FUNDAMENTAL ASPECTS OF INVENTORY MANAGEMENT IN THE SUPPLY CHAIN

This text is the first of two articles dedicated to the analysis of inventory management, from a perspective different from the traditional view, focused exclusively on a single company in the supply chain.

Initially, two tools are reviewed that, even though developed for a single company, are still extremely useful for implementing inventory policies in the supply chain: the modeling of material consumption through sawtooth graphs and the analysis of trade-offs in costs between inventories and other logistical functions, such as transport and storage.

Then, the main factors that are motivating supply chains to continuously reduce their inventory levels are presented: a greater diversity in the number of products and markets served, the high opportunity cost of capital and the growing managerial focus on control and reduction in the group of accounts belonging to Net Working Capital. In addition, attention is called to the fact that reducing inventory levels, without a preliminary analysis of the degree of efficiency of transport, storage and order processing, can generate an increase in the total logistical cost of the operation.

Finally, it discusses how three basic transformations in the supply chain have allowed companies to operate with increasingly lower inventory levels: forming partnerships, the emergence of logistics operators and the adoption of information technologies to capture and exchange data. Specifically with regard to the integration between retailers and manufacturers, a concrete example illustrates how the exchange of sales data can contribute to a substantial reduction in safety stock levels.

In the next article, the four main decisions necessary to formalize an inventory policy in the supply chain will be discussed: (1) level of inventory centralization, (2) resupply postponement strategies versus reorder point calculation, (3) policies to correct safety stocks due to shortage and excess costs and (4) adoption of the economic purchase lot versus just-in-time resupply.

BASIC TOOLS FOR INVENTORY MANAGEMENT

Regardless of the existing motivations for reducing inventory levels, the dynamics of material consumption in a given link in the supply chain can be represented by sawtooth graphs, as illustrated in Figure 1.

In an ideal world, with no uncertainty, the average consumption rate (D) of products is completely predictable from day to day. In this way, it is possible to know exactly when the stock level will reach zero, that is, the moment of replenishment, that is, when we should schedule the arrival of new products. Just break it down into the resupply lead-time (LT), from the moment of refueling, to determine the moment to ask for resupply. The Order Point (PP) is simply the ordering time converted to the stock level across the product between the average consumption rate per replenishment lead time (D*LT).

On the other hand, in the real world (with uncertainty) the rate of consumption of products is not entirely predictable, and may vary considerably around average consumption. In addition, the resupply lead time may also vary, causing delays in delivery. To protect themselves from these unexpected effects, companies size safety stocks, based on an acceptable probability of product shortages in stock.

Another element of the dynamics of inventory management that remains unchanged, regardless of the drivers behind the reduction in inventory levels, is the cost trade-off between inventory and other logistics functions.

Let's imagine, for example, a distribution center (DC) that has an average annual demand of 300 units for a given product and consider two alternative policies, as shown in figure 2. In the first policy, 6 shipments of 50 units are sent throughout the year . In the second policy, the 300 units are shipped at once. What are the advantages and disadvantages of each policy?

In the first policy, the company incurs a lower opportunity cost of maintaining inventories, as it operates with an average level of only 25 units. Transport expenses, however, are higher: the freight bill is higher not only due to a greater number of trips, but also proportionally more is spent on transport per ton-kilometer due to the lack of scale in the operation with fractional shipments .

On the other hand, in the second policy, the opportunity costs of maintaining inventories are higher (an average level of 150 units of product is maintained in stock) but, on the other hand, not only is the freight account lower because there is only one trip, but also the unit cost of freight is also lower, due to possible economies of scale arising from sending consolidated shipments.

Equilibrium, or the ideal resupply policy for this DC, is reached when we balance the opportunity cost of holding inventory with the unit cost, in this particular example, of transporting to the DC.

As we can see in Figure 3, the objective of supply chains with regard to inventory management should be to determine the most appropriate resupply batch size for their level of efficiency in the material handling process. In this example, the balance is not located either to the left of the graph, as policy 1 suggests, nor to the right, as policy 2 suggests.

In practice, it is very difficult for companies to properly assess where their current inventory policy sits on the graph. However, it is possible, through the generation of scenarios and incremental analyzes of inventory and material handling costs, to determine whether an alternative operation will entail a lower total logistics cost.

In this way, it is possible to avoid the perception that isolated reductions in stock levels, without taking into account impacts on other logistical functions, such as transport, storage and order processing, allow for a lower total cost resupply operation. In fact, as illustrated by Figure 3, companies should seek to minimize the total logistical cost of inventory, transportation and order processing due to a certain product availability desired by the end customer.

WHY AND HOW TO REDUCE INVENTORY LEVELS

More and more companies are seeking to guarantee product availability to the end customer with the lowest possible stock level. There are several factors that have been determining this type of policy, as described below.

• The increasing diversity in the number of products, making the continuous management of stock levels, reorder points and safety stocks more complex and laborious. It is worth exemplifying the case of Brazilian breweries that, in 1985, offered a single flavor (pilsen) in a single package (the 600 ml bottle) and currently offer several flavors (bock, draft, light, etc.) in other types of packaging (can, long neck etc).
• The high opportunity cost of capital, a reflection of prohibitive interest rates in Brazil, has made the ownership and maintenance of inventories increasingly costly.
• The managerial focus on reducing Net Working Capital, one of the measures adopted by several companies that wish to maximize their Value Added by the Market indicators.

On the other hand, several factors have influenced inventory management in the supply chain in order to increase the efficiency with which companies operate material handling processes (transport, storage and order processing) as illustrated in figure 4. Increase the efficiency of these processes simply means shifting the unit cost curve of material handling downwards, allowing operations with smaller resupply batch sizes, without, however, affecting the product availability desired by end customers or incurring cost increases total logistics.

We highlight three factors that have contributed substantially to the reduction of unit costs of material handling, whether in transport, storage or order processing activities:

• formation of partnerships between companies in the supply chain;• emergence of logistics operators;• adoption of new information technologies for capturing and exchanging data between companies.

Below we describe the impact of each of these factors.

FORMATION OF PARTNERSHIPS BETWEEN COMPANIES IN THE SUPPLY CHAIN

The formation of partnerships between companies in the supply chain, a phenomenon initially verified between automakers and suppliers in the Japanese automobile industry, has allowed reductions in purchasing costs through the elimination of several activities that do not add value. As the final objective is the just-in-time resupply of parts and materials, tasks such as quality control upon receipt, bids and price quotations were practically eliminated in the commercial relationship between companies, through the establishment of partnerships.

EMERGENCE OF LOGISTICS OPERATORS

The emergence of companies such as TNT, FedEx, Ryder and several others, which have been assuming an increasing importance in the supply chain, offers the possibility of reducing the unit costs of moving products between companies. This is because these companies have know-how, economies of scale and focus on various logistical operations related to material handling and transportation. For example, it is quite common for a logistics operator to consolidate fractional shipments from several companies in order to complete a trailer, making it possible to dilute the fixed costs of this transport by a larger apportionment base.

ADOPTION OF NEW INFORMATION TECHNOLOGIES (ITS) FOR DATA EXCHANGE

Finally, the adoption of new ITs such as bar codes, EDI, POS automation, etc., brought several benefits inherent to the capture and availability of information with a greater degree of precision and punctuality. We draw attention in particular to the elimination of errors and rework in order processing, a fact that substantially reduces the costs associated with this activity, and to the reduction of uncertainty in relation to future demand, by sharing the series of sales for the final customer by all companies in the chain.

HOW CAN IT ADOPTION CONTRIBUTE TO THE REDUCTION OF SAFETY STOCKS?

We noticed that, on the one hand, the ITs made it possible to reduce order processing costs, by eliminating errors resulting from human interference in placing orders, enabling a resupply operation with smaller lot sizes. On the other hand, the possibility for companies in the chain to exchange information has contributed to reducing the lack of visibility in the supply chain about the real demand of final consumers, a factor that directly influences the formation of safety stocks.

Figure 5 illustrates a real case between a soft drink manufacturer and a retail chain in the city of São Paulo verified over the months of February, March and April 1996. This figure compares the real sales profile of the soft drink over a 12-week period X in the retail network for its final consumers (blue line) with the profile of withdrawals of this same product by the retail network at the manufacturer (purple line).

We can clearly see from this figure that, as we move away from the final consumer in the supply chain, fluctuations in existing stock levels in companies (in this case between the retail chain and the manufacturer) contribute to substantially shifting the amount of replacement or be bought (taken from the retailer at the manufacturer) of the true demand profile (in this case approximated by the sales of the retail chain).

The availability via IT of sales collected in real time in retail can allow a leaner planning of several manufacturer's operations. This is because, when we analyze the two series of sales over the twelve weeks, the average demand for soft drink X in the retail chain is 143 boxes per week (with an associated standard deviation of 53 boxes per week), while the average demand at the manufacturer is 221 boxes per week (with a standard deviation of 271 boxes per week), as shown in Figure 6.

It should be noted that, although in the long term the average demand is approximately equal for the entire supply chain, in the short term fluctuations in stock levels between companies mean that in the same week the retail chain sells a different quantity than that invoiced by the manufacturer. This effect entails substantial impacts on the planning of several operations in the short term, such as purchasing, production, distribution and sizing of safety stocks.

With regard specifically to safety stock levels and assuming normally distributed demand in the two links of this chain, the safety stock level (ES) that guarantees a 98% probability of not having a product shortage is approximately equal to two standard deviations of demand. As previously explained, because the average demand varies in the short term depending on the company's position in the supply chain, the manufacturer must dimension a safety stock of refrigerant X a little more than 5 times greater than what is necessary for the retailer.

CONCLUSION

This article explored the main drivers for reducing inventory levels and the pitfalls present in the traditional view, as well as the transformations in the supply chain that are allowing companies to operate with ever smaller resupply lot sizes.

Specifically, it was discussed that reducing intercompany resupply batch sizes should not simply be an objective isolated from other logistics functions, but rather a direct consequence of increasing efficiency in transportation, warehousing, order processing, etc. activities. The significant impact that the exchange of information between companies can have on the dimensioning of safety stocks was also evaluated.

In the next article, the main aspects and issues relevant to the formalization of inventory policies for supply chains will be discussed.