Complexity is the property of a project that makes it difficult to understand, foresee and control its behaviour due to interrelatedness between various project attributes such as tasks, parties, disciplines, and interfaces. Project complexity is an inherent part of mega construction projects due to their large scale, long period, the multiplicity of disciplines, the high number of stakeholders, multi-nationality, and high level of public attention, as well as political interest. This paper depicts findings of a research project that aims to identify characteristics of and factors that contribute to project complexity by referring to the experiences of professionals involved in mega construction projects. For this purpose, data has been collected by interviewing 18 participants from 11 mega construction projects. Communication complexity, which reflects the amount of distributed communication channels between various stakeholders, has been identified as one of the major drivers of project complexity in megaprojects. Based on literature and semi-structured interview findings, a causal structure to represent communication complexity comprising of three components, which are; project, stakeholders, and communication management, was developed. The causal map, which represents the opinions of a limited number of experts, cannot be generalised; however, it can provide useful insights about drivers of communication complexity that can be used to develop strategies to manage complexity.

Risk assessment in projects requires the integration of various information on project characteristics as well as external and internal sources of uncertainty and is based on assumptions about future and project vulnerability. Complexity is a major source of uncertainty that decreases the predictability of project outputs. In this research, the aim was to develop a decision-support tool that can estimate the level of risk and required contingency in a project by assessment of complexity factors as well as contextual information such as contract conditions and mitigation strategies. A process model and a tool were developed using the data of 11 mega construction projects. The tool was tested on a real project, and promising results were obtained about its usability. The tool has the potential to support decision-making during bidding in construction projects with its visualization and prediction features. On the other hand, as a limited number of cases and experts were involved in this study, findings on its performance cannot be generalized. The identified complexity and risk factors, proposed process model, and visual representations may help the development of similar decision-support tools according to different company needs.