Management situation is influenced by unexpected events and random shocks,new requirements are proposed on the effectiveness of time-series dynamic comprehensive evaluation method.The complexity and variability of the external environment makes the fluctuation of a class of time-series dynamic comprehensive evaluation value significantly different from the previous state,such as the evaluation value has plummeted and then slowly climbed to a lower level and continued to fluctuate.At this time,the evaluation value of the evaluated object has penetrated the negative incentive line of the classical dual incentive control line and continued to be in a downward state,which makes the effectiveness of the classical incentive evaluation method to be improved.In the face of this situation,the recursive incentive evaluation method is proposed with contextual adjustment characteristics.For the split SERU production system employee characteristics,through the production state of SERU multi-capable workers through the Markov process of prediction,the next production state of the probability distribution is firstly predicted for the multi-capable workers.Based on the probability distribution,the recursive value is then obtained at each stage,the recursive value of the multi-capable workers is incentivized with the combination of the dual-incentive line method.And finally combined with the hesitant fuzzy function,the time weights are solved with the degree of subordination,and the evaluation value is obtained with time weights with the degree of subordination.By considering the combination of SERU production system,which is a new production system of the era,it is proved that this method has better flexibility and robustness.