如何让无疲劳的计算机代替易疲劳的用户是交互式遗传算法研究的一个重要内容.本文首先研究了机器代替用户的3个基本内容,即,机器代替用户扮演环境角色,对个体进行评价;机器采样和代替用户的时机是进入偏好稳定阶段;机器寻优结果由样本和代替用户策略决定.其次,给出基于基因意义单元“适应值”估计的个体适应值估计方法.最后,通过比较实验验证了本文方法的有效性. How to make fatigue-free computers replace fatigue-prone users is an important part of the research on interactive genetic algorithms.This paper firstly studies three basic elements of machine replacement users, that is, the machine replaces users and plays an environmental role to evaluate individuals. The time to sample and replace users is to enter the phase of preference stabilization. The result of machine optimization is decided by sample and user’s strategy.Secondly, the method of individual fitness estimation based on gene unit “fitness value” is given.Finally, Experimental results verify the effectiveness of the proposed method.