Event Modeling for Demand Planning
Traditionally planners have asked how to differentiate between baseline forecasts and incremental deltas. How do you differentiate a baseline forecast from an event-driven spike or drop in demand?
Baseline is business as usual. This is normal demand that occurs when there are no unusual events or promotional incentives offered to customers. Ironically, everyday may be a promotion day in the world of Consumer goods companies. So care should be taken to modeling for events.
Good demand planning software applications provide features to plan for events:
1. Some offer statistical models that can help precisely model the impact of events.
2. Many others offer features to plan and differentiate promotional plans manually through planner intervention but perhaps not advanced models that can forecast lift.
See the chart above that illustrates baseline and event spikes.
A typical intervention model using either Box-Jenkins or Holt-Winters methods can help model events as interventions. This allows you to model even unusual events such as the 2020 Covid19 Pandemic in the United States. A simplified event modeling equation will look like -
Unusual events are also known as black swans. How do we prepare for Black Swan events?
Black Swan events are those highly unpredictable events that occur out of nowhere and cause a systemic shock to the process and the way we do things. They may perennially alter the way we conduct our business.
Click here to read more about The importance of Black Swans in our planning process - https://forecastingblog.com/?p=586