چگونه می توانیم عملکرد قراردادهای زنجیره تامین را بهبود دهیم؟ یک مطالعه تجربی

Although optimal forms of supply chain contracts have been widely studied in the literature, it has also been observed that decision makers fail to make optimal decisions in these contract setups. In this research, we propose different approaches to improve the performance of supply chain contracts in practice. We consider revenue sharing and buyback contracts between a rational supplier and a retailer who, unlike the supplier, is susceptible to decision errors. We propose five approaches to improve the retailer’s decisions which are in response to contract terms offered by the supplier. Through laboratory experiments, we examine the effectiveness of each approach. Among the proposed approaches, we observe that offering free items can bring the retailer’s effective order quantity close to the optimal level. We also observe that the retailer’s learning trend can be improved by providing him with collective feedbacks on the profits associated with his decisions.

Supply chain contracts have been extensively studied by researchers. A large stream of research in this field considers a two echelon supply chain consisting of a supplier (seller) and a retailer (buyer) who sells a seasonal (fashion) product to a market with random demand. Due to usually lengthy production and distribution lead times (Fisher and Raman, 1996), the retailer has to decide about the order quantity (initial inventory level) long before the start of the selling season. Under this setup, the retailer faces a classical Newsvendor inventory problem. That is, if the retailer’s order quantity is less than the realized demand, the retailer faces with inventory shortage (unmet demand), while if the order quantity is more than the realized demand the retailer is left with unsold inventory, which should be discarded or salvaged with a very low price. The classical Newsvendor solution identifies the optimal order quantity which maximizes the retailer’s expected profit. In a simple wholesale price contract, the retailer faces all the risk and the wholesale price that maximizes the supplier’s profit causes the retailer to order a quantity less than the value that maximizes the channel profit (Spengler, 1950). To avoid this situation, the supplier can offer a contract in which she provides the retailer with proper economic incentives to order the quantity that maximizes the supply chain profit (a coordinating contract). In this research, we consider two types of coordinating contracts: revenue sharing and buyback. In a revenue sharing contract, the supplier offers a relatively low wholesale price but asks the retailer to share part of the revenue of every item sold. Revenue sharing contracts have been used successfully (among other industries) in the video-rental industry (Cachon and Lariviere, 2005). In a buyback contract, the supplier buys back any unsold item from the retailer with a price lower than the wholesale price. Buyback contracts are common practice in the publishing, software, and pharmaceutical industries (Padmanabhan and Png, 1995). In both contracts, the supplier shares part of the retailer’s risk of facing a random demand. Although the theoretical benefits of optimal Newsvendor solutions and coordinating contracts have been widely studied, it is also known that retailers fail to place the optimal order quantities in practice. Fisher and Raman (1996) and Corbett and Fransoo (2007) show industry evidence that managers’ inventory decisions systematically deviate from the optimal quantities. Fisher and Raman (1996) show that managers’ less-than-optimal production quantity, at a ski apparel manufacturer, resulted in a profit which was 60% less than their calculated optimal profit. Corbett and Fransoo (2007) study inventory decisions of 51 small businesses. They show that the inventory decisions deviates from the optimal decisions calculated by a Newsvendor model. They show that the deviations are consistent with the prospect theory predictions. Almost all the research papers in this field have focused on finding how and why decision makers’ order quantities deviate from the optimal values (we will briefly review these papers in Section 2). The more important question of how this deviation could be avoided, however, has received little attention in the existing literature. As an attempt to fill this gap, we explore possible ways through which we can improve the performance of a supply chain by inducing the retailer to choose order quantities close to the channel’s optimal order quantity. Here, we consider an ideal supplier whose decisions are rational and sets the parameters of the contract according to their theoretical optimal values. The retailer, however, is assumed to be prone to behavioral misjudgments and errors. Therefore, the order quantities chosen by the retailer can systematically deviate from the optimal values. The retailer’s suboptimal decision has a negative impact on his profitability as well as the supplier’s and the channel’s profitability. Hence, the supplier tries to design the contract terms or offer additional information to address the inefficiency in the retailer’s decision and increase her (and consequently channel’s) profit. We explore five approaches which could possibly improve the performance of a revenue sharing or buyback contracts. We first identify the concept or logic behind each approach and then verify its effectiveness through laboratory experiments. Three of these approaches concern the contract terms which the supplier offers the retailer. The other two approaches concern providing the retailer with additional information or feedback that might help him to make better decisions. In our first approach we consider a new type of contract which is a combination of revenue sharing and buyback contracts. The second approach examines the possibility that risk-aversion is the source of suboptimal decisions. If this is the case, then a coordinating contract that is designed for a risk-averse (not a risk-neutral) retailer should result in an optimal order quantity. The third approach considers the offering of free items by the supplier. If the number of free items offered increases with the size of the order, the retailer might be encouraged to increase his order quantity. Moreover, these free items increase the number of items in the supply chain. In our fourth approach we examine the impact of providing the retailer with visual information about the nature of demand uncertainty. This could possibly discourage the retailer to follow shortsighted strategies such as demand chasing. In our last approach we provide the retailer, in each decision round, with a new performance measure that shows the collective impact of last decision if the current order quantity were the decision for previous decision rounds as well. This new piece of information should also discourage the retailer to follow a demand chasing strategy. The remainder of this paper is organized as follows. Section 2 reviews the related literature. Section 3 presents the theoretical background of the problem, explains the general experimental setup, and shows the results of our benchmark experiments. 4, 5, 6, 7 and 8 present the five studies through which we explain and investigate the effectiveness of each of our approaches to improve the performance of the supply chain. Section 9 concludes the paper with a summary of our results.

In this research, we examine a two-echelon supply chain consisting of a rational supplier and a retailer who is prone to behavioral errors. We show (like others before us), when the supplier offers a coordinating contract (either revenue sharing or buyback), the retailer systematically fails to place an order with optimal quantity. This sub-optimal behavior, in turn, results in less-than-optimal profits for all parties and the supply chain as a whole. We contribute to the existing literature by proposing five approaches to improve the decisions made by the retailer. We verify the effectiveness of each approach through laboratory experiments. The first three approaches concern the contract terms offered by the supplier. These approaches are (1) combined contracts, (2) contracts designed for risk-averse retailers, and (3) contracts with free-item offering. Among these approaches, we show, only the contracts with free-item offering can actually bring the order quantities close to the optimal level and coordinate the supply chain. The next two approaches concern extra information and feedback for the retailer. These are (4) providing the visual pattern of demand randomness and (5) providing a collective feedback on each decision. We show that the collective feedback (would-be total profit) can create a stronger learning process in the revenue sharing contract, which means decision makers can learn from their prior decisions and eventually place close-to-optimal order quantities. This approach is not effective in the buyback contract. A general takeaway from this research is that it is possible to improve the results of decisions made by the retailer either through a contract mechanism or through carefully designed feedback. It is interesting to note that the two effective approaches that we find in this research improve the performance of the contract in two very different ways. The free-item approach (or its adjusted counterpart) does not improve the retailer’s decisions. Instead, it adds a proper number of items to the items ordered by the retailer. Therefore, it increases the total number of items in the supply chain. The required change in the wholesale price is so small that it does not change the decision maker’s ordering behavior. Hence, the resulting total number of items (effective order quantity) increases to a number very close to the optimal order quantity. On the other hand, the collective feedback approach improves the retailer’s decisions by weakening the demand chasing behavior. One of the reasons behind the retailer’s suboptimal decisions is argued to be the decision maker’s limited attention span. Having a limited attention span, the decision maker mostly focuses on the feedback from the latest decision which is either a shortage or excess of inventory. This shortsightedness results in the demand chasing behavior. Providing a collective feedback (the potential impact of a decision on all previous selling seasons) can help the decision maker to overcome the tendency to chase the random demand. The collective feedback, in fact, shows a more realistic value of each decision in each selling season. It is also interesting to note that these two approaches are less effective on Buyback contracts. The reason behind this behavior can be attributed to the decision makers’ loss aversion behavior (Katok and Wu, 2009). In a buyback contract the wholesale price is higher than the wholesale price in a revenue sharing contract. In the free-item approach, higher initial wholesale price means that the price increase due to offering free items is more noticeable by the decision maker. Therefore, the retailer tends to order fewer items to reduce the risk of loss due to inventory overage. Similarly, in the collective feedback approach, when the wholesale price is high, the risk of loss due to inventory overage attracts the decision maker’s attention. This prevents the retailer from paying enough attention to the collective feedback. Therefore, the demand chasing behavior prevails. This research also contributes to the supply chain contracting literature by introducing theoretical forms of three new contracts, which are extensions of revenue sharing and buyback contracts. These are (a) revenue sharing contracts for risk-averse retailers, (b) combined contracts (combination of revenue sharing and buyback), and (c) revenue sharing and buyback contracts with free items. For each type of contract, we derive contract parameters that theoretically coordinate the supply chain. Having the results of this research, it would be interesting to explore the possible approaches that can improve other forms of contracts (such as two-part tariffs and quantity discount) in practice. This could be a possible avenue for future research.