Abstract: Aiming at multi-phases information aggregation question in timing ordering dynamic comprehensive evaluation, a self-mutual incentive aggregation method reflecting the development trend is proposed. Firstly, different prediction points are obtained according to different development trends of the evaluated objects, andthe prediction points are connected to get the prediction line, then the self-incentive of the evaluated object is carried out on the state based on the prediction line. Secondly, the difficulty of growth is measured by the size of the original evaluation value and combined with growthrate to carry out mutual incentive on the growth level. Finally, an example is given to verify the effectiveness of the method. The aggregation method considers the development trend and relative growth level of the evaluated objects from the incentive perspective.Accordingly, differentiated incentives are applied to objects with different starting points, fully respecting individual development differences.

Key words: dynamic comprehensiveevaluation, information aggregation, development trend, relative growth level

摘要： 针对时序动态综合评价中多阶段信息集结问题,提出一种体现发展趋势的自-互激励型集结方法。首先,根据被评价对象不同的发展趋势得出不同的预测点,连接预测点得到预测线,以预测线为基准对被评价对象进行状态上的自激励;其次,依据原始评价值的大小衡量增长难度,并与增长速度结合,进行增长水平上的互激励;最后,通过算例验证方法的有效性。该集结方法从激励的视角考虑被评价对象自身的发展趋势和相对增长水平,据此对不同起点的对象实行差别化激励,充分尊重了个体发展差异。

关键词: 动态综合评价, 信息集结, 发展趋势, 相对增长水平