A novel computational framework that enables reactive flow simulation without using an a priori reaction mechanism is proposed in this work. The proposed computational framework is based on the integration of automated mechanism generation and on-the-fly reduction based on element flux analysis. Stepwise implementation of the integrated framework is developed to perform reactive flow simulations. The computational framework starts the simulation without any prior knowledge of the mechanism with only fuel and air as the initial species and performs the stepwise mechanism generation and on-the-fly reduction iteratively to obtain the entire simulation results. In each local computation step, a mechanism is generated automatically starting with the reduced mechanism at the end of the previous step, and flux-based on-the-fly reduction is applied to the generated mechanism to obtain reactive flow simulation results for the current step. The reduced mechanism at the end of the current step is then used as the starting point for the next mechanism generation step. Case studies of methane oxidation simulation in plug-flow reactor (PFR) model are carried out to demonstrate the proposed computational framework. The simulation results show satisfactory accuracy of the proposed framework compared to the traditional framework of reactive flow simulation. The proposed computational framework provides a novel methodology for conducting reactive flow simulations. By using the proposed computational framework, we are able to start reactive flow simulation without the need for a readily developed reaction mechanism beforehand. The application of on-the-fly reduction also provides the potential to reduce the computational costs during the simulation.