Supply Chain Frontiers #51 :: A Model for Transporting Goods in Brazil

Supply Chain Frontiers issue #51

When shipping product into a country as large and complex as Brazil, the choice of transportation routes has a critical impact on supply chain costs. Four GCLOG students* developed a decision support system that enables a leading luxury goods company to determine the optimal routes for product imported into the Brazilian market.
  Luxury Goods Corporation (LGC) markets its products under 24 brand names in three channels: wholesale, retail, and licensing. The items fall into four classes of product: apparel, accessories ranging from eyewear to luggage, home products such as bedding, and fragrances. Some 800 factories supply goods to around 10,000 delivery points worldwide, requiring the organization to make more than 100,000 deliveries monthly.   In general, these products are shipped from factories to regional distribution centers, where they are consolidated, packed, and transported with other products to a destination country and distributed to retail outlets.
  In Brazil, the company decided to switch from a wholesale model – i.e. other retailers buy and sell its merchandise – to a retail operation where the enterprise sells its own products direct to consumers.   The strategy will give LGC a competitive advantage in the region – provided it can develop a supply chain that minimizes costs while maintaining high service levels. Standard operational goals, it could be argued, but achieving these twin objectives involves overcoming some unique challenges.
  First, the complexity of LGC’s global distribution network makes the selection of transportation routes far from trivial. In addition, Brazil’s byzantine import system includes numerous classes of duties and taxes that are levied at the national and local levels. The way these payments are structured depends, to a large extent, on the routing of goods within the country. Since taxes are paid on the value of the product plus freight and other costs, high-value goods such as luxury items are particularly sensitive to these logistics factors.
  LGC’s dilemma was how to create a supply chain model that accounts for these variables, and identifies the optimal transportation routes for the company’s new distribution strategy in Brazil.   It is a tough challenge. The company’s supply chains are built around four general echelons: production in the factories, consolidation and primary distribution in regional DCs, consolidation in nodes within destination countries, and retail outlets. There are a number of operational components in the mix as well such as factory licensees, global and regional transportation networks, customs requirements, product returns, and liquidation.
  Defining the best combination of freight transportation routes from suppliers to final customers in terms of logistics costs, duties and taxes, and time constraints involves the evaluation of numerous alternatives and tradeoffs.   The students selected a model based on a Decision Support System (DSS) to do the job.   DSSs excel at helping managers to synthesize information from a wide range of sources and to make decisions. The tool often uses data inputs from enterprise-wide information systems to address specific business problems and has the capacity to organize and disseminate knowledge.
  Crucially, DSSs can be used to carry out tradeoff studies. This is important since LGC has to evaluate many routing options, and the main objective of a tradeoff study is to generate a single, final score for each of a number of competing alternatives. Also, managers can challenge different model configurations and test the robustness of each decision.   The model performed two key tasks. First, it helped decision makers to analyze multiple transportation options for the Brazilian market and to compare the pros and cons of each one. Second, it provided a foundation for analyzing LGC’s entire supply network.
  “Our main goal in the capstone project was to provide a tool so LGC could plan their supply network in Brazil considering their global network, although we conducted some scenario analysis to show how the DSS could be used to optimize global logistics networks,” explains Luiz de Andrade, one of the students involved in the project who is now Director at TEVEC Systems, a logistics technology provider. The scenario tested could be considered as a baseline scenario in the sense that it didn’t have accurate data such as cost data and demand forecasts, he notes.   An example of the results obtained from the DSS is an optimal network that included a mixed transportation strategy. This involved shipments from a cargo consolidation center in Panama to a port in southwestern Brazil and an airport in the central portion of the country. Prior to the project, Panama as a consolidation center was an obvious choice due to its proximity to Brazil. How much product should be carried in each transportation mode wasn’t obvious, however.   Another result from the DSS model clarified the relation between tax cost and logistic cost of product at the final destination. The model showed that tax cost is much more significant than logistic costs.
  “We also compared the global network cost as a function of desired service level. We could actually see how the network costs vary as you relax service level requirements,” says de Andrade.
  Since the DSS model was broadly defined for planning global networks, it can easily be applied to other industries. For example, the model can be used to define how much product should flow in each of the routes in a global network, where the entrance points should be in a specific country, and how the goods are to be distributed.   However, as de Andrade pointed out, “one of the key issues faced by global sourcing decisions is the tax structure to be considered in the decision model, and since we were focusing on the Brazilian market, we customized the model to cope with this specific country tax structure.”   In other decision scenarios it might be necessary to adjust the DSS to incorporate other specifics tax regulations. But this is a relatively straightforward task.

*The research was carried out by Graduate Certificate in Logistics & Supply Chain Management (GCLOG) students Luiz de Andrade, Javier Gotschlich, Edson Trevisan, and Wildor Vargas, for their capstone project titled Developing A Decision-Support System for Global Sourcing: A Case Study for Importing in the Brazilian Market, under the supervision of Dr. Roberto Perez-Franco, Director of the GCLOG Program. For more information on the research, contact Luiz de Andrade (luiz.andrade@tevec.com.br).