Probabilistic load flow has gain attractive attention in electric power system planning, operation, and control. Probabilistic load flow is an efficient tool to access the performance of the power system considering the uncertainties of the load demand and generation. However, future operating condition and operating criteria could not be predicted without considering uncertainty. Due to integration of renewable energy sources into power system network brought uncertainty and dependence factor. Most of the uncertainties in power system are load increment, generation scheduling, and network topology in which engineers are dealing with them. In this paper, Box-Muller sampling algorithm is proposed to solve the probabilistic load flow problems considering uncertainty with generation and correlated loads. The main advantage of the proposed approach is that accurate solution can be obtain with less computation time and can address the dependence between variables. Also, it is almost unconstrained for the probability distributions of the input random variables. The proposed method is compared with correlated simple random sampling Monte Carlo simulation. For the demonstration purpose, the proposed method is investigated using IEEE 14-bus and IEEE 118-bus test systems. The simulation results indicate that Box-Muller sampling method is a promising approach in probabilistic load flow evaluation.