Use of Crack Characteristics in Crack Sealing Performance Modeling and Network-Level Project Selection

Crack sealing (CS) and crack filling (CF) are commonly used crack treatment methods. However, the study of their performance is still very limited, making it difficult for highway agencies to systematically and optimally select network-level CS-CF projects within available budgets. To address this issue, a generalized performance model for CS-CF–treated pavements is proposed. Detailed crack characteristics—including crack type, density, and width—are employed in the model. In the proposed performance model, crack density related to three types of cracks (transverse cracks, nonwheelpath longitudinal cracks, and wheelpath longitudinal cracks) is used to determine performance gain. Two discount functions are incorporated to consider the negative impact caused by alligator cracks and cracks that are very tight or very wide. The proposed model is instantiated and estimated using the practices of the Georgia Department of Transportation on CS and CF and the department’s pavement distress survey protocol. The case study—which uses three-dimensional laser data collected from a 1-mi pavement section on State Route 26 (US-80) near Savannah, Georgia, from 2011 to 2016—validates the feasibility and reasonableness of the model. An integer programming method is formulated for network-level CS-CF project selection. The testing results of 53 pavement segments show that the model and programming method can be used to select CS-CF projects within budget constraints while maximizing the length-weighted average performance gain. The proposed performance model and integer programming method show promise for use in incorporating CS and CF into a highway agency’s pavement management system. Conclusions and recommendations are offered.