With the diversified development of user demand, demand response can facilitate unit maintenance scheduling. This paper proposes a multi-objective power generation maintenance scheduling model that considers demand response, takes the objectives of environment, economy and system reliability into account, and uses a lexicographic algorithm to solve the problem. First, a four-stage solution process and the lexicographic algorithm are proposed, with the explanations of the incentive index and the system reserve level indicator. A corresponding model for demand side management and demand response is built. Then this paper establishes a multi-target unit maintenance scheduling optimization model. It takes into account the impact of regional emission values and emission levels in terms of environmental objectives, and accounts for the next phase of constraints. It optimizes the minimum operating, maintenance and backup costs, and considers the maximization of reliability indicators with demand side management. Finally, the model is simulated in a 24-node system, and the effectiveness and applicability of the model are illustrated. This work is supported by National Natural Science Foundation of China (No. 51877190) “Optimization Operation and Risk Scheduling Method of Electrical Integrated Energy System”.