The ongoing climate change has obligated stricter limits for CO2 emissions on the automotive industry. Research on the efficient production of lithium-ion batteries (LIB) for electric vehicles is underway. However, LIB production faces challenges due to the high level of complexity and uncertainty in the correlations between its process parameters. Therefore, as a powerful tool for uncertainty modeling, Bayesian networks (BN) play an important role in analyzing LIB production data. However, most of the process parameters in the LIB production data take continuous values and contain missing records, which makes the BN learning process difficult and inefficient. This paper pro-poses an iterative algorithm based on discretization and imputation techniques to learn BN structure from LIB production data with continuous variables and missing values. This algorithm combines the cluster-based discretization technique for continuous variables and the data augmentation technique for missing data. The experimental results show that this algorithm can improve the quality of the learned network structure.     «

The ongoing climate change has obligated stricter limits for CO2 emissions on the automotive industry. Research on the efficient production of lithium-ion batteries (LIB) for electric vehicles is underway. However, LIB production faces challenges due to the high level of complexity and uncertainty in the correlations between its process parameters. Therefore, as a powerful tool for uncertainty modeling, Bayesian networks (BN) play an important role in analyzing LIB production data. However, most o...     »