Mind the Gap between Demand and Supply: A behavioral perspective on demand forecasting Defended on Friday, 8 January 2016

Prior academic research has recognized human judgment as an indispensable decision-aid in demand forecasting although it is subject to a number of biases. Therefore, it is important to understand human judgment in forecasting to explain poor forecast decisions and attenuate their negative consequences for many related operational decisions. This dissertation is part of a growing research field in which human behavior and cognition are incorporated into analytical models of operations management. It aims to provide new insights into the role of human judgment in demand forecasting.

 

The functional specialization and differentiation inherent to most organizations usually shapes forecasting behavior in such a way that it benefits departmental goals and agendas. Lack of clear forecast ownership, diffused responsibilities and varying interests and incentives are often at odds with the organizational goal of producing accurate forecasts. The first study identifies and describes the potential benefits of forecast ownership and mechanically combined departmental forecasts on the tendency to over- and under-forecast demand. The findings of the second and third study show that departmental roles offer a particular frame with which forecasters interpret information and make decisions. The fourth study examines how forecasters use historic data and extrapolate information to determine future trends. The findings indicate that forecasters cannot clearly distinguish actual from illusionary trends and persistent changes from random variations in time-series.

 

Together these studies underscore the importance of understanding the underlying cognitive and motivational processes in order to devise strategies for making accurate forecasts.

Keywords

Demand forecasting, behavioral operations, sales and operations planning, forecast ownership, negotiation, social value orientation, goal concern, illusionary trends, time-series forecasting


  • Share on