TY - JOUR
T1 - Pavement Maintenance Threshold Detection and Network-Level Rutting Prediction Model Based on Finnish Road Data
AU - Dettenborn, Taavi
AU - Hartikainen, Ari
AU - Korkiala-Tanttu, Leena
PY - 2020/6/1
Y1 - 2020/6/1
N2 - Accurate prediction models for road structure deterioration increase the cost-effectiveness of road construction and the scheduling rehabilitation and maintenance of road structures. In this paper, a method to detect the minimum maintenance operation detection (MMOD) threshold and network-level pavement rutting prediction model are described. The MMOD threshold has the potential to filter network-level pavement rutting measurement data and improve prediction models. The model is a multilevel statistical time series model for rutting prediction without the need for measurement history. The model parameters used are pavement type and average daily traffic. The road maintenance planner estimates the need for a minimum sampling rate for future pavement performance measurements and predicts the pavement rut behavior. For asphalt concrete and soft asphalt concrete, the model gives realistic predictions for the first 10 years. For stone mastic asphalt, the realistic prediction window is the first six years.
AB - Accurate prediction models for road structure deterioration increase the cost-effectiveness of road construction and the scheduling rehabilitation and maintenance of road structures. In this paper, a method to detect the minimum maintenance operation detection (MMOD) threshold and network-level pavement rutting prediction model are described. The MMOD threshold has the potential to filter network-level pavement rutting measurement data and improve prediction models. The model is a multilevel statistical time series model for rutting prediction without the need for measurement history. The model parameters used are pavement type and average daily traffic. The road maintenance planner estimates the need for a minimum sampling rate for future pavement performance measurements and predicts the pavement rut behavior. For asphalt concrete and soft asphalt concrete, the model gives realistic predictions for the first 10 years. For stone mastic asphalt, the realistic prediction window is the first six years.
KW - Multilevel model
KW - Pavement performance model
KW - Rutting prediction
UR - http://www.scopus.com/inward/record.url?scp=85083217222&partnerID=8YFLogxK
U2 - 10.1061/(ASCE)IS.1943-555X.0000539
DO - 10.1061/(ASCE)IS.1943-555X.0000539
M3 - Article
AN - SCOPUS:85083217222
VL - 26
JO - Journal of Infrastructure Systems
JF - Journal of Infrastructure Systems
SN - 1076-0342
IS - 2
M1 - 04020016
ER -