Optimizing the batch scheduling under time-sensitive electricity prices is effective to reduce electricity costs in production, and an excellent algorithm is very important to obtain significant economical benefits in the scheduling. In this paper, an bi-objective model is introduced to implement economical hot rolling batch scheduling, and an improved NSGA-II based algorithm is proposed to obtain Pareto optimal solutions. A hybrid chromosome code with mixed natural numbers and real numbers as well as code mapping are customized to guide the search in feasible domain for dealing with multiple complex constraints, a dynamic change strategy is designed to adjust crossover and mutation probability, and an iterative greedy strategy is developed to solve the subproblem of allocating idle time for rolling units. Finally, a comprehensive optimal solution is recommended through multicriteria decision-making. Computational results show that the proposed algorithm is effective to solve the problem and outperforms the corresponding algorithms in performance.