In recent years, interdisciplinary methods integrating deterministic and probabilistic approaches have been gaining popularity due to their effectiveness in decision making for the design and operation of socio-technical systems. This paper demonstrates the value of combining the Bayesian Belief Networks (BBN) and System Dynamics (SD) for socio-technical predictive modeling. BBN is a technique for depicting probabilistic relations among elements of the model, where objective data are lacking and use of expert opinion is necessary. This is beneficial for the quantification of socio-technical models, dealing with the soft nature of human and organizational factors, however, BBN is inadequate for capturing dynamic aspects including feedback loops and delays. Combining SD with BBN can compensate for these BBN deficiencies. As an application, SD-BBN methodology is integrated with classical Probabilistic Risk Analysis (PRA) techniques in order to enable the Socio-Technical Risk Analysis framework to capture dynamic interactions of causal factors within their ranges of uncertainty.