This PhD thesis was motivated by the simple observation that the objectives of distinct supply chain managers are often conflicting. This problem is usually addressed via supply chain contracts that are designed to align the incentives of the different supply chain partners to the overall benefit of the entire supply chain, when seen as a whole. In this way, the long-term prosperity and viability of all the firms that participate in the supply chain can be ensured. In order to study the efficiency of different supply chain contracts in attaining the theoretical optimum performance, there exist a number of standard normative models that predict the decisions of perfectly rational decision makers. But supply chain partners might in reality not make the perfectly rational decisions that these theoretical models predict. This may be because they may lack the required information, or experience cognitive limitations and individual preferences or have only a finite amount of time available. For this reason, they might have to settle at satisficing choices. The result of these ‘boundedly rational’ decisions is a real world of different than expected interactions. Since in this world the standard normative models retain limited predictive power, this PhD thesis aims to explore the true efficiency of the simplest supply chain contract that can exist, namely, the wholesale price contract. In addition, this PhD thesis provides some useful recommendations that aim to help supply chain managers make price and order quantity decisions that would be better aligned with the interests of the overall supply chain. To this end, this study applies an original approach that supplements experiments with human subjects with Agent Based Simulation experiments. In greater detail, informal pilot sessions with volunteers were first conducted, during which knowledge of the underlying decision making processes was elicited. Appropriate Agent Based Simulation models were subsequently built based on this understanding. Later on human subjects were asked to interact with specially designed versions of these Agent Based Simulation models in the laboratory, so that their consecutive decisions over time could be recorded. Statistical models were then fitted to these data sets of decisions. The last stage of this approach was to simulate in the corresponding Agent Based Simulation models all possible combinations of decision models, so that statically accurate conclusions could be inferred. This approach has been replicated for both the simple newsvendor setting and the beer distribution game. The results that are obtained indicate that the overall efficiency of the wholesale price contract differs significantly from the theoretical prediction of the corresponding standard normative models. It varies greatly and depends largely on the interplay between the pricing and ordering strategies that the interacting supply chain partners adopt. In view of this, real world echelon managers are advised to use prices as an effective mechanism to control demand and, also, keep their total supply chain profits in mind when making their respective decisions.
展开▼
