Edge computing has become a prominent computing strategy as mobile devices and Internet of Things (IoT) became popular in the last decade where cloud computing proved partly insufficient meeting the computational requirements of these devices/applications. Unlike cloud, edge computing can provide low latency in communication, high quality of service, and support for high mobility. Connected and autonomous vehicles scenarios can be considered as an important application field for edge computing as these are the key requirements to implement a vehicular network. In this paper, we aim to present a remedy to one of the crucial problems in vehicular networks: efficient RSU placement by addressing network coverage and computational demand. We propose an RSU placement framework for generating placement models based on traffic characteristics of a target area. Our work differs from previous studies in that we focus on both communication coverage and the computational demand aspects simultaneously. The proposed framework in this study can be used by infrastructure providers for designing an efficient RSU placement while building a smart city. Moreover, our work includes extending capabilities of a simulation framework designed for edge computing scenarios. To demonstrate the effectiveness of our proposal we evaluated the performance of various placement models in realistic settings.