Objective To analyze the differences in mineral element contents among different pineapple samples and explore the feasibility of applying a discriminant model based on elemental fingerprint analysis to the identification of Shenwan pineapple, a geographical indication product.Methods Inductively coupled plasma mass spectrometry was employed to determine the contents of 28 mineral elements in 5 different pineapple samples. Combined with principal component analysis, partial least squares discriminant analysis, and Fisher discriminant analysis, the classification characteristic elements were identified. Discriminant models for pineapples from different producing areas and different bases within the same producing area were established and evaluated.Results The characteristic mineral elements (Ca, K, Mg, and Mn) for identifying pineapples from three different producing areas, namely Mangchong Village of Shenwan Town, Yugonglou Village of Xuwen Town, and Wuhe Village of Jiangchuan Town, are screened out. For pineapples from three different planting bases (Mangchong, Dapai, and Modao) in Shenwan Town, the characteristic mineral elements identified are Ca, K, Mg, Na, and B. The Fisher discriminant model established based on these characteristic mineral elements achieves an overall discrimination accuracy of 100% for pineapple samples from different producing areas and 95.9% for those from different bases.Conclusion The multivariate statistical analysis method based on mineral elements can effectively discriminate pineapple samples from different producing areas or different bases and can be applied to the traceability of pineapples.