The lighting system as a crucial urban infrastructure needs to be maintained to uphold good performance. The planning and scheduling of maintenance for geographically distributed lighting systems is a complex research problem as well as a challenge in practice due to resource constraints and the complexity of route planning. This study proposes a multiple-objective decision-support model (MODSM) for lighting maintenance routing, which considers both the carbon emissions generated during the maintenance process and the loss of efficiency due to workload imbalance among maintenance crews. In the proposed model, carbon emissions are converted into a carbon tax index as part of the total cost index (along with the pure cost index), while the workload balance is measured by the standard deviation of workload indexes among the crews. To address the identified problem, an improved genetic algorithm is used to identify the optimal solution for group relamping. A numerical example is provided to demonstrate the suitability of the developed model to the selection of an optimal maintenance scheme, including the number of maintenance crews and their corresponding maintenance routes. Finally, the performance of the proposed model is assessed by sensitivity analysis and comparison with the traditional method. Integrating the interests of multiple stakeholders and environmental protection, the MODSM for lighting maintenance routing developed in this study can provide a helpful tool for researchers and practitioners in the construction engineering and management community to formulate a more comprehensive decision into maintenance management.