Prediction of Road Pavement Conditions Using the Markov Chain Method in Regency Road Programming

Abstract

The use of budget in road management must be effective and efficient by providing the optimal pavement performance values. The problem in cost optimization of the road handling programs in West Bangka Regency is the lack of information regarding with the changes in conditions that will occur in the future due to the pattern of road handling that carried out, so that required a method that capable to predict road pavement conditions. The purpose of this study is to determine the cost of future road handling programs based on the value of road pavement conditions from the prediction results using the Markov chain method. Modeling requires an initial condition vector and a Transition Probability Matrix (TPM). The main data used in the development of this model is pavement condition data and road handling history data for 2012 - 2017. The application of the Markov chain model on the road network in West Bangka Regency in the period of 2018-2022 shows a drastic decrease in pavement conditions, if not conducted the handling action on the road damage, with a change in the value of Good (B) condition  from 63,7% in 2017 become 12,3% in 2022. Based on a simulation of a road management program during this period, produce an estimated cost of Rp. 45.338.471.000 by providing a change in the value of Good (B) conditions at the end of 2022 by 58,6%. The results of the study are expected could assist road managers in the context of the preparation of an optimal road management program.