In view of the uncertainty of remanufacturing processing time and multi-objective parallelism that exists in the job-shop scheduling problems. A multi-objective scheduling model of remanufacturing job-shop considering completion time, cost, load balancing and energy consumption is established based on triangular fuzzy operation time of work-piece. Besides, a multi-population hybrid artificial fish swarm algorithm is applied to solve the proposed model. In this algorithm, a multi-population co-evolution strategy is used to improve the efficiency of the algorithm. And then, the Pareto solution set is updated with individual dispersion as the evaluation criterion. Finally, one case of the remanufacturing job-shop scheduling problem is presented to illustrate the feasibility of the proposed method.