This paper deals with the performance optimisation of a multistate operating system with hot redundancy using genetic algorithm (GA). A makeup water system, which is one of a unit of a thermal plant, has multistate system comprises of four subsystems working in series and parallel configuration. The failure and repair rates follow exponential distributions. The mathematical formulation is carried out using probabilistic approach and the Markov birth-death process is used to develop the Chapman-Kolmogorov difference differential equations. The expression for steady state availability has been derived after solving differential equations recursively. GA technique is used to compute optimum values of availability using various feasible failure/repair rates of each subsystem of makeup water system. The optimum availability value and steady state availability value obtained through Markov analysis is also compared in paper. So, the findings of the present paper will be highly useful to the plant management for enhancing system performance.