This study develops a Bayesian method for the analysis of multiple structural changes in a long memory process. Our method is based on the irreversible Markov-switching ARFIMA model. We apply this method to the estimation of the number and the points of structural changes in daily realized volatility, which is the sum of the squared intraday returns over a day, of Nikkei 225 stock index. We also analyze how the estimate of the difference parameter may change by taking into account the structural changes to examine whether the long memory property is spurious and caused by structural changes.