Date of Award


Document Type

Open Access Dissertation


Moore School of Business


Business Administration

First Advisor

Manoj Malhotra

Second Advisor

Pelin Pekgün


Profitably balancing demand and supply is a continuous challenge for companies under changing market conditions, and the potential benefit of collaboration between supply chain partners cannot be overlooked by any firm who strives to succeed. One of the key elements to successful collaboration is sharing of forecast information between supply chain partners. However, when supply shortage is expected, buyers may inflate order quantities and/or order forecasts to secure sufficient supply. An important question that arises is how the supplier should allocate inventory to customers when shortage exists. Literature shows that certain allocation policies can reduce buyers’ order inflation behavior. However, this has not yet been empirically shown for order forecast inflation behavior, nor incorporating the behavioral aspects of decision makers. In this dissertation, through behavioral experiments using a supply chain simulation game, we investigate the impact of different capacity allocation mechanisms and information disclosures of a supplier on buyers’ forecast sharing and ordering behavior.

We first investigate the buyers’ order forecast sharing behavior in a single-suppliertwo- buyer supply chain. Our behavioral study shows that forecast-accuracy based allocation, where the supplier allocates more capacity to the buyer with better forecast accuracy, can significantly improve order forecast accuracy relative to uniform allocation, where the supplier equally allocates capacity to the buyers. Under both policies, particularly uniform allocation, the order forecast accuracy is improved with the supplier’s information disclosure on the policy. Next, we focus on buyers’ ordering behavior, and formulate a single-supplier-single-buyer base-stock inventory model under constrained supply. We validate our analytical results through numerical simulation, which is then extended to the single-supplier-two-buyer case. We next compare the buyers’ optimal decisions from the simulation with the actual decisions in our behavioral study, and find that buyers in the experiment show a significantly lower profit performance ranging from 0.8% to 14.1%. Using structural estimation modeling techniques, we estimate the buyers’ perceived overage/underage cost ratios from the experiment, and conclude by conducting a detailed analysis on the factors that affect buyers’ ordering decisions.

In addition to academic contributions, our results provide insights for practitioners to understand buyers’ strategic behavior and help with designing capacity allocation strategies.


© 2017, Minseok Park