Predictive Analytics in Industrial Processes Using LSTM Networks
Rahul Arulkumaran, Dignesh Kumar Khatri, Viharika Bhimanapati, Lagan Goel, Om Goel
Abstract
It is now well acknowledged that predictive analytics is an indispensable instrument for boosting the efficiency and effectiveness of industrial operations. Long Short-Term Memory (LSTM) networks, which are a form of recurrent neural network (RNN), have shown substantial promise among the numerous methodologies that have been used. This is because of its capacity to collect and describe temporal relationships in sequential data sequences. The purpose of this study is to investigate the use of LSTM networks in predictive analytics for industrial processes, with a particular emphasis on the advantages, disadvantages, and uses of these networks in the actual world. An overview of predictive analytics and the significance of its role in the optimisation of industrial processes is presented at the beginning of the research. The process of analysing previous data to make predictions about future occurrences is known as predictive analytics. This helps companies improve their decision-making processes and their operational efficiency. Industrial processes, which are characterised by data that is complex, dynamic, and time-dependent, may considerably benefit from sophisticated predictive models that are able to manage such complexities.