LOGISTICS POSITIONING STRATEGY: CONCEPTS, IMPLICATIONS AND ANALYSIS OF THE BRAZILIAN REALITY

The purpose of this article is to present a concept that is allowing several companies to compete efficiently in the supply chain, through the integration of production and logistics processes, in order to minimize the total cost for a given level of service - the logistics positioning. Currently, there are several motivators that lead to a growing search for the integration of production and logistics operations within the supply chain.

1. Pressure to reduce inventory levels due to high opportunity costs of holding inventories.
2. Pressure to streamline customer service, reducing delivery time and increasing availability, in view of the growing customer demand in recent years.
3. Pressure to mass customize, that is, to offer a wide variety of customers products exclusively designed to meet their specific needs, still as a reflection of the growing demands in recent years.

A logistics positioning strategy is made up of five decision categories that must be articulated and consistent with each other over time, in order to allow a company to achieve its cost and service level objectives.

1. Coordination of the product flow: should the product flow be pulled, that is, activated by the link that is closest to the final consumer or pushed, that is, coordinated by the link that is closest to the initial supplier?
2. Production policy: should a company produce to stock, based on forecasts of future sales, or produce to order, always meeting actual demand only at the time the customer places the order?
3. Inventory Allocation: Should inventories be centralized in a single location, or decentralized across multiple facilities?
4. Transport policy: should a company operate with slower and cheaper modes of transport, such as rail and sea, or faster and more expensive, such as road and air? Should you pursue consolidation shipping or express delivery?
5. Network sizing: how many facilities should a company have, where are they located, what products and markets should each facility serve?

As indicated in the previous items, there are, therefore, different possibilities for integrating production and logistics systems, that is, for logistical positioning, within the scope of the supply chain.

PRODUCT FLOW COORDINATION

The product flow coordination decision is basic for the logistics positioning strategy, strongly affecting all other decisions, especially the production policy. As seen, coordination is related to which stage of the chain will trigger the flow:

• Closer to the end customer: PULL
• Closest to initial vendor: PUSH

Usually, the decision to pull or push depends on the joint analysis of two factors: demand visibility and resupply and distribution cycle times. Demand visibility refers to the fact that a supply chain company has access to consumer/end customer demand information in real time. It should not be confused with the predictability of demand, or the degree of accuracy/accuracy in the sales forecasting process, which depends on several factors: quality of historical information, forecasting method, number of competitors, substitute products, etc. Supply and distribution cycle times refer to the average time it takes to receive the input that takes the longest to produce and deliver the product to the customer. They allow us to answer the following question: “if there were no more stocks in the supply chain, how long would it take for the customer to have the product and hands?”.

Demand visibility allows product flows to be pulled, that is, coordinated by the stage closest to the final consumer, based on real-time sales information captured by information technology.

Figure 1 illustrates two situations for a company that supplies consumer goods, formed by a factory and a distribution center. In the first, only the retailer has real-time access to sales information, so the company has to push the flow of products based on sales forecasts. In the second, the company has access to sales information in real time through agreements with retailers and the adoption of information technology, so the flow of products could be pulled, that is, driven by real demand. Note that in the first situation, the coordination base is inside the company (these are sales forecasts) and in the second situation, the coordination base is outside the company (actual sales).

A clear example of a supply chain that operates entirely with product flows pulled is the Direct Model that Dell Computers has organized with its suppliers and customers. In this model, all of Dell's commercial, manufacturing and distribution operations were centralized in a single location. The customer places his orders through a telemarketing center or the Internet, signaling the need to assemble a new computer. Dell suppliers such as Sony and Intel continually access the Dell Intranet to track order status and ship needed parts and components. With this model, Dell's inventory turns over 180 times a year, while the competition turns over only 6 times a year. In this way, with the Direct Model, Dell articulated the following logistical positioning decisions: pull production and distribution, production against order, centralized inventories, express transport to the end customer.

On the other hand, supply and distribution cycle times make it possible to answer whether the flow of products can be pushed or pulled, when compared with the response time required by the end customer, as illustrated in figure 2.

If the response time required by the end customer is longer than the duration of the supply/distribution cycle, the flow can be triggered by the stage closest to the end customer (pulled). This is the case for Dell customers who are willing to wait 3 to 7 days for a new computer that has a maximum two-day supply-production-distribution cycle (the basis of coordination is outside the company). If the response time required by the end customer is less than the duration of the supply/distribution cycle, the flow will be coordinated by the stage closest to the initial supplier (pushed), and guided by sales forecasts that signal the formation of inventories (the basis for coordination is outside the company).

The ECR program, adopted by several retailers and manufacturers of consumer goods in Brazil, is an example where part of the product flow is pulled, that is, coordinated by the retailer, and part of the flow is pushed, that is, coordinated by the manufacturer. This is due to the difference between the response time required by the retailer and the supply/distribution cycle time of the manufacturer. In the ECR, the automatic replenishment of retail shelves must take place within a time window of up to 24 hours, guided by real-time sales information collected at the POS. On the other hand, some manufacturers such as breweries experience supply and production cycle times well in excess of 24 hours. In these circumstances, the flow of production and purchases must be pushed based on forecasts of future sales so that there is no shortage of stock.

Research on logistical positioning decisions, carried out by the Center for Studies in Logistics, with almost 30 Brazilian companies throughout 2001, whose combined revenues reach close to USD 20 billion, provides further details on how the coordination of the flow of products is organized in different sectors of the economy. Each company surveyed reported its decisions for two types of product, a Class A product in revenue and another Class E product, also in revenue. With this, the research totaled almost 60 cases to be analyzed among the food, electronics, pharmaceutical, chemical and petrochemical, automotive and technology and computing sectors. The sectors where pushed flows predominate, driven by sales forecasts and coordinated by the stage closest to the initial supplier, are the food, pharmaceutical, electro-electronic and chemical and petrochemical industries. The sectors where pull flows predominate, driven by real demand and coordinated by the stage closest to the final consumer, are automotive and technology and computing. Graph 1 presents these results.

Graph 2 indicates that, as the percentage of cases with visibility of demand across the various sectors increases, the percentage of cases with product flows driven by real demand increases. For example, the automotive sector showed demand visibility in almost 70% of its surveyed cases, reflecting pull flows in almost 70% of them as well. On the other hand, the pharmaceutical sector showed demand visibility in only 20% of the cases surveyed, not reflecting pulled flows in any of the cases, only pushed flows. Finally, the food sector showed demand visibility in less than 10% of the cases surveyed, reflecting product flows pulled in 30%, which indirectly indicates effects of ECR ​​programs on product flow coordination.

In Graph 3, it can be seen that as the response time (in days) of the supply and distribution cycles increases throughout the different sectors of the economy, the percentage of cases with pushed flows increases. For example, the chemical and petrochemical sector has an average response time for the supply and distribution cycle of approximately 140 days, reflecting pushed flows in 75% of the cases surveyed. The automotive sector, on the other hand, presented an average response time of just under 40 days, reflecting pushed flows in almost 35% of the cases surveyed.

These analyzes and conclusions lead us to a 2×2 matrix for decision-making on product flow coordination. According to this matrix, represented by figure 3:

• If demand visibility is none and response time is long, product flow must be pushed, driven by sales forecasts, with the basis for coordination closer to the initial supplier.
• If the demand visibility is total and the response time is short, the flow of products can be pulled, being directed by the real demand, being the basis for coordination closer to the end customer. – DELL
• If the demand visibility is total and the response time is long, or the demand visibility is none and the response time is short, push-pull hybrid systems are organized in the supply chain, where in part of the chain the flow is pulled, usually close to the final consumer, and elsewhere the flow is pushed, usually close to the initial supplier. This is the ECR example seen earlier.

The decision to coordinate the flow of products is the main decision of a logistics positioning strategy, affecting all other decisions, especially the production policy, as illustrated in figure 4. A decision to push the flow of products, taken based on on demand visibility and supply/distribution cycle time, it always implies the use of sales forecasts in anticipation of future demand, as a basis for planning. Using sales forecasts often means producing, distributing, storing and transporting quantities in excess of actual demand at a given time. In this way, pushing the flow of products will imply the decentralization of inventories by many facilities, associated with a policy of production for inventory and the consolidation of transport through the use of cheaper and slower modes. On the other hand, a decision to increase the flow of products may imply both the use of sales forecasts and the use of real demand at the stage closest to the final consumer. In the case of targeting real demand, pushing the flow of products will imply the physical centralization of inventories, on-demand production and the use of premium transport by contracting more expensive modes.

PRODUCTION POLICY

Another decision of the logistical positioning strategy is the definition of the production policy, whether it will be to order or for stock. Producing to order means postponing the purchase and transformation of inputs into a finished product as much as possible, the same only being done when the final customer places the order. Making to stock means buying and turning inputs into finished goods now and in anticipation of future demand, based on sales forecasts. In order to define the most adequate production policy, not only the decision on the coordination of the product flow must be observed, but also other characteristics of the product and the process.

Among the characteristics of the product, the total added cost stands out, which can be measured in cost accounting as the COGS (cost of goods sold), the degree of obsolescence (reflection of the product's life cycle) and the degree of perishability ( reflects the expiry date of the product). Among the characteristics of the process, the structure of fixed and variable costs stands out, that is, if the production process is more intensive in fixed costs and presents potential for economies of scale and if the process is of continuous flow (eg steelworks, refinery ) or discrete flow or assembly (eg automotive, electronics, etc.). Other factors, such as the profile of added costs in the value chain, indicate how much a certain activity added costs and with what duration in relation to the total production process.

Graph 4 shows how the production policy is organized in different sectors of the Brazilian economy. The sectors where production to stock predominates are: food, electro-electronics, pharmaceuticals, chemicals and petrochemicals. The sectors where production to order predominates are: automotive and technology and computing.

Graph 5, on the other hand, illustrates that as the COGS increases across the different sectors surveyed, the percentage of cases surveyed for on-demand production increases. For example, the automotive and computer technology industries have COGS greater than $10.000,00 reflecting at least 50% of cases surveyed as make-to-order.

Graph 6 illustrates that, as the degree of perishability increases across the various surveyed sectors, the percentage of production-to-stock cases increases. For example, the food and pharmaceutical sectors have a degree of perishability greater than 0,20 (equivalent to a shelf life of less than 5 months), reflecting production for stock in 100% of the cases surveyed. Finally, graph 7 illustrates that, as the degree of obsolescence increases across the various sectors surveyed, the percentage of cases of production against order increases. For example, the technology and computing and automotive sectors have a degree of obsolescence greater than 0,06 (equivalent to a product life cycle of less than 17 months), reflecting at least 50% of production to order.

 

This apparent contradiction between the impact of the degree of obsolescence and the degree of perishability in defining the production policy can be explained by the different nature of products subject to obsolescence and those subject to perishability. Products subject to obsolescence are usually durable consumer goods, in which the purchase decision is more sensitive to product characteristics, such as functionality, performance and design. Differentiation in this context normally stems from high investments in research and development and product design, which can contribute to increasing the magnitude of COGS. Products subject to perishability are normally non-durable consumer goods, whose purchase decision may be comparatively more sensitive to price, implying the need to explore economies of scale to reduce the magnitude of COGS.

ALLOCATION OF STOCKS

Another decision in the logistical positioning strategy concerns the allocation of inventories, whether they will be centralized or decentralized. The centralization of inventories means postponing the transport of products as much as possible, only being moved when the end customer places his order. On the other hand, decentralizing inventories means anticipating their transport/movement through other intermediary facilities at the present time, based on future sales forecasts. In order to decide on the allocation of stocks, characteristics of the product and demand must be observed, in addition to the decision to coordinate the flow of products.

Product characteristics include the density of added costs, that is, the ratio between the product's COGS and its volume or weight, in addition to the degree of obsolescence and perishability. Demand characteristics include turnover, that is, the ratio between the sales level and the average inventory level resulting from a given replenishment policy, and the sales range, the ratio between the maximum and minimum sales levels. Other factors involve contracting premium transport, more expensive and faster modes and exploring economies of scale in transport, by moving large quantities over long distances.

Graph 8 illustrates how stock allocation decisions are organized in different sectors of the Brazilian economy. The sectors where decentralization predominates are: food, automotive, electro-electronic, chemical and petrochemical. The sectors where centralization predominates are pharmaceuticals and electronics. Graph 9 illustrates that the smaller the range of sales across the different sectors of the economy, the greater the percentage of researched cases of decentralization of inventories. For example, the food sector has a sales range slightly greater than 1, reflecting in 100% of the cases surveyed the decentralization of inventories. On the other hand, the pharmaceutical sector has a sales range just below 3, reflecting the decentralization of inventories in 20% of cases. Finally, graph 10 illustrates that as the density of costs increases across the different sectors surveyed, the percentage of cases surveyed where inventories are decentralized decreases. For example, the pharmaceutical sector has a cost density of almost $10.000/kg, reflecting only 20% of the cases surveyed with decentralized inventories. The food sector, on the other hand, has a cost density of just over 1 $/kg, reflecting decentralization of inventories in 100% of the cases surveyed.

 

 

DIMENSIONING OF THE FACILITIES NETWORK

The decision regarding the sizing of the facilities network is strongly associated with the same characteristics that influence the allocation of inventories. However, some considerations must be made regarding the impact of increasing the number of facilities on different components of the logistics system: service level, transportation costs, opportunity costs and storage costs.

CHOICE OF TRANSPORT MODE

Basically, there are two criteria adopted by a shipper when choosing a mode of transport: price/cost and performance. Normally, the performance dimension is measured through the average delivery time, its absolute and percentage variability and the average level of losses and damages that occur in transport. Firms are willing to incur a price or freight cost level commensurate with a given performance. In addition to these elements, the characteristics of the product and demand must be considered when choosing modes.

The characteristics that must be observed when choosing the modal are the density of added costs and the range of sales. The implications of a low density of added costs are related to the choice of modes of transport whose unit cost is compatible, at most equal to the cost density. In this case, slower and cheaper modes of transport such as rail and sea have greater loading capacity, allowing to generate scale to reduce unit costs.

The implications of a high range of sales are related to a mode of transport that provides enough flexibility to keep up with changes in demand, minimizing the chances of wrong decisions such as sending wrong quantities, of the wrong product to the wrong place. In this case, faster and more expensive modes of transport such as air and road have less loading capacity, allowing consolidation in less time and generating the necessary flexibility to keep up with changes in demand. An example of a company that operates entirely with air transport is the North American IBM, in the delivery of spare parts for mainframes to its customers. IBM centralized spare parts inventory in Mechanicsburg. As soon as a need arises from its customers, the part, previously produced and in stock, is immediately sent by air to the place of use. In this way, IBM articulated the following logistical positioning decisions: fractional air transport, production to stock, centralized inventories and pulled flows.

In addition, the research revealed that, when the cost density increases along the different sectors, the percentage of use of air transport increases in the researched cases and the total number of storage points is reduced. A greater number of storage points creates the need to consolidate shipments and explore economies of scale in transportation, which is often only achieved by scheduling shipments.

CONCLUSION

The question that arises at this point is: how to ensure consistency over time between logistical positioning decisions? The success of a strategy, in its strict sense, depends on the degree of integration and mutual reinforcement of the different decisions, actions and course corrections taken within a given time horizon. The element that unites these decisions is the opportunity cost of holding stocks, as each of these five decisions will imply higher or lower levels of cycle, transit and safety stock. The opportunity cost of holding inventory, however, must be calculated based on a view of the flow of products in the supply chain. According to this view, at any point in time, any company in the supply chain will own inventories, whether they are in-process, finished or in-transit. Also from this point of view, what matters is the opportunity cost of the chain as a whole, and not of a particular company.

As opposed to the view of the flow of products in the supply chain, there is the traditional view of the opportunity cost of maintaining inventories, directly related to managing the cash cycle, which is done individually by each company in the chain. Under this approach, one of the main objectives of managing the cash cycle is to lengthen the number of days in accounts payable, that is, the time elapsed between receiving the inventory and paying the inventory in cash to the supplier; and the reduction of the number of days in accounts receivable, that is, the period between the sale to the client/consumer and the receipt of the same. With this, it is possible to reduce the period for conversion into cash, which represents the need to finance the portion of working capital that was tied up in inventory. Although, from the perspective of a single company in the chain, these actions make it possible to reduce the opportunity costs of holding stocks, the rest of the chain, that is, customers and suppliers, will experience an increase in them, implying distortions in the decisions of logistical positioning strategy that would be suitable for the chain as a whole.

BIBLIOGRAPHY

Leeuw, SD and Goor, AR (1999), “The Selection of Distribution Control Techniques”, The International Journal of Logistics Management, Vol.10, No.1, pp. 97-112

Landvater, D. (1997), World Class Production & Inventory Management, 2nd Ed., New York, John Wiley & Sons.

Inman, R. (1999), “Are You Implementing a Pull System by Putting the Cart Before the Horse?”, Production and Inventory Management Journal, Second Quarter, pp.67-71

Christopher, M. (2000), “The Agile Supply Chain – Competing in Volatile Markets”, Industrial Marketing Management, Vol. 29, pp.37-44.

Sharman, G. (1984), “The Rediscovery of Logistics”, Harvard Business Review, Vol. 62, Iss.5, September/October, p. 71.

Amstel, MJ and Amstel, W. (1985), “Economic Trade-offs in Physical Distribution – A Pragmatic Approach”, International Journal of Physical Distribution & Materials Management, Vol.17, No.7, pp.15-54

Carter, JR and Ferrin, BG (1996), “Transportation Costs and Inventory Management: Why Transportation Cost Matter”, Production and Inventory Management Journal, Third Quarter, 58-62.

Silver, EA and Peterson, R. (1985), Decision Systems for Inventory Management and Production Planning, 2nd Ed., Wiley & Sons;

Jayaraman, V. (1998), “Transportation, Facility Location and Inventory Issues in Distribution Network Design”, International Journal of Operations & Production Management, Vol.18, No.5, pp. 471-494.

Closs, DJ and Roath, AS (1998), “An Empirical Comparison of Anticipatory and Response Based Supply Chain Strategies”, International Journal of Logistics Management, pp. 21-34.

Zinn, W. and Bowersox, D. (1988), “Planning Physical Distribution with the Principle of Postponement”, Journal of Business Logistics, Vol.9, No.2, pp.117-136,

Hoek, RI (1998), “Reconfiguring the Supply Chain to Implement Postponed Manufacturing”, International Journal of Logistics Management, Vol.9, No.1, pp.95-110

Pagh, JD, Cooper, MC (1998), “Supply Chain Postponement and Speculation Strategies: How to Choose the Right Strategy”, Journal of Business Logistics, Vol. 19, No. 2, pp.13-33