Fuzzy Time Series and Grey Theory Forecasting for the Sales of Cleaning Products

Abstract

The Covid-19 pandemic has brought too many concerns in businesses whether they will survive or fail. This is what makes forecasting crucial since it guides businesses in order to make appropriate decisions. In a pandemic environment, forecasting is critical because there is little historical data available. Taking this into consideration, in this study, fuzzy time series (FTS) and grey model (GM), which do not require long past time series data are implemented for short term forecasting of a product sold on a dedicated social media account. After adopting monthly sales data of the related product, accuracy of fuzzy time series and grey model are improved with parameter optimization. According to the results of absolute percentage error (APE), both methods demonstrate superior forecasting accuracies when the product shows an increasing sales trend. This study can also be a reference for businesses that are positively affected by the pandemic due to increased sales and willing to act on the opportunities it has already emerged. © 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.