Cyber physical systems, as a backbone of the Fourth Industrial Revolution (IR 4.0), and all their enabling technologies introduced a relatively newer concept named the digital twin. The interest in the digital twin technology has been growing in academia and industry. This is evident in the increasing number of published research and patents concerning digital twin development along with their various industrial applications. However, there is no framework that could be used to evaluate current digital twin development techniques available in the literature and, Using the proposed framework, identify corresponding strengths and shortfalls. In this paper, prominent approaches related to the development of digital twin were analyzed. Accordingly, a framework was built to compare between the digital twin approaches in terms of research domains, technologies, and models employed in the digital twin, and validation methods used. Using the proposed framework, gaps and future directions for digital twin research are identified from five aspects: digital twin definitions, applications, integration, modeling, and data.