Mathematical programming framework for modeling and comparing network-level pavement maintenance strategies

This study proposes a mathematical programming framework to model and quantitatively compare different maintenance strategies for network-level highway pavements. The study develops mixed-integer linear programming models for various maintenance strategies that are commonly adopted in practice and in the literature. In developing these models, traffic, pavement age, and maintenance actions with heterogeneous effects are considered. The strategies include optimization-based, worst-first, best-first, and threshold based strategies. To demonstrate the flexibility of the framework and present a practical situation in which engineering judgment is sometimes incorporated in pavement maintenance strategies, we further develop a mixed strategy. A solution procedure combining the off-shelf mixed-integer programming solver, greedy algorithms, and Lagrangian relaxation algorithms is developed to efficiently solve the models. Finally, a numerical example of a hypothetical network is established. Different maintenance strategies are applied given different budget levels, traffic loadings, and initial pavement conditions. The results of the numerical example are reasonable, and they provide insights into the efficient implementation of maintenance strategies. Results also show that the framework has the potential to aid maintenance agencies in evaluating maintenance strategies before they are implemented, improving pavement conditions, and reducing the budget for transportation infrastructure. (C) 2018 Elsevier Ltd. All rights reserved.