Effect of Model Accuracy on Maintenance and Rehabilitation Benefits

Abstract

To achieve maximum benefit and cost-effectiveness, the application of pavement maintenance and rehabilitation treatments should be performed within optimal timing widows. In pavement management systems, these timing windows are primarily determined using pavement deterioration models and decision trees embedded in the system. However, owing to the prediction errors of the pavement performance models, the obtained time windows might not be optimal, making it difficult to achieve the expected maximum benefit and cost-effectiveness. To date, there is very little research quantifying the effect of model errors on the difference between the obtained benefit and the expected maximum benefit. Considering this, in this study, a reliability-based approach was developed to analyze the effect of model accuracy on the treatment benefit and extra cost. Through a case study, the effect of model accuracy on the treatment benefit was quantified. An increase in the model error was found to decrease the treatment benefit, while the extra cost needed to achieve the same benefit as the reference scenario increased. Although these findings might seem intuitive, this study has provided an analysis method to quantify such disbenefits.