In this paper, we propose a multiobjective model of portfolio rebalancing problem considering return, risk and liquidity as key financial criteria. Further, a more realistic situation of financial market is considered where the portfolio, at the end of a typical time period, will be modified by buying and/or selling asset(s) in response to changing conditions. We assume that the transaction costs are paid on the basis of incremental discounts and are adjusted in the net return of the portfolio. A real-coded genetic algorithm (RGGA) is developed to solve the portfolio rebalancing problem and build an optimal portfolio. An empirical study is included to illustrate the behaviour of the proposed model using data of some randomly selected assets listed on the National Stock Exchange (NSE), Mumbai, India.