Empowering Supply Chain Transparency and Risk Management through Hybrid CNN and BiGRU Approach

Everything from needs to luxuries has been impacted by the COVID-19 pandemic in the manufacturing sector, and that includes the automotive industry. Thus, there is an immediate need for research into the automotive supply chain’s sustainability, transparency, and competency development practices. For the automotive supply chains to successfully apply the Transparency for Sustainability (TfS) architecture in the long run, it also specifies a set of skills that are necessary. The three steps that comprise this proposed method are preprocessing, feature selection, and model training. The data had to be preprocessed in order for it to be ready for analysis. The feature selection procedure is comprised of the LDA topic model, term frequency, and correlation analysis. A GRU-DT was utilized for the model’s training. The suggested method outperforms GRU and DT with an average accuracy of 96.29%.