Although transportation has long been recognised as one of the critical factors for socio-economic development, it is responsible for many global greenhouse gas emissions and significant pollutants that cause severe health problem, especially in urban areas. Besides, the rapid development of transportation also brings great concern about energy security. To reduce the transportation system's side effects, Electric vehicles (EVs) have emerged as a promising solution toward sustainable transportation due to their positive impact on environmental issues and energy crisis. However, the adoption of EVs is still minimal compared to conventional gasoline vehicles due to the lack of appropriate charging infrastructure. Furthermore, the electrification of transportation may cost a significant amount of money and result in more congestion (i.e., en-route and charging congestion) due to EVs' routing and charging behaviours. Having these concerns in mind, in this study, we focus on answering the questions of where and how to deploy the charging facilities to promote the widespread adoption of EVs and improve the system performance in the presence of EVs. The objective is to minimise the investment cost of charging infrastructure and the en-route and charging congestion by capturing travellers' routeing choice behaviours with stochastic demand and driving range. The problem is first formulated as a bi-level optimisation problem to capture the mutual interaction between planning decision and traffic flow pattern on the network and then solved by a meta-heuristic. Finally, the proposed framework is tested through the numerical test, and some managerial insights into the facility planning and system performance also be provided.