The short-term hydro-thermal scheduling involves scheduling of hydro and thermal generations to minimise the fuel cost of thermal plants in a horizon of planning time while meeting various hydraulic and electric system constraints. The problem is considered as a non-linear and non-convex problem involving valve-points effects and transmission losses with a set of equality and inequality constraints. In this study, a novel promising approach is represented with an innovated mutation method utilising genetic algorithm (GA). The proposed real coded GA approach with random transfer vectors-based mutation (RCGA-RTVM) is applied to two test systems with different characteristics. The effectiveness of the proposed RCGA-RTVM is validated by the obtained results. Comparison of the obtained results with recently developed methods shows the superiority of the proposed algorithm.