The integration of data provenance with mass customization production models is the basis for achieving data quality improvement and technological innovation development. Based on the random nature of orders, a series of opportunities and challenges were brought to data quality by a large variety of products and small batches, the data provenance model based on the open provenance model was proposed. Combining the characteristics of the data lifecycle in mass customization, by finding the source data from the massive amount of data and recording the changes in the process of data dissemination to manage them, a data provenance information exchange model was built in the manufacturing sector under mass customization. The open provenance model for the supply chain, the open provenance model for product attributes and the open provenance model for the production stage were established, thereby the problem of information exchange among or within enterprises in mass customization was solved and the data tracking function and the control and improvement of data quality were realized.