Network selection is very important for a successful handover in a multi-tier heterogeneous networks. However, the primary challenges currently faced by research community is the lack of availability of network information at the mobile node side for efficiently select the most appropriate target network. It is practically difficult for an UE to get network information from base stations/access point of the neighbouring networks before connecting to them. In response to this, this paper proposes a network selection method that applies the knowledge of mobility data and the network load information to carry out an efficient handover for vehicle-to-infrastructure communication over multi-tier heterogeneous networks. We first derive key parameters, such as relative direction index, proximity index, residence time index, and network load index to select the best candidate network. A moving vehicle would be able to select the most appropriate target network by selecting one or more of the above parameters. We then test our algorithms by developing a dual mode vehicle On-Board Unit equipped with both Long Term Evolution-Advanced (LTE-A) and Wi-Fi network interface cards in OPNET simulator. The performance of the proposed handover method is evaluated by extensive OPNET-based simulation experiments. In the simulation model, we consider a multi-tier heterogeneous network comprising of a macro and multiple small cells of LTE-A and IEEE 802.11n technologies. Results show that our proposed handover method offers about 50% higher throughput and up to 43% higher packet delivery ratio than the conventional received signal strengths based network selection method.