution normal and mixed, and in this regard, a simulation study compares the superiority proposed method over classical method.