In order to study feeding traits of breeding pigs and to meet the demand for intelligent development of measuring equipment，a set of information management system based on internet of things and web technology was designed.The system selects CAN bus for bi-directional data transmission，and adopts B/S （Browser/Server） front and back-end separation architecture.The front-end was based on Reactjs technology stack，and the back-end was developed with flask web framework and MySQL database.The results of two measurement tests showed that the data transmission reliability of the system is high，and functional modules meet the expected design requirements and the practical application requirements of performance measurement of breeding pigs.The phenotypic correlation between feeding traits and growth traits was analyzed in Duroc breeding pigs，and a partial least square （PLS） prediction model was constructed for breeding pigs up to 100 kg body weight.The cross-validation results showed that an average determination coefficient of PLS test set was 0.68 and the average MAE was 3.2 day.The PLS model was tested using experimental data，and the results showed that combination of feeding traits and growth traits could be used to predict the growth rate of breeding pigs in advance.