This paper reviews the rate constant matrix contraction (RCMC) method, a kinetics-based graph clustering approach developed for the network of paths of chemical reaction elementary steps. First, fundamentals of the transition state theory and the concept and algorithm of the RCMC method are introduced. Next, the RCMC method is explained for kinetic simulation and kinetics-based navigation of automated reaction path search. The RCMC-based kinetic simulation enables quick evaluation of population of chemical species involved at a specific time without explicit time-evolution of the rate equations. The kinetics-based navigation combined with the single component artificial force induced reaction (SC-AFIR) method dramatically reduces the cost of SC-AFIR searches by restricting the search areas to those that are kinetically accessible from the initial structure under a given reaction temperature and timescale. These approaches are presented together with some example applications.