Conference Agenda

The normative (expected profit-maximizing) theory assumes that decision-makers are fully rational, but in reality, they deviate from the optimal response. In contract negotiations, contract parameters are often optimized based on the assumption of rational behavior. Deviations from rational behavior can lead to significant costs for all parties involved. To address this, we propose robust optimization to obtain contract parameters that are robust to deviations from rational behavior.

We apply this optimization approach to obtain the robustly optimal contract parameters for the buyback contract when there is limited knowledge of the buyers' responses.

We compare the robust contract to the normative contract if people behave according to known behavioral models, such as mean-anchoring, bounded rationality, or a behavioral model explicitly for the buyback contract. Additionally, we test different robust contracts in a lab experiment with actual human decision-makers. The results demonstrate that a robust contract, based on simple assumptions, can outperform the normative contract in terms of risk while not sacrificing average supply chain profit. This approach offers a practical way to set contract parameters without relying on assumptions about the distribution of responses, except for the support.

When binary classifications are wrong, operations managers face misclassification costs. While false positive outcomes lead to unnecessary mitigation efforts, false negative outcomes imply overlooking the class of interest. Translating the prediction probability into a positive or negative classification can be customized by adjusting the decision threshold, i.e., the cut-off probability. Our study shows that, despite being provided with all relevant information, decision makers do not select the optimal threshold that minimizes misclassification costs. We conducted incentivized, controlled laboratory experiments to investigate which decision problem characteristics and biases explain this boundedly rational behavior. Our findings are twofold. First, participants systematically select sub-optimal thresholds. We observe a significant interaction effect of class and cost imbalance on the deviation from the optimum. The deviation from the optimum increases in high stakes settings where more extreme thresholds would be optimal. Second, we find that the pull-to-center behavior can be explained by an anchoring and insufficient adjustment heuristic. In particular, we show that anchoring on the mean threshold does not explain the behavior sufficiently in settings with cost imbalance. In line with the notion of impulse balance, we find an additional reference point for which the expected misclassification costs for false alarms and missed hits are equal. Our findings suggest that this impulse balance equilibrium also serves as reference point for ex ante decisions and without loss aversion. Operations managers need to be informed about these boundedly rational preferences as sub-optimal decision threshold preferences result in 52% higher misclassification costs, on average.

Packaging parcels into delivery vehicles is a central logistic activity and, despite technological advancements, requires the work of human logistics professionals. Here, packing speed and efficiency are key factors, as inefficiencies result in higher transport costs and increased CO2 emissions. In an experimental study, we investigate how standardized packages influence human packing behavior and efficiency. We examine two opposing effects: Package standardization enhances efficiency by facilitating the easy placement of similarly sized packages, enabling workers to quickly find space-efficient packing solutions. Yet, standardized packaging results in unused space between items and their packaging, decreasing delivery vehicle utilization. The latter is a common phenomenon criticized by customers and media amid the ongoing sustainability debate. Insights from our experiment contribute to this discussion as they shed light on whether the empty space within packages can be sustainable if it boosts human packing efficiency and sufficiently reduces the empty space between packages.