Polycyclic aromatic hydrocarbons (PAHs) are a class of organic compounds that are composed of aromatic rings containing only carbon and hydrogen atoms. They are one of the widespread environmental pollutants in the world. In recent years, many scholars have focused on the inhibition, formation mechanism, content of active components, and biodegradation effect of polycyclic aromatic hydrocarbons. They summarized the research progress of pretreatment methods for detection, but rarely discussed the experimental dataset for comprehensive analysis of pollution sources and the impact of different pretreatment technologies on the extraction of different substrates. What is more, computer simulation has not been mentioned. In this study, the pollution sources of polycyclic aromatic hydrocarbons (PAHs) are reviewed, and the related applications of various pretreatment methods such as gel permeation chromatography (GPC) are summarized. Finally, the computer simulation of the response surface method is introduced. The concentration of polycyclic aromatic hydrocarbons is tested or predicted by combining the neural network with the alternating trilinear decomposition (ATLD) algorithm, artificial population algorithm (ABC), and hierarchical genetic algorithm (HGA). Its future development trend is discussed and prospected, which provides a reference for solving the pollution problem. We look forward to providing help for the follow-up research of scholars in this field.