The paper introduces a data-driven framework seeking to realistically model the multi-stakeholder decision-making process for reaching opinion consensus. The proposed process models the sequential steps starting from opinion elicitation and the creation of the decision-makers network using a novel data-driven method based on Bayesian Networks. The realization of consensus is then achieved either through deliberations modeled using Opinion Dynamics, or through the intervention of an arbitrator modeled using the Consensus Reaching Process framework. The common issues of opinion inconsistency and multitude are addressed applying a linear optimization and a fuzzy c-means clustering algorithm, respectively. An application of the methodology on a network-wide traffic management strategy is presented, using opinion and interaction data elicited from multiple decision-makers using a structured questionnaire survey.