In the context of carbon trading policy, enterprises need to consider efficiency and environmental factors integrally to maximize their benefits. In this paper, an energy-saving scheduling problem for distributed dual-resource flexible job shop considering machines and workers is investigated, and a mixed-integer planning model is developed, with the objectives of minimizing the makespan, energy consumption and carbon trading cost. Aiming at the multi-resource constraints of this problem, an improved sparrow search algorithm is designed in this paper, which embeds an active decoding strategy based on the three-layer encoding of "factory-operation-machine&worker" to improve the resource utilization rate. To expand search space, 6 local search strategies are introduced into the algorithm. The improved sparrow search algorithm is compared with the other three algorithms through 30 sets of instances in 2, 3 and 4 factories. The results show that the improved sparrow search algorithm is superior to the other comparison algorithms, and the comparison results verify the effectiveness of this improved algorithm.