Adaptive grey model (AGM) approach for judgemental forecasting in short-term manufacturing demand

Abstract

The covid-19 pandemic has created problems in every manufacturing sector and has posed considerable challenges to pharmaceutical, healthcare, and sanitation companies. The challenges faced are particularly daunting for pharmaceutical companies producing vaccines with ever-growing demand and shorter and shorter deadlines to fulfill them. Further, due to the vaccine's novelty and unprecedented demand, there is a lack of any available data on which traditional forecasting methods can be used. In this paper, we attempt to propose a solution by utilizing the Grey Systems Theory, particularly the AGM (1, 1) model, which has been used to significant effect for problems involving uncertain / lack of data to forecast the demand for vaccines. The experimental results obtained showed that our proposed model successfully generated accurate forecasts with a small dataset and minimal error. Additionally, judgmental forecasting has been used to qualitatively assess the future scope of vaccine manufacturing as well as the use cases of the model. We can thus effectively say proposed AGM (1,1) model is a lucid method to forecast the demand for vaccines.