Digital twin (DT) is commonly known as a virtual model of a physical object, a process or a system. In order to create DT, the emphasis is on building a Cyber-Physical System (CPS), where information collected from physical processes is used for computation to generate a real-time monitoring cyber model. At the same time, changes of the cyber model will either be used for prediction or reflection in the physical system to achieve the control function. With the development of digital and information communication technologies, DT has been applied in various areas such as manufacturing and aerospace industries. However, studies on the application of DT in the building construction sector were limited. Such studies were particularly rare about volumetric modular building which has attracted a burgeoning interest in both academia and practice globally. This paper aims to contribute a better understanding of integrating the use of DT for Modular Integrated Construction (MiC) by examining the opportunities and challenges. The paper first reviews the state-of-the-art DT applications across various industries, and then focuses on publications about DT applications in the building construction sector through conducting a systematic search of literature using Scopus. After review and analysis of the identified publications, the paper explores the opportunities and challenges of applying DT in MiC. The review results show that DT is often considered with other digital technologies for integration. A conceptual framework is developed for applying DT in module installation in MiC within the context of Hong Kong as a typical case of high-density cities. Promising opportunities with huge benefits are speculated from the application of DT in MiC, including enhanced coordination of logistics and construction management by using DT to monitor on-site progress during the module installation stage. Nevertheless, challenges are also identified, which exist in not only the sensing technology and cyber system, but also the social-political environment supporting innovation and regulation. The developed framework should provide a useful guide to address the challenges and shape future research on DT in MiC.