Monitoring of multivariate calibration models in the absence of new reference values: The regression case

Multivariate calibration models are one of the most predominant tools in analytical chemistry for the non-destructive quantification of constituents. While pipelines, methods, and software are widely available for training calibration models, the monitoring strategies to track the prediction performance in time are still limited and are one of the emerging research topics in chemometrics and machine learning. Moreover, for the sake of efficiency, the development of modern monitoring strategies requires tools that do not require new reference values at each time point. The current study addresses the problem of monitoring calibration models for regression. We present a monitoring paradigm and propose two methods to monitor the performance of the model. The two methods focus on bias drift and spectral variation drift. We present the derivation of the methods and their application to two real case studies from the agrofood industry involving milk composition and apple harvesting. Conclusions and recommendations are presented for practitioners, as well as views for future work in this domain of research.

Automated summary

Multivariate calibration models are commonly used in analytical chemistry for non-destructive quantification of constituents. While training calibration models is well-established, monitoring strategies for their prediction performance over time are still limited. This study addresses the problem of monitoring calibration models and proposes two methods focusing on bias drift and spectral variation drift. Real case studies involving milk composition and apple harvesting are used to demonstrate the application of these methods. Conclusions, recommendations, and future research directions are provided.