The development of a progressive modern highway system is essential for the enhancement of the road capacity, safety, efficiency, commerce, and national defences of a country. These highway systems consist of various integrated individual asset components that undergo constant deterioration during their usage. The difficulty with maintaining these infrastructure systems is, the various asset components have a different service life and erode with a different rate during the lifespan of the system. This research study proposes the use of a multi-objective predictive maintenance optimization system using a non-dominated sorting-based multi-objective evolutionary algorithm (MOEA), for the optimum upkeep of a highway infrastructure project. The model has been applied on a pavement system in this study, but the framework can be effectively applied on other multi-asset infrastructure systems as well. The algorithm aims to find a spread of Pareto-optimal solutions by concurrently optimizing two objectives consisting of minimizing the life cycle cost (LCC), and maximizing the level of service (LOS) throughout the life-cycle. A case study was developed to compare the effectiveness of the model, based on the maintenance data from the asset management plan of the California department of transportation published in October 2017. It is acknowledged in this research study that the two objectives have a conflicting nature of various degrees and thus the research suggests a set of solutions for different ranges rather than a single value solution. The approach proposed in this research study will also analyze the role of a robust multi-objective optimization (MOO) system for highway maintenance through application to a deteriorating highway project. The results from the study will be helpful in developing promising techniques for the application of various multi-objective optimization (MOO) systems and thus pave the way for efficient decision-making tools for the maintenance of highway infrastructure projects.