Gonul, Sinan, Paul Goodwin, Paul Goodwin and Önkal, Dilek (2020) Why/when can scenarios be harmful for judgmental demand forecasts and the following production order decisions? In: ISF 2020: 40th International Symposium on Forecasting, 26-28 Oct 2020, Rio de Janeiro, Brazil.
|
Text
ISF2020_presentation_Oct2020_MSG_v3.pdf - Published Version Download (2MB) | Preview |
Abstract
Judgmental demand forecasting constitutes an integral part of inventory management and production planning activities within organizations. Among forecasting academicians and practitioners, there is the generally accepted belief that the presence of scenarios is largely beneficial for future planning and may aid the decision makers in producing these demand predictions. However, there is only circumstantial evidence and some studies report controversial findings. One recent experimental work (Gonul, Goodwin & Onkal, ISF2019) investigated the interaction between the existence of optimistic and pessimistic scenarios and the presence of time-series information alone in the task of generating demand forecasts and the following production order decisions. The findings revealed that providing scenarios worsened forecast accuracy and swayed the production order decisions further away from the optimality. What were the reasons underlying these results? Why did scenarios degrade forecasters’ accuracy? This current work is an attempt to disentangle this puzzle by trying to shed some light on these controversial findings through the application of a Generalized Estimating Equations (GEE) model. The findings from this analysis will be discussed to guide future research on scenarios and judgmental forecasting.
Item Type: | Conference or Workshop Item (Other) |
---|---|
Subjects: | N100 Business studies |
Department: | Faculties > Business and Law > Newcastle Business School |
Related URLs: | |
Depositing User: | John Coen |
Date Deposited: | 03 Mar 2021 15:14 |
Last Modified: | 31 Jul 2021 15:17 |
URI: | http://nrl.northumbria.ac.uk/id/eprint/45607 |
Downloads
Downloads per month over past year