Partial periodic patterns are a key form of regularity observed in temporal databases. Most previous studies tried to find these patterns in binary databases by ignoring the critical information about the quantitative values of the items. This paper leverages the concept of fuzzy sets to propose a novel model of fuzzy partial periodic frequent patterns (FPPFPs) that may occur in a quantitative temporal database. An efficient depth-first search algorithm is also introduced to discover all FPPFPs. Our algorithm introduces an innovative pruning method that streamlines the search space and lowers the computational cost of identifying the desired patterns. Experimental results validate our algorithm’s effectiveness. Finally, a case study on air pollution data is presented to illustrate the effectiveness of the proposed patterns.