| Our Guest Post comes from recently retired Temple U. Professor Howard Weiss, creator of our POM and ExcelOM software |
There have been many articles and blogs written about the effect that COVID-19 is having on operations. One of the major areas that needs to be addressed is forecasting. Many companies rely on time-series forecasting as indicated in Chapter 4 of your textbook, but COVID-19 has caused the demand for some products to fall sharply, the demand for other products to rise sharply and the demand for other products to have temporary spikes mainly due to hoarding. Examples of each are in the figure and table below.
Using any of the time-series methods in these situations will cause the forecasts for 2021 and beyond to be incorrect. To overcome these problems one needs to look at using a different method or try to adapt the time-series method by replacing the outliers with demands that are more reasonable to assume had COVID not been a factor.
As the forecasting chapter indicates, rather than looking at past demand the forecasting method “may
be a subjective or intuitive prediction. It may be based on demand-driven data, such as a customer plans to purchase and projecting them into the future. Or the forecast may involve a combination of these, that is, a mathematical model adjusted by a manager’s good judgement.”
Classroom discussion questions:
- How could you modify the data outliers if you want to adjust a time-series method?
- If you change the method during COVID, how would you measure the quality of the forecast when using a new technique?