Abstract: To establish a reliable in vitro cigarette smoke exposure assessment model to study the toxicological effects of smoke exposure on human lung epithelial cells, a human lung epithelial microfluidic chip was designed and fabricated to simulate the airway-liquid interface, and smoke exposure analysis was performed based on the microfluidic chip. Cell viability, apoptosis and tight junction, calcium influx and gene expression were analyzed using immunofluorescence staining, flow cytometry and fluorescence quantitative PCR. To identify further biomarkers related to smoke stimulation, transcriptomic analysis using R language was performed on smoke stimulation-related RNA sequencing datasets (GSE186359 and GSE92662) and lung single-cell sequencing data from smokers and non-smokers (GSE131391), and experimental validation was also performed. The results showed that: 1) Within 72 h after short-term smoke exposure, human lung epithelial BEAS-2B cells grew normally on microfluidic chips without apparent cell apoptosis or tight junction damage. 2) Smoke exposure significantly increased calcium ion influx and upregulated the expression of the capsaicin receptor TRPV1/4, but did not induce the expression of pro-inflammatory factors. 3) The smoke of three types of cigarettes with similar tar and nicotine deliveries but different sensory scores could all induce the expression of aldoketo reductase AKR1C2, and the expression level was positively correlated with the smoke stimulation score. The aldoketo reductase AKR1C2 gene may be a biomarker related to the level of smoke irritation. The developed smoke exposure assessment method based on the lung epithelial microfluidic chip provides a new research tool for evaluating the quality and toxicity of cigarettes.