The Folly of Forecasting: The Effects of Sales Forecast Accuracy and Bias on inventory and Production Decisions under Aggregated and Disaggregated Forecasting Regimes

In this study, we use field and archival methods to examine the effects of sales forecast accuracy and bias on inventory and production planning within a large manufacturing organization. While the operations literature is replete with analytic and simulation studies of sales and operations planning (S&OP) systems, this research fails to account for the role that incentives, biases, and organizational dynamics play. In stark contrast to prior literature, we find no associations between sales forecast accuracy and either inventory levels or shorthorizon production plan stability at our research site, suggesting that operations manager discount the value of sales forecasts and, instead, rely on their own historical data. However, during the period of study, our research site modified its S&OP process to incorporate a mechanism for sales managers to disaggregate sales forecasts into its two constituent components: sales forecasts of relatively certain demand information, and highly uncertain contingent demand information. Our findings show sales forecast accuracy improved following sales forecast disaggregation with corresponding improved production plan stability. However, while greater forecast accuracy was associated with lower inventory, this was offset by higher inventory owing to increased positive sales forecast bias. More generally, the disaggregation of sales forecasts restored the associations between sales forecast accuracy and both inventory and production stability presumed in operations research, suggesting improved reliability of sales forecasts from the perspective of production planners. Our study provides novel insights regarding the role of forecasting within the organizational context.

This seminar is organised by the Erasmus Accounting Research Group.