A pavement management system (PMS) consists of a set of defined procedures for collecting, analyzing, maintaining, and reporting pavement data to provide an aid tool in the decision-making process. Modeling of pavement performance is essential to pavement management. A transition probability matrix (TPM) can be used to model pavement deterioration with time. In the present investigation, the derivation of transition matrices was based on historic data of the International Roughness Index (IRI). Data used on this investigation was collected on a specific program called Corredores de Mantenimiento integral implemented by the Colombian National Highway Institute (INVIAS) from 2005 to 2009. Results indicate that the IRI was sensitive to short-term effects; therefore, it was possible to establish a model to accurately predict its condition for the available data.