Abstract Number: 003-0090
Title of the paper: An investigation of the impact of VMI on supply chain performance
Conference: Sixteen Annual Conference of POMS, Chicago, IL, April 29-May 2, 2005.
Authors: Chungsuk Ryu
School of Management, State University of New York at Buffalo
310G Jacobs Management Center Buffalo, New York 14260-4000
E-mail: ; Phone: (716) 743-8266; Fax: (716) 645-6117
Nallan C. Suresh
School of Management, State University of New York at Buffalo
355 Jacobs Management Center Buffalo, New York 14260-4000
E-mail: ; Phone: (716) 743-3279; Fax: (716) 645-6117
Paper title: An investigation of the impact of VMI on supply chain performance
Abstract:
This study investigates the impact of vendor managed inventory (VMI) systems on supply chain performance under different conditions of decision authority and information sharing. Based on both cost-based and profit-based models, supply chain performance is evaluated in terms of costs and profits, and compared with the traditional buyer managed inventory system. The analytical model including only inventory costs confirms that VMI results in less supply chain costs than the traditional system. The numerical examples with the profit-based model, however, show that VMI does not always outperform the traditional supply chain system. The cost savings due to VMI are due to the dramatic reduction of buyer's costs, but, in most cases, the supplier appears to incur higher expenses. The study seeks to identify the conditions that may remedy the supplier’s losses caused by VMI. These results also suggest the contractual elements and forecasting needed for more equitable benefits for both supplier and buyer.
Key words: vendor-managed inventory, supply chain coordination, economic modeling
1. Introduction
There have been various supply chain initiatives that have appeared to be effective tools for supply chain management, such as QR (Quick Response) and CR (continuous Replenishment). VMI is another kind of these supply chain initiatives and recently has received much attention from both academia and industry. In fact, VMI has been already applied to various industries, including consumer goods retails such as Wal-mart and Kmart stores (Aviv 1998, Gnanasekaran 2000, MacKenzie 1999), telecommunication such as Bell Canada (APICS 2003), healthcare systems such as a multi-hospital integrated delivery system (Haavik 2000), home delivery services such as e-grocery (Smaros and Holmstrom 2000), and the automotive industry (Cooke 1998). Despite the recent introduction of VMI into industrial applications, the VMI program has resulted in significant success in lowering inventories and increasing customer service levels.
The main goal of this study is to determine the impact of VMI on performance of the supply chain system based on the mathematical model. The performance evaluation is conducted in two ways – through model analysis and numerical examples. First, the simple inventory management models that represent two distinct supply chain systems are analyzed under optimal conditions. Second, the complex mathematical models are developed to include more detailed cost items as well as revenue and numerical examples are used to compare the profits of two different supply chain systems under specified contexts. Based on the results of model analysis and numerical examples, this study attempts to determine whether the VMI system is superior to the traditional system, as most previous studies and real cases had supported. Further detailed investigation is also conducted to examine the pattern of VMI's performance under different operational contexts.
The major contributions of this research are twofold. First, this study not only examines the financial superiority of the VMI supply chain system over the traditional one but also discovers the potential that VMI can be beneficial to every participant of this program. By acquiring sufficient demands and order quantity, even the supplier can be financially benefited by participating in the VMI program. Second, the numerical examples in this study reveal the specific conditions that enlarge the cost savings due to the VMI program. Both supplier and buyer can optimally exploit the advantage of VMI by properly managing forecast and price contract.
2. Background of Research
In order to develop the model for the VMI system and to analyze its impact on the supply chain system, this study first conducts investigations on fundamental characteristics of VMI. VMI originated from the buyer-supplier partnership that is designed to improve the performance of the whole supply chain system (Simchi-Levi, et al. 2000). Due to the nature of the business, the supply chain system has an inherent limitation, in that there have been trade-offs of profits and conflicting goals between supplier and buyer. Reduction of inventory in a supplier's warehouse implies cost savings for the supplier, but the reduced inventory may sacrifice the service level and cause loss of sales, which may be counterproductive.
There are distinct types of retailer-supplier partnerships. The quick response (QR) strategy enables the supplier to receive point-of-sales (POS) data from retailers and use this information to improve production and inventory management in accordance with actual sales of retailers. The basic form of QR focuses solely on information-sharing activities between the supplier and the retailer, and other activities, including ordering, inventory management, and replenishment, are independently performed by the retailer and the supplier.
In a continuous replenishment (CR) strategy, the supplier uses POS data received from retailers to schedule shipments to meet the previously agreed upon inventory level or service level. By applying sophisticated techniques for inventory control and shipment decisions, both the retailer and the supplier pursue continuous improvement in terms of cost savings of inventory and shipment operations.
Vendor Managed Inventory (VMI) is considered to be the advanced form of the buyer-supplier partnership, because it goes beyond information-sharing activities and also covers the consignment scheme. In the VMI system, a supplier has full responsibility for managing the inventory at buyers' warehouses. According to the detailed agreement between the buyer and the supplier for initiating the VMI program, the supplier controls the buyer's inventories by determining appropriate levels of orders and inventory based on POS data from the buyer. Thus, the VMI program enables the buyer to eliminate the burden of ordering and inventory management. For the supplier, the VMI program is an efficient tool for synchronizing inventory, replenishment, shipment, and production activities with buyers' actual sales, and systematically reducing the total costs associated with those activities.
The main goals of the VMI are twofold - to lower the inventory level and to improve the service level at the same time (Levy and Grewel 2000). Traditionally, it is thought that these two goals cannot be achieved simultaneously, but they are rather compromised. Both the buyer and the supplier hold a certain level of inventories in their own warehouses to secure product availability for their customers. Keeping safety stock is a traditional way to minimize the occurrence of stockouts. In fact, it used to be the norm for inventory holding cost and customer service level to be positively correlated, and lowering the inventory level and increasing the service level have not been possibly achieved at the same time through any traditional management techniques. VMI was created to overcome this limitation of traditional management techniques. In the VMI system, the buyer eliminates inventory holding costs. The supplier also reduces his inventory holding cost and increases the service level by controlling the buyer's inventory according to his or her best interest in scheduling production, delivery, warehousing, and replenishment.
Inventory management is the key operation that is most affected by the VMI program, and it is also the process where the most benefits of VMI take place. First of all, according to the formal definition of VMI, the responsibility for managing buyers' inventories transfers to the supplier. In addition to the change in the right of management, the VMI implementation brings significant changes in inventory management, including the ordering process, replenishment policy, and information usage (Achabal, et al. 2000, Holmstrom 1998a, Holmstrom 1998b). Most of the existing studies addressing the inventory control issue in the VMI system focus on the evaluation of VMI performance, particularly compared with the traditional buyer managed inventory system (Aviv and Federgruen 1998, Disney, et al. 2003, Disney and Towill 2002, Dong and Xu 2002, Southard 2001, Waller, et al. 1999). According to the literature review, existing studies about inventory control under the VMI program address an almost ubiquitous issue of whether VMI is superior to the non-VMI program. However, they develop various models for the VMI system based on different operational contexts. In many cases, they use very complicated stochastic models, and evaluate their own VMI systems in terms of different performance criteria. By implication, the current stream of studies shows that there exists no consensus on the proper supply chain model for the VMI system. Therefore, it may be a good idea to develop a general VMI model representing basic features of VMI rather than to start with a complex and specified VMI model in the first place, and then expect development of further advanced models that are sufficient to reflect a real-life of the particular VMI system.
3. Analytical Comparison of Supply Chain Costs
In this section, different supply chain systems are compared in terms of their economic performances based on analytic models. Two supply chain systems are considered - (1) traditional supply chain system and (2) VMI system. Each supply chain system is defined as a distinct mathematical form, which represents unique properties of cost calculation and decision variables. The performance measurement considered in the comparison between different supply chain systems is total cost, which is related to only the inventory management. The total cost is composed of setup cost (or ordering cost) and inventory holding cost in both sides of buyer and supplier. The performance to be evaluated is the total cost of a whole supply chain system and the analyses do not account for the comparison between costs of individual supplier chain players. The inventory management in the analyses is based on the EOQ model and every analytical model follows all EOQ principles for cost calculation and ordering decision.
Each supply chain model indicates that either buyer or supplier decides the order quantity in a way to minimize his or her own cost. Detailed cost calculation and decision variables are defined according to the specific type of the supply chain system. Table 1. describes variables used in the analytic models and their definitions.
Table 1. Model notation for cost comparison
Supplier side / Buyer side/ Total cost / / Total cost
/ Estimated total cost / / Annual market demand
/ Estimated annual market demand
/ Annual production rate / / Order quantity
/ Planned annual production rate / / Estimated order quantity
/ Transfer price paid by buyer / / Sales price to the market
/ Profit margin
/ Annual inventory holding cost / / Annual inventory holding cost
/ Unit production cost / / Annual stocking cost for inventory
/ Unit transportation cost / / Annual financing cost for inventory
/ Setup cost per buyer’s order / / Ordering cost per order
3.1. Traditional supply chain system
The traditional supply chain system is defined as the system within which the buyer and the supplier have independent decision authority and cost structure. In this system, the buyer possesses the authority over a decision on ordering products to the supplier. The total cost of the buyer is composed of the ordering cost and inventory holding cost occurring at the buyer’s warehouse as the following Equation (1).
(1)
In the supplier’s side, the total cost is represented as the setup cost plus the inventory holding cost occurring at the supplier’s warehouse. Equation (2) shows the supplier’s cost (Banerjee 1986).
(2)
In the traditional system, the buyer decides the order quantity based on the information about the annual market demand and cost items. Through the simple first order condition in Equation (3), the optimal order quantity to minimize the total cost can be obtained in Equation (4).
(3)
(4)
With the optimal order quantity decided by the buyer, the total supply chain cost can be represented as following.
(5)
Equation (5) indicates the total cost of the traditional supply chain system with the optimal order quantity determined by the buyer.
3.2. VMI system
The main feature of VMI, which differentiates the VMI system from the traditional supply chain system, is that the supplier has a full authority over the inventory management at the buyer’s warehouse. Thus, the decision on order quantity is given to the supplier in the VMI system, contrary to the tradition case wherein the buyer has it. The supplier may pay costs associated with the buyer’s inventories and ordering. In some cases, the buyer may be responsible for paying those costs. In this analysis, we assume that the supplier has a full responsibility for paying any costs associated with the buyer’s inventory and ordering. Equation (6) indicates the total cost of the supplier in the VMI system.
(6)
When the model accounts for only the inventory management in the supply chain system, the VMI system, in fact, denotes the centralized system within which the supplier as a sole decision maker decides the order quantity based on every information available from both buyer’s and supplier’s sides in a way to minimize the total cost of the whole supply chain system (here, the total cost of the supply chain system is identical to the total cost of the supplier).
Based on the supplier’s cost in Equation (6), the optimal order quantity can be obtained by the following equations.
(7)
(8)
According to the optimal order quantity in Equation (8), the total supply chain cost of the VMI system can be represented as it appears in Equation (9).
(9)
Equation (10) shows the comparison between the traditional supply chain system and the VMI system in terms of the rate of total supply chain costs.