Controlling inventories in a supply chain: A case study
Introduction
One of the most important aspects affecting the performance of a supply chain is the management of inventories, since the decisions taken in this respect have a significant impact on material flow time, throughput and availability of products. Particularly interesting is the problem of coordination in the replenishment of multiple products when they share common resources (e.g. same mode of transportation), with the idea of saving fixed costs. Quite some papers have been published on this joint replenishment problem (JRP), but case studies are rare. In order to keep the analysis tractable, most studies on the JRP assume simplified problem settings, such as a simple transportation cost structure.
The objective of this article is to present a case study on joint replenishment policies in which we observed non-linear transportation costs and minimum order quantities for the individual items. This in contrast to existing JRP models (see the literature review in Section 3), which use in their formulation item set-up costs instead of minimum order quantities and consider constant transportation cost.
In the next section we describe the case study. Section 3 presents a literature review on joint replenishment. Section 4 describes a simulation model and its main assumptions. In Section 5, we discuss the results of the experiments carried out with the simulation tool. Section 6 presents analytical considerations using the EOQ procedure, and the final conclusions are presented in Section 7.
Section snippets
Case study
The company: We consider a start-up company that sells gift items through retail stores in the Netherlands and Belgium. The company orders the items from a manufacturer located in China, which in turn receives raw materials and components from a variety of suppliers (see product description). The company keeps inventory of items at a distribution center (DC) in the Netherlands by means of a Vendor-Managed Inventory contract with the distributor, who is responsible of sending out the items to
Literature review
Inventory models found in papers related to the JRP literature with stationary demand basically fall in two main categories according to the nature of demand: deterministic and stochastic models. In the deterministic methods it is assumed that the major ordering cost is charged at a basic cycle time T and that the ordering cycle of each item is some integer multiple of T, which is called a policy. In this line of research Goyal (1974) proposed a solution method based on enumeration,
Simulation model
A simulation model was built to analyze the problem in which the net inventory and inventory position at the DC are controlled for each item with a time step of one week. The demand of each item is considered normally distributed with parameters given in Table 3. Demands not met from stock are lost. For demands with large values of the c.v. we cut off the negative part of the normal distribution by setting to zero all negative demands, which caused only a marginal distortion. In Table 4 we
Experiments and discussion of results
Case 1a. No intermediate stocks (see Section 4): Following the algorithm presented in Section 4 we defined the six experiments showed in Table 5, starting out with experiment 1 in which we selected the initial values for the and then improved them by using different subsets of these integers as defined by experiments 2–6. The best strategy was found in experiment 2, for which a combination of full containers and different sizes of LFC is sent alternately. In this way we exploit the
Some considerations using the EOQ method
Assuming a deterministic demand and a constant set-up cost, we perform some calculations using the EOQ procedure to investigate the behavior of the system when using a full container under minimum order quantities. Accordingly, consider the following data from the case study: Total average demand of the system: items/week Set-up cost for an order: euros (for a full container) Annual holding rate at the DC: Unit cost: euros
Conclusions
For a real supply chain we showed that we can achieve coordination of orders while adhering minimum order quantities, and at the same time exploit the economies of scale of a transportation system with a non-linear cost structure. Particularly, we found that a policy performs better than focusing directly on a specific container size, because inventory costs in the studied case are higher than transportation costs. The method does however, have a varying order size, which is not
References (11)
- et al.
An efficient optimal solution method for the joint replenishment problem
European Journal of Operational Research
(1997) - et al.
Periodic versus `can-order' policies for coordinated multi-item inventory systems
Management Science
(1988) On a basis class of multi-items inventory problems
Management Science
(1964)- et al.
Periodic review and joint replenishment in stochastic demand environments
IIE Transactions
(1998) Determination of optimum packaging frequency of items jointly replenished
Management Science
(1974)