Soil health evaluation needs to consider regional natural endowment and environmental characteristics, and the selection of evaluation indicators should be determined based on different regional characteristics. To ensure that the selected indicators can represent the key attributes and functions of farmland soils in Yinchuan Plain, in this study, 147 farmland soil samples were collected, 16 soil physical, chemical and biological indicators were determined, principal component analysis (PCA) was used to screen and construct the minimum data set(MDS) for farmland soil health evaluation, and soil health index (SHI) method was adopted to evaluate the health status of farmland soils. The results showed that the MDS for soil health evaluation contained eight indicators: soil water content, water-stable macroaggregates, pH, total salt, total nitrogen, available phosphorus, organic matter and microbial biomass carbon. The linear regression equation fitted between soil health indices derived from the MDS and the total data set (TDS) using the weighted synthesis method showed a good performance, which indicated that the constructed MDS could represent the TDS for soil health evaluation in the study region. The soil health index for farmland in the Yinchuan Plain, calculated based on the MDS, ranges from 0.18 to 0.78 with an average value of 0.52. Spatially, the soil health index exhibits a general trend of higher values in the south and lower values in the north. However, distribution remains uneven in certain areas, indicating significant spatial heterogeneity. The results of this study can provide theoretical support for the evaluation of farmland soil health in the Yinchuan Plain as well as for green agricultural production.