Image Image Image Image Image Image Image Image Image Image
Scroll to top

Top

Case Study: Supply Chain Improvement

| On 07, Jan 2008

By Getúlio Kazue Akabane, Odair Oliva de Farias and Wellington Barros Bonfim Filho

Abstract

This paper describes a new model of logistics integration based on the utilization of the contradiction matrix and collaborative planning forecast and replenishment (CPFR). The new model has been tested to develop a type of supply chain management system for small companies (SCM-ME).

Before running the phases between planning and operation control, the model starts with an analysis of management process using the contradiction matrix to identify gaps between initiatives and results. CPFR is used to drive and improve system results.

The paper describes one of the pilot cases – a forklift parts distributor from São Paulo where the main functionalities of the SCM system (demand forecast, collaborative integration with suppliers and customers and strategic network design) was implemented with better than expected results.

Keywords

Logistics, supply chain management, collaboration, TRIZ

Introduction

In recent years, many studies and projects have worked to improve supply chain management. Consulting firm Booz Allen Hamilton has shown that supply chain management systems investments often have unsatisfactory results, which can be attributed to a lack of understanding the basic principles of strategic supply chain politics, trade-offs, holistic analysis and a use of trans-functional support. These characteristics are well suited to the use of TRIZ and CPFR.

Applying the TRIZ tools of inventive problem solving in engineering successfully replaces the unsystematic trial-and-error method in the search for solutions in the everyday life of engineers and developers. The majority of organizational management decisions made by executives and managing directors however remain based on intuition and personal experience. Complex contexts are often simplified, alternatives ignored, constraints avoided, risk not evaluated correctly and resources, and knowledge and potentials not utilized for the best problem solving at the right time.1

In the Brazilian forklift market, the competition is different throughout the companies that distribute parts, services and the heavy equipment. There are dealers from brands all over the world, but in this example an independent forklift parts distributor is the main focus.

A growing market of forklift companies challenges the understanding of the real needs of this market and demand problems. Back orders, extended lead times and high competition are logistical situations that managers regularly deal with – with low success. The main causes of deficient results in supply chains are that managers do not understand how to forecast products and services and then cannot create a supply chain to satisfy such a forecast. Managers are unable to create an efficient and/or agile supply chain. An efficient chain coordinates the flow of products and services to stock reduction and maximizes manufacturer and service companies’ efficiencies in the chain. Agile supply chains are built to react fast to a changing market and are able to protect companies from uncertainties.4 A combination of these two models sets the stage for TRIZ.

The best environment and principle characteristics of those supply chains are described in Tables 1 and 2.

Table 1: Best Environment for Supply Chains
FactorEfficient Supply ChainAgile Supply Chain
ForecastPredictable, low level forecast mistakesUnpredictable, high level forecast mistakes
CompetitionLow cost, on-time deliveryShort lead times, volume flexibility
New ProductsInfrequentlyFrequently
Products VarietyLowHigh

Table 2: Characteristics of Supply Chains
FactorEfficient Supply ChainAgile Supply Chain
Operations StrategyStock production, standard services and products, high volumeProduction by order, customized services, high variety
Suppliers SelectionLow prices, on-time deliveryFast delivery time, customization, volume flexibility
InvestmentLow with high level stock turnsNecessary to allow fast delivery time

The potential for new forecast model adjustments using TRIZ increased the possibility of new collaborations, combined innovative solutions and distributed costs throughout the supply chain.

Forklift Distributor Case

The project started as a pilot for an independent forklift parts and accessories distributor client. With a decentralized market and the growing of internal forklift parts production, the distributor’s clients began to leave, decreasing revenue and increasing costs (e.g., the period for stock on the shelves). To make the situation more complex, all the parts sold by our client are imported from Europe, the Unites States and Japan, and then distributed all over the Brazilian states by road and air. This international commerce has an important influence in importation lead times and parts availabilities. A logistics solution that can deal with those contradictions and analyze the forecast planning among the companies is needed.

The first step was to perform a strategic evaluation using a strengths, weaknesses, opportunities, threats (SWOT) matrix combined with a performance indicator structure noting the main functions in the logistics and supply chain as time, resources and information. This step included direct input from the client’s management and employees.With this picture of the company and market completed, TRIZ was brought in to help understand the main contradictions and diagnose the problems at the project’s outset.

The main contradiction is not how to reduce available stock, but how to adapt stock purchasing volume to the changing Brazilian market. How can the distributor have the necessary stock to supply its customers without risking a stock out? How can the company reduce its inventory costs? The trade-off between supply cost and stock-out risk prevention characterizes the inventory sizing dilemma.2 The costs of ordering, transportation, nationalization and third party, warehousing, packing and selling costs can be reduced as a supply cost and shared among the involved companies.

Collaborative Planning, Forecasting and Replenishment (CPFR)

CPFR combines the intelligence of multiple trading partners in the planning and fulfillment of customer demand. CPFR links sales and marketing best practices (e.g., category management) to supply chain planning and execution processes to increase availability while reducing inventory, transportation and logistics costs. The collaborative process among companies is best defined as any administrative practice that adopts a close and organized collaboration, or administered, among independent companies.3 Planning and sharing information throughout the chain will be a challenge.

TRIZ has many secrets – more precisely 17.7 Two of them are used in this example: 1) to identify the “local zone” of the problem, finding where the problem really exists and 2) considering the ideal final result.

Figure 1 represents a CPFR cycle, showing the companies’ responsibilities and workflow integration.

Figure 1: CPFR Model
CPFR Model

With business integration, the companies should find common points for learning and creating a growing network. Because a lot of different themes will be worked together, training will be necessary for all involved in the process.

To understand the connections among supply chain partners, it is important to visualize and separately study each part of the business process for a better global vision. The global result is more than the sum of each part’s evaluation and steps; in this case the companies are organized and united to contribute to the global vision.8

The CPFR reference model is designed to fit many scenarios. Any individual CPFR program must adapt to the particular needs of the trading relationship. CPFR must be done in different modules (e.g., purchase confirmation, purchase management, forecast management). For this case, forecast planning was selected as it is the main focus for product supply.4

The Evaluation

TRIZ was first used to deal with the logistics systems; the matrix pointed out the integration between all the companies in the forklift market that the distributor is included and the problems of cost, stock and purchase lead times of the forklift parts as a main contradiction. Inventive principle 2, take out, says to “separate an interfering part or property from an object or system, or single out the only necessary part (or property) of an object or system.”9 In a supply chain model, the take out principle could mean separating and sharing the activities among the chain’s companies. In technical or logistical systems, this principle leads to an idea of better stock management and using special softwares. Both are incorporated in CPFR concepts.

The distributor need to improve forecast variation flexibility with short lead times, integrate departments and manage the forecast, start a collaborative relationship with clients and suppliers, and equalize the forecast trough strategic marketing and analyze the competitive companies.

Particularly in small companies, there exist purchase systems that have poor process management, which can lead to overstocking and stock out situations. Purchases are made by studying past selling and projecting future consumption, but the past does not always predict the future. An analogy can be done to a typical inventive and creative process. The inventor has an idea and spends much time making adjustments to improve the invention without much effective success in the face ofthousands of available alternatives.5

The analysis of the purchase process of the forklift distributor started with the identification of the main products sold. Identifying the primary products, a relationship to its inventory category and the forklift market was defined. The following is one such example.

Product 1: Master brake cylinder
Category: Brake cables, brake shoes, wheel cylinders and others
Forklift market: Information about how many forklifts exist in Brazilian market that uses that master brake cylinder.
Representation: 2.5 percent of all sold parts and 3 percent of all sold value

Figure 2:Product 1 Forecast

Figure 3:Product 1Category Forecast

Product 2: Electronic accelerator
Category:Potentiometers, contactors and velocity controllers
Forklift market:All electric forklifts that use this electronic accelerator
Representation:1.3 percentof all sold parts and 2.8 percentof all sold value

Figure 4: Product 2 Forecast

Figure 5: Product 2 Category Forecast

After selecting five main products and developing their complete informations, it was possible to begin to manage the exceptions and try to make direct forecast replenishment – planning with the target to adapt the stock levels and raise the services levels.

A starting point for extrapolating the forecast was the theoretical method that is traditionally used when working with predictable, seasonal series and the most used method for forecasting and replenishments, multiple exponential suavization. The following formula represents the combination of the factors of seasonality, tendencies and periods of consumption:

Ft+1 = F1 + α (D1 – F1)
where Ft+1= forecast planning for period t+1, F1 = forecast planning for period 1, α = suavization constant and D = real demand for period 1

In order to narrow the forecast evaluation, this will allow the company the ability to define back-up stock – the minimum quantity of parts in stock, the right moment for order replenishment, the right quantity of parts to be purchased by lots – to reduce costs and configure an investment planning.

Comments and Future Work

The SWOT analyses pointed out purchase processes and adjustments in stock management as strategic problems. The analysis of just one trade-off was able to raise the level of responsiveness of the supply chain. As a non linear system, though, the unpredictability of the market remains challenging. The difficulty of combining different companies’ interests in a decentralized market was the engine that inspired the study based in CPFR. Efforts in marketing, transportation, international commerce and, mainly, the security information trade among forklifts firms are pieces of the supply puzzle that need to be studied in order to improve business and discover new solutions and methodologies.

Conclusion

As a consequence of this study, an improvement in the quality of the supply and a high level forecast management was proposed.

The main logistics variables of time, resource and information were used with the TRIZ principle take out. It is important to realize that a supply chain is based in the quality of its relationships and the level of trust that exists among them. Using TRIZ will help develop those relationships and accomplish a true partnership among chain partners.

The use of TRIZ to support company purchasing decisions will bring a competitive advantage to a supply chain. TRIZ will help deal with emerging contradictions and combining analysis with global and individual activities in the supply chain, finally creating a circle of study, quality and growth.

References

  1. Ruchti, Bruno and Livotov, Pavel, TRIZ-based Innovation Principles and a Process for Problem Solving in Business and Management, The TRIZ Journal, December 2001.
  2. Teti, Roberto and D’addona, Doriana, TRIZ-based Tool Management in Supply Networks. TRIZ Futures 2006 Conference.
  3. Stallkamp, Thomas T. (2006) Score! A Better Way to Do Busine$$: Moving from Conflict to Collaboration, Bookman Press.
  4. Ritzman, L.P.; Krajewsky, L. J. (2003). Administração da produção e operações. Pearson Education do Brasil, pp. 249-253.
  5. Salamatov, Y. (1999). TRIZ: The Right Solution at the Right Time. Insytec BV, The Netherlands.
  6. Andraski, J. (2004). Voluntary Interindustry Commerce Standards, VICS, pp 5.
  7. Domb, Ellen, Using the Ideal Final Result to Define the Problem to be Solved, The TRIZ Journal, June 1998.
  8. Morin, E. (2005). Introdução ao pensamento complexo. Sulinas, pp 85-86.
  9. Zhang, Jun, Chai, Kah-Hin and Tan, Kay-Chuan, 40 Inventive Principles with Applications in Service Operations Management, The TRIZ Journal, December 2003.

Note: This paper was presented at 2º Congreso Iberoamericano de Innovación Tecnológica Monterrey, N.L. 30 de octubre al 1 de noviembre de 2007.