Time-of-day division is an important step in the formulation of time-of-use electricity prices. The minimum and maximum continuous duration in each period have an important impact on the adjustment effect of time-of-use electricity prices. However, at present, previous studies on the time-of-use electricity price division have not taken into account the time preference of the policymaker. The different needs of policynakers for the total duration of the session, the minimum continuous duration of the session, and the maximum continuous duration within the session are considered in this paper. On the basis of cluster analysis, a time-of-use electricity price time division model considering the preference of policymakers is constructed. The proposed model calculates the distance between load values through affiliation function and Euclidean distance, with the objective function of minimizing the total similarity within each cluster. In this paper, the typical daily load data of a region in China is used to verify the rationality and effectiveness of the proposed model. The results show that the proposed method can obtain the time division results that meet the preferences of decision makers. It is found that the clustering analysis of time period using 0-1 integer linear programming is an effective method to solve the problem of policymakers' preference for duration. Under the traditional time-of-use electricity price time division model, the modified load is inconsistent with the time period electricity price. However, the time-of-use electricity price time division model considering the time preference of decision-makers has a significant corrective effect on this phenomenon.