A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Evaluation of Flexible Pavement Performance Models in Mississippi: A Neural Network Approach
One of the main responsibilities of state transportation departments is the construction of new roadways and the maintenance of existing highways. The main concern for transportation agencies is prioritizing the decisions for maintenance and rehabilitation actions. Mississippi Department of Transportation (MDOT) currently utilizes Markov probability matrices, which provides the probability of a pavement section that needs to be rehabilitated. Flexible pavements are affected by various factors such as design parameters, traffic load, climate, and environmental factors. The probability matrices utilized are not based on these factors, which significantly affect the rate of deterioration of the roads and highways. Critical maintenance scheduling is a needed action to limit further problems as the deterioration rate increases over time. To properly assess the condition of the pavement and operate maintenance, prediction models with significant design and condition variables are essential. The rate of deterioration of the pavement and its condition need to be predicted, so that the effectiveness and timing of the maintenance can be estimated. In this study, an artificial neural networks (ANNs) approach was used to develop pavement performance prediction models. The effect of rehabilitation actions were included as part of the inputs. The number of outputs were determined after various trials. The two output model for predicting Performance Condition Rating (PCR) and International Roughness Index (IRI) have found to be the most promising. ANN model successfully characterized the behavior even though the statistical measures are not within a suitable range. Rehabilitation actions were efficiently incorporated in the model and were found to be accurate.
Evaluation of Flexible Pavement Performance Models in Mississippi: A Neural Network Approach
One of the main responsibilities of state transportation departments is the construction of new roadways and the maintenance of existing highways. The main concern for transportation agencies is prioritizing the decisions for maintenance and rehabilitation actions. Mississippi Department of Transportation (MDOT) currently utilizes Markov probability matrices, which provides the probability of a pavement section that needs to be rehabilitated. Flexible pavements are affected by various factors such as design parameters, traffic load, climate, and environmental factors. The probability matrices utilized are not based on these factors, which significantly affect the rate of deterioration of the roads and highways. Critical maintenance scheduling is a needed action to limit further problems as the deterioration rate increases over time. To properly assess the condition of the pavement and operate maintenance, prediction models with significant design and condition variables are essential. The rate of deterioration of the pavement and its condition need to be predicted, so that the effectiveness and timing of the maintenance can be estimated. In this study, an artificial neural networks (ANNs) approach was used to develop pavement performance prediction models. The effect of rehabilitation actions were included as part of the inputs. The number of outputs were determined after various trials. The two output model for predicting Performance Condition Rating (PCR) and International Roughness Index (IRI) have found to be the most promising. ANN model successfully characterized the behavior even though the statistical measures are not within a suitable range. Rehabilitation actions were efficiently incorporated in the model and were found to be accurate.
Evaluation of Flexible Pavement Performance Models in Mississippi: A Neural Network Approach
Lecture Notes in Civil Engineering
Tutumluer, Erol (editor) / Nazarian, Soheil (editor) / Al-Qadi, Imad (editor) / Qamhia, Issam I.A. (editor) / Duckworth, Patrick (author) / Yasarer, Hakan (author) / Najjar, Yacoub (author)
2021-08-31
12 pages
Article/Chapter (Book)
Electronic Resource
English
Nondestructive Flexible Pavement Evaluation Using ILLI-PAVE Based Artificial Neural Network Models
| British Library Conference Proceedings | 2006
| Taylor & Francis Verlag | 2018