Multiobjective Optimization Model to Coordinate between Segment and Network Level for Managing Pavement and Sustainability

Abstract

Traditionally, maintenance optimization for large-scale pavement networks usually has been ignored. This study developed a new multiobjective segment-linked decision-making model (MOSL-DM) integrating five modules to optimize pavement preventive maintenance plans at the network level: decision tree, decision objective selection and calculation, decision-making unit partition, multiobjective optimization (MOO), and segment-network conversion modules. This MOSL-DM can be used to solve large-scale pavement network maintenance optimization problems while coordinating between the segment level and the network level. The framework of MOSL-DM is introduced, and the operation method and function of each module are described thoroughly. The MOO mathematical formulation in MOSL-DM considering agency costs, environmental costs, and performance improvements is established, and the calculation method of each objective function is presented. An illustrative example of an expressway network with 1,100 pavement segments shows the application of MOSL-DM. In this application, 155 sets of feasible combinations of preventive maintenance plans were obtained for decision makers to choose from, each of which is nondominated by the others, and the conversion function embedded in MOSL-DM that can coordinate between segment-level and network-level was presented. The principal contribution of this study is to provide a tool for highway agencies to optimize sustainably the maintenance plans for large-scale pavement networks.