Product recycling, which aims to maximize lifecycle value creation and to conserve environment, becomes important. However, much of product recycling has been done by third party remanufacturers, which is complicated by the need to deal with the various models manufactured by various manufacturers. Furthermore, the working method can be varied depending on the recycling option that comes from lifecycle history, even in the same model. This characteristics impose a heavy burden on the worker, which results in problems of inefficiency caused by human errors. To deal with this problem, an augmented reality based approach for worker support on the complex remanufacturing shop floor, which considers efficiency of information delivery, is proposed in order to minimize human errors.