Many studies on land-use transition have been carried out using a Markov chain model. Its transition probability matrix on land-use is usually estimated from current or historical data in the study area. In this process it is assumed that every place in the area is changeable following to the Markov chain process. However, there are some places whose land-use is in steady state. In this paper we propose a method for estimating the transition probability considering the existence of such steady places. Also effectiveness of this method is examined by numerical examples.
