ABSTRACT
During the COVID-19 pandemic, the information related to medical oxygen demand and supply was on the rise. However, it was noticed that there was lack of method or platform to predict the oxygen demand for future requirements, which was essential in the supply chain and logistics of oxygen to treat COVID patients. Through Data Analytics techniques implemented in this chapter, an effort has been made with Susceptible, Infectious, and Recovered (SIR) model and Polynomial Regression machine learning model to create the methodology and platform for predicting medical oxygen demand and to gather relevant data. ) This chapter has shown the scheme for the collection of oxygen-related data, its visualization, and forecasting oxygen demand by utilizing data available related to COVID-19. This chapter discusses a way to predict medical oxygen demand in a particular region based on outputs from standard SIR model and Polynomial Regression machine learning model. Further, it has given a way to predict region- or location-specific oxygen demand based on different input parameters provided. The results shown by this model are reasonably close when compared with actual data. For the first time, SIR model has been successfully used to predict the oxygen demand with an accuracy of around 84.12%. Such a timely prediction pertaining to oxygen demand will be helpful to bridge the gap between its demand and supply and to have an effective distribution to the end user by means of a Web portal.
