Decision biases in revenue management revisited: Dynamic decision-making under stationary and nonstationary demand

Abstract

State-of-the-art revenue management systems combine forecasting and optimization algorithms with human decision-making. However, only a few existing contributions consider the behavioral aspects of revenue management. To extend the related research, we examine the impact of nonstationary demand and two dynamic decision tasks. We examine human decision-making strategies and biases by implementing a related experimental design in a laboratory study and comparing participant decisions to systematic heuristics. Our results highlight that participants struggle to accommodate a nonstationary willingness to pay. In that, they exhibit a combination of optimism and loss aversion biases. We further find that participants anchor their decisions on customers' willingness to pay. We draw implications and further research opportunities to behaviorally inform the design of symbiotic analytics systems from these results.