Abstract: For the multi⁃performance evaluation model in the complex coking production process, the evaluation index is single, and the evaluation index can only reflect the production one⁃sidedly.The process without considering the problem of global factors, use the Analytic Hierarchy Process (AHP) to comprehensively consider safety, stability and economic aspects, and establish a global performance evaluation model. In order to solve the problem that decision makers are more likely to assign uncertainties to multiple indicators, use the intuitionistic fuzzy set (IFS) to obtain the preference relationship including membership degree, non⁃membership degree and hesitation degree. Consider the irrationality of consistency threshold, introduce a new threshold, and use entropy and cross entropy to calculate the index weight, and finally obtain the performance state of the target system. Finally, two different operating states of the coking production process were evaluated to prove the effectiveness and feasibility of the IFAHP method.

Key words: Coking process, Intuitionistic fuzzy analytic hierarchy process, Performance evaluation, Consistency threshold, Entropy and cross entropy

摘要： 针对复杂的焦化生产过程中多性能评价模型存在的评价指标单一、评价指标选取仅能片面反映生产过程，未考虑全局因素的问题，利用层次分析法（AHP）从安全、稳定、经济方面综合考虑，建立全局性能评估模型。为解决决策者对多个指标分配偏好关系易出现不确定评估的问题，采用直觉模糊集（IFS）获取包含隶属度、非隶属度、犹豫度的偏好关系。考虑一致性阈值的不合理性，引入新的阈值，采用熵和交叉熵计算指标权重，最终得出目标系统的性能状态。对焦化生产过程两种不同运行状态的评估结果，证明了IFAHP方法的有效性和可行性。

关键词: 焦化过程, 直觉模糊层次分析法, 性能评估, 一致性阈值, 熵与交叉熵