Abstract: The dynamic development of things increases the demand of dynamic evaluation methods. The static evaluation method solves problems of “utility measurement” and “ranking” that can be compared between the evaluated objects (decision making units).On the basis of the static evaluation, the dynamic evaluation method needs to solve the problem of comparing the evaluated objects’ “utility measurement” and “ranking” at their different periods. It requires the evaluation model and data standardization method possessing more long-term utility.At present, there are two widely used ways of dynamic evaluation: In the first one, according to the index information of each period, the traditional static evaluation is carried out to obtain the evaluated objects’ evaluation value, which is later utilized and processed to obtain a comprehensive evaluation value of various periods on multiple indexes of period.The second one integrates the index value of the evaluated object through the time synthesis on the index system, and then used the static evaluation method to get the final dynamic evaluation value. Both the dynamic evaluation ways need to determine the weight for the evaluation values of each time period.These two mentioned standardization methods, whether respectively standardized according to each time period, or standardized the data of several time periods, share the same problem of dynamic evaluation: the former one, which maps the data of each period to a unified interval, covers the overall differences of evaluated objects in each period, which may lose the compatibility and objectivity of the evaluation results of different period. The latter one, which can support the dynamic comparison between various periods, is sort of a “static” process that runs through these periods. Further, its evaluation result is only valid within the unified standardization of the various periods and is difficult to adapt to a new period. Therefore, although there are several methods of dynamic evaluation have been presented at present, there is still a lack of research on the analysis and construction of dynamic evaluation methods from the viewpoint of the effectiveness of dynamic evaluation. In this paper, a principle of building a long-term effective dynamic evaluation mechanism is put forward:based on judgment and processing of the long-term effectiveness of various evaluation techniques, this principle is composed of the classification of index types, the data standardization method of “datum period regression” which lacks the long-term effectiveness, and the selection of long-term effective evaluation methods, and integration of the overall evaluation mechanism.This paper is organized as following: Initially, the common evaluation data is classified into three types (natural data, social data related to money and artificial data obtained through expert consulting methods) and a long-term processing method for these data is put forward. Furthermore, a data standardization method called “datum period regression” is presented. Additionally, the Jingyou evaluation method based on linear weighted model is adopted to construct a dynamic evaluation mechanism of objective and democratic. Finally, a dynamic evaluation is conducted on technological innovation capability for four municipalities in China, by which the practical effectiveness of the result of this paper is proved. The evaluation method proposed by this paperenriches the comprehensive evaluation method and can be studied and applied more widely.

Key words: dynamic evaluation; effectiveness; evaluation mechanism; long-term effective integration principle