Pavement Layer Data Repository Using a Spatiotemporal Block Model

Abstract

Historical and in-service pavement layer data along with pavement maintenance, rehabilitation, and reconstruction (MR&R) history is critical information for developing a reliable and accurate pavement performance model that is indispensably necessary for the support of a pavement management system. The conventional homogeneous segmentation method for pavement layer data repository has the problem of excessive segmentation, which may result in inefficient data storage and incomplete data. To address this issue, a spatiotemporal block model is proposed to eliminate the need of homogeneous segmentation, and thus avoid the excessive segmentation and the creation of tiny pavement segments. Three time factors are defined for each pavement layer block to maintain the temporal information of a pavement layer structure. A depth-first-search (DFS)—based cluster searching algorithm is used for reconstruction and visualization of pavement structures as well as for data manipulation. A case study is presented to demonstrate the application developed using the proposed model and algorithms. The computing time and merits of the proposed methodology are also discussed.