<p>The biodegradability evaluation of petrochemical wastewater is vital for regulating the petrochemical wastewater treatment process. Nevertheless, the essential datasets derived by instruments with different sampling scales are characterized by multiple time scales, making it challenging for the existing data-driven biodegradability evaluation methods to achieve feasible results. In this paper, an intelligent evaluation method is proposed based on multiple time-scale analyses to ensure real-time and accurate biodegradability evaluation of the petrochemical wastewater treatment process. Firstly, a multiple time-scale reconfiguration method is introduced to regularize the datasets consistently by regulating the time-series characteristics of the collected variables. Moreover, missing data for large time-scale variables are supplemented by linear interpolation. Secondly, a multi-scale feature extraction algorithm based on partial least squares is designed to obtain biodegradability feature variables and remove noise and redundant information. Thirdly, an intelligent evaluation model based on a dynamic fuzzy min-max neural network is established to realize the classification of biodegradability. Finally, the proposed evaluation method is applied to the practical petrochemical wastewater treatment process. The experimental results demonstrate that the proposed method can provide real-time and accurate evaluation of the petrochemical wastewater biodegradability.</p>