Software effort estimation depends on the standard of features that are considered. Considering irrelevant features will effect the model by degrading its performance. Most Research studies focus on Single Objective optimization for feature selection. A Multi-Objective Binary Rao Algorithm for Feature Selection is proposed in this work to estimate software development effort. We have validated the performance of proposed approach using four bench mark datasets and compared with single objective Binary Rao algorithm and NSGA-II algorithm. It is observed that our approach is able to provide better performance than the compared models.