Road side units (RSUs) are considered to be one of the most important components in vehicular networks to send and receive data from other components inside the networks. Regarding the fact that vehicular network performance and vehicular environment coverage are highly related, the maximum coverage via RSUs improves the efficiency of vehicular networks. However, because of high costs of RSU placement, it would be impossible to have full coverage through RSUs. Hence, it would be necessary to optimally install a limited number of RSUs in the most appropriate locations. In this article, the problem of RSU placement is formulated to binary integer programming. The objective of the present study is to find locations which have higher impacts on efficiency of vehicular networks. The objective function is minimising the costs of RSU placement. Moreover, coverage of the most important locations by RSUs has been applied in the form of optimization constraints. The performance of the given mechanism has been assessed by real traces. Simulation results indicate that the method is able to find optimal locations of RSUs in highway and urban environments. Furthermore, the results show that by optimal RSU placement inside the environment, the performance of vehicular network is improved.