FORMALIZING AN INVENTORY POLICY FOR THE SUPPLY CHAIN

This text is the second of two articles dedicated to the analysis of inventory management, from a perspective different from the traditional view. in the first article the main motivations for reducing inventory levels were seen, the pitfalls to be avoided according to the traditional view and which transformations in the supply chain are allowing companies to operate with smaller resupply lot sizes.

In this article, the four fundamental decisions for the formalization of an inventory policy in companies will be addressed, namely:

• Where to locate inventories in the supply chain? This decision refers to their centralization or decentralization, vis-à-vis the analysis of some relevant dimensions such as turnover, added value and required service levels.
• When to order replenishment? This decision seeks to determine whether or not the company will follow the methodology suggested by the ordering point.
• How much to keep in safety stocks? When calculating the safety stock as a function of variability in demand and lead-time for resupply, companies must determine whether it is possible to reduce it without prejudice to the levels of product availability required by the market.
• How much to ask? Finally, we seek to determine whether it is more appropriate for a company to adopt the economic batch methodology or implement a just-in-time replenishment regime.

WHERE TO LOCATE STOCKS IN THE SUPPLY CHAIN?

This decision seeks to determine whether stocks should be centralized (in a single distribution center/warehouse) or decentralized (in more than one distribution center/warehouse) in the supply chain. In addition, depending on the specific characteristics of each business, the location of inventories may in some cases involve consignment decisions or the decision not to keep a certain material in stock.

Basically, as illustrated in Figure 1, there are four dimensions that influence the location of inventories in the supply chain: material turnover, response lead-time, level of availability required by the markets and added value of the material.

Let us now individually analyze the impact of each of these dimensions on the decision to locate materials in the supply chain.

• Material turnover: the greater the tendency towards decentralization across different warehouses or distribution centers, as the risks associated with material perishability and obsolescence are lower. In addition, we should note that materials with high turnover absorb a smaller share of fixed storage costs, compared to materials with lower turnover.
• Response lead-time: the longer the response time from placing the order to serving the final customer, the greater the tendency to decentralize inventories, with a view to providing faster service. Companies must assess, in incremental terms, whether the reduction in opportunity costs of holding inventory in transit more than compensates for opening a new warehouse.
• Level of availability required by markets: the higher the level of service, the greater the tendency to position materials close to the end customer. In this case, the same incremental analysis described for response lead-time must be performed.
• Added value: finally, the higher the value, the greater the tendency towards centralization, contrary to what was exposed for the three previous dimensions. This is because materials with high added value imply high opportunity costs for inventories, which can become prohibitive when they are decentralized. Real cases of several companies indicate that decentralization leads to a significant increase in the levels of safety stock, proportional to the square root of the ratio between the final number and the initial number of storage points. For example, when moving from two to three warehouses, a company is expected to experience, in the medium term, an increase in safety stock levels of around 22,5% (obtained from the expression (v(3/2) – 1)*100% ).

When we jointly analyze the effect of these four dimensions, the situation becomes extremely complex due to the different possible iterations. Thus, without claiming to be exhaustive in terms of product categories, or to accurately address all possible decisions regarding the location of stocks in the supply chain, we draw attention to two particular cases: consignment of materials and non-maintenance. of a material in stock.

Favorable conditions for the consignment of inventories, for example, from a supplier of raw materials to its industrial customer, arise when:

• the material in question has a high added value, significantly affecting, from the perspective of the industrial customer, the opportunity cost of keeping it in stock;
• it is an extremely critical material for the industrial customer, that is, it has a high requirement regarding its immediate availability, and must be located close to the production process;
• the material has a high turnover, allowing the supplier to maintain or increase its return on investment even if there is a reduction in the contribution margins per unit of product due to the lengthening of the cash cycles and the period in which it remains the owner of the inventories.

Some materials that would fit in this case, for example, would be some parts and components used by the automobile industry. On the other hand, favorable conditions for not keeping a certain material in stock in the supply chain occur when:

• the material has high added value;
• It is a low-turn material. This characteristic not only makes the option for consignment difficult, but also increases the risks associated with obsolescence and perishability, and the share of fixed storage costs to be absorbed by the material;
• the material presents little requirement regarding its immediate availability.

Some products that would fit in this case, for example, would be capital goods and some more expensive hospital equipment, such as computerized tomographs. Usually, the company responsible for the commercial representation of this type of product uses its previous customers as showrooms to serve new customers in the decision-making phase for the purchase.

WHEN TO ASK FOR A REFUND?

According to the reorder point methodology, as seen in the first text, the replenishment request (time to order) depends directly on the average consumption of materials and the response lead-time, as illustrated in Figure 2. Depending on each company's holding and transportation cost structure, it may be economically feasible to order replenishment before or after the date indicated by the order point.

There are situations in which it may be more interesting to postpone the replenishment request until the last possible moment before the replenishment moment, one of which is when there is the possibility of contracting express or premium transport. For materials with high added value, low unit weight and high risk of obsolescence or perishability, companies must assess whether the increase in spending on contracting express transport is more than offset by the reduction in the opportunity cost of keeping inventory in transit .

For example, the transport of electronic microprocessors from Asia to the US and the transport of tropical fruits from Central America to Japan are done by air rather than by sea. It was noticed in these cases that the increase in freight costs would be more than offset by the reduction in the risks of obsolescence and perishability, respectively, associated with a shorter transport lead-time.

On the other hand, there are situations in which it is interesting to postpone the resupply request with a view to consolidating the shipment. For materials with low added value, considerable unit weight, and low risk of obsolescence/perishability, companies should assess whether the increased opportunity cost of holding additional safety stocks is more than offset by reductions in transportation expense.

For example, some suppliers of electrical and welding materials located in Rio de Janeiro and São Paulo schedule weekly vehicle departures to serve their branches located in the Northeast. As the daily demand for cargo to the Northeast is extremely low, these companies can wait up to five business days after placing the order for consolidation of the shipment. By adopting this transport policy, the safety stocks in the Northeastern branches must be adequate to cover a longer response lead-time. We must remember that the purpose of safety stocks is to protect the company from unexpected increases in demand once the reorder point is reached and replenishment is requested.

HOW MUCH TO KEEP IN SAFETY STOCKS?
Generally, safety stocks are determined by assuming that demand variability follows a Normal probability distribution. Considering this premise implies a decreasing return on safety stocks for the purpose of product availability, as illustrated in Graph 1.

According to this graph, for example, a safety stock equal to one standard deviation of demand guarantees a little less than 85% chances of not having a lack of product (stock-out). Two parries guarantee just under 98% chance and three parries just under 99,9%. Ultimately, safety stocks will never guarantee a 100% chance that there will be no product shortages.

The practical implications of diminishing returns from safety stocks for companies operating in highly competitive markets are extremely significant, for two main reasons.

• First, the greater the level of competition in a given market, the greater the errors associated with the demand forecasting process. Larger forecast errors, as they indicate a standard deviation of greater magnitude, imply greater safety stocks.
• Second, competitive markets often demand greater product availability. The greater the required availability, the greater the number of standard deviations of demand used in determining safety stocks.

Based on these two reasons, companies must consider not only demand variability and desired product availability, but also the costs associated with excess and shortage of products in stock, as illustrated in Figure 3. level of risk associated with the maintenance of safety stocks, that is, what are the chances of the company investing in a level of safety stock to guarantee a certain product availability and the actual demand to be below the expected level.

A preliminary analysis of this level of risk can be done quickly through the equation expressed in Figure 4, where Cf represents the cost of shortages and Ce the cost of excess. The cost of shortages encompasses not only the contribution margin (sales price minus the acquisition cost) lost when there is no product available in stock, but also any damage to the company's image that can be made tangible in some way. The cost of excess involves not only the opportunity cost of maintaining safety stocks, but also possible losses due to obsolescence or perishability of the product.

For products with a high added value, with a high rate of obsolescence or a high degree of perishability, the risk associated with maintaining safety stocks is considerable (approximately 1). In this case, the safety stocks must be undersized. On the other hand, products that guarantee high contribution margins for the supply chain, or that a momentary unavailability substantially affects customer loyalty, should have safety stocks conservatively sized. We must remember that the type of adjustment made will depend on each case, and there is no single solution to be followed.

HOW MUCH TO ASK?

In the era of Just in Time (JIT) management of inventory levels, the concept of the Economic Purchase Lot (LEC) seems to be a bit outdated. Is the mathematical formula conceptually wrong or has it simply fallen into disuse?

The LEC formula calculates the optimal lot size from the trade-off between the costs of holding inventory and the cost of processing the order (transport, credit assessment, equipment set-up, etc.). The traditional wisdom of lean sourcing, however, tells us that unit lot size should be the primary objective to pursue. We clearly see that these two approaches are antagonistic. But can they be used together?

The reconciliation of these two approaches comes through recognizing that the LEC formula is valid, but it perceives the problem less dynamically than advocates of lean resupply. For example, the LEC assumes that order processing costs are given by the system and therefore calculates the batch size that will dilute this cost without, however, incurring excessive inventory holding costs. The JIT practice works the other way around: given that the ideal batch size is one, the company must strive to reduce order processing costs. As order processing costs decrease, the optimal lot size calculated by the LEC shape also decreases.

Proponents of JIT and ECR in retail tend to place greater weight on the cost of carrying inventory. In his perspective, inventories are a resource used to hide inefficiencies in production and distribution systems. Using the famous lake analogy, JIT advocates argue that as the water level (storage) drops, rocks appear (problems or deficiencies in the system). From there it is possible to direct efforts to eliminate these problems allowing the boat (flow of products and materials) to sail more smoothly. ECR advocates, on the other hand, acknowledge that the cost of carrying inventory has often been miscalculated. From the retailer's perspective, if on the one hand the cost of lost sales is extremely high, on the other hand a large volume of products in stock is liquidated through promotions to stimulate demand. The solution involves a smaller, better-sized inventory level supported by an efficient replenishment system.

As such, advocates of ECR ​​and JIT should not reject the LEC formula. A lot size that balances the costs of processing orders with the costs of actually carrying inventory leads to the lowest total cost of operation. Leading companies have realized not only the importance of reducing inventories, but also the need to continually improve order processing to ensure lean sourcing is the lowest total cost operation.

CONCLUSION

This article commented on the four main decisions needed to formalize an inventory policy for the supply chain: where to locate, when to order, how much to keep in safety stocks and when to order. Specifically for each of these decisions, we highlight the main points below.

• The location, or the level of centralization of inventories in the supply chain, is a process that fundamentally depends on the iteration of several dimensions characteristic of each material: turnover, added value, availability and required response lead-time.
• The decision on how much to order also depends on incremental analyzes of inventory holding and transportation costs, especially when postponing or resupply consolidation strategies are being evaluated.
• The dimensioning of safety stocks will depend not only on the product availability demanded by the markets and the variability of demand, but also on an analysis related to the costs of shortages and excess.
• With regard to how much to order, the LEC methodology and JIT resupply are not mutually exclusive approaches and can be used together to evaluate and continually reduce lot sizes.

BIBLIOGRAPHY

• BALLOU, RH, 1992, Business Logistics Management, 4th ed, Prentice Hall.• BOWERSOX, DJ, CLOSS, DJ, 1996, Logistical Management – ​​The Integrated Supply Chain Process, 1st ed, McGraw-Hill.• WANKE, P., FLEURY , PF, 1999, “The Lean Resupply Paradigm: Pitfall in Material Flow Management between Links in the Supply Chain”, Anais do XXIII ENANPAD.• WATERS, CDJ, 1992, Inventory Control and Management, 1 ed, New York, Wiley&Sons.