Procurement of mechanical distribution products at Cargill July 2009 Author: Olivier Vegter Principal company: Cargill Supervisor: Harm-Jan Meijer University of Groningen 1

(1)

Master thesis Technology Management

Procurement of mechanical distribution products at Cargill

July 2009

Author: Olivier Vegter

Principal company: Cargill

Supervisor: Harm-Jan Meijer University of Groningen

(2)

2

Preface

In January I started my internship at Cargill to conduct a research on the procurement of mechanical distribution products. The research resulted in this research report. The six months I spend at Cargill were dedicated to all the ins and outs regarding the mechanical distribution product group. Additionally, this project is the final step of my study Technology Management at the University in Groningen. Cargill as an organization, and the research project, turned out to be a very suitable way to continue the theory of the study Technology Management in a more ‘practical’ manner.

Most of the time I worked from Cargill’s office in the port of Amsterdam at the central procurement department. This was added with frequent visits to production locations and meetings with suppliers. I found the dynamics of a multinational company like Cargill very interesting and did not just learn a lot about the research subject, but also about the characteristics of a company and its stakeholders. For this experience I would like to thank everybody I got to meet in the past months.

Some people I would like to thank in particular. First of all I want to thank Harm-Jan Meijer. He contacted me last year with a research project in mind and from then on provided me with direction, challenged me when necessary and was great to work with. Next, I would like to thank my colleagues at the Amsterdam office: Bart, Deborah, Jasper, Jenny, Martin, Otto, Rob, Stefan and Ted. They received me as part of the team, helped me when possible with their experience in procurement and created a pleasant atmosphere.

Furthermore I would like to thank my supervisors from the University of Groningen. Dirk-Jan Kamann provided valuable feedback with his expertise as a professor in procurement. Martijn van der Steen was my second supervisor and helped to give direction regarding the financial part of the research.

Olivier Vegter

(3)

3

Management Summary

Cargill is a company active in the purchasing, distribution and processing of agricultural commodities like food ingredients, cattle feed and pharmaceuticals. These activities involve 14 processing plants in The Netherlands. The product category of Maintenance, Repair and Operations (MRO) comprises all products that are used to keep the production sites functioning and perform maintenance and repairs on them. Mechanical distribution products (MeDi) is a subcategory of MRO and concerns non-electronic products that are acquired through distributors. It represents a spend of ca. ‘confidential’ in The Netherlands for Cargill in fiscal year 08/09. The main product groups in this category are: bearings, power transmission gear, hand tools, nuts and bolts, (mechanical) seals and hoses and appendages. Despite central procurement efforts much of the MRO procurement is done locally by employees in the various plants and offices. There is a lack of strategy and cooperation regarding the procurement of this product category. Therefore, the research objective is: Develop a strategy and best practices for the organization

of the procurement of mechanical distribution products in The Netherlands.

Evaluation of available theory and best practices showed that there are several elements of a successful MRO purchasing organization which are interdependent and are related to reducing the complexity of the supply base, increasing process efficiency and inventory optimization.

Firstly the supply base, procurement policies, the procurement organization, relevant processes and stakeholders were evaluated. Then the total cost of ownership was calculated. It turned out that the purchase price and inventory carrying costs accounted for over 90% of the total cost of ownership. From this point a strategy and best practices for the procurement of MeDi products were designed.

(4)

4

1. INTRODUCTION

1.1 Context

Cargill is a company active in the purchasing, distribution and processing of agricultural commodities like food ingredients, cattle feed and pharmaceuticals. Additionally they provide their clients with risk management and financial solutions. The company is a US-based multinational, headquartered in Minneapolis. It had 120 billion US$ of revenues in fiscal year 2008 and 160.000 people employed in 67 countries. It is the largest privately owned company in the world. In The Netherlands Cargill employs about 2300 people in their production plants and corporate offices and belongs to the 20 largest companies in The Netherlands.

The company is organized in business units, primarily structured on the basis of the type of raw materials processed. Some business units have a geographical focus as well. The majority of the business units is also assigned to platforms. In total there are 75 business units and 7 platforms. The platforms are:

 Cargill animal nutrition and salt  Cargill agricultural supply chain  Cargill animal protein

 Cargill bio-fuels, bio-industrial & emerging business  Cargill energy, transportation and industrial  Cargill financial services

 Cargill food ingredients & systems.

(7)

7 Cargill is active in The Netherlands with the following activities:

 Cacao and chocolate processing  Cattle feed processing

 Crushing and refining of grain and oilseeds  Bulk terminal

 Malt processing

 Ocean & transportation, chartering, trading, operations, logistics and risk management for internal as well as external clients

 Poultry processing  Sugar trading

 Sweeteners and starches processing  Texturisers and emulsifiers processing  Refining of vegetable oils

These activities involve 14 processing plants in The Netherlands (Figure 1). The procurement function is executed locally as well as centrally. Raw materials are mostly centrally purchased by traders on global commodity markets. Non raw materials, indirect costs, are purchased both locally and centrally. Currently, the location of Cargill in Amsterdam houses a division that is responsible for part of the indirect spend of Cargill in Europe. Indirect spend refers to the procurement of products and services that cannot directly be allocated to the output of a business. Examples of these expenditures are mechanical and electrical parts and equipment used in the production process, safety equipment, office supplies and security costs. Indirect costs are divided in the main categories of MRO/CaPex (Products used for Maintenance, Repair and Operations and Capital Expenditures), Processing aids, Packaging and Services and IT. Cargill’s corporate procurement mission is to ‘Enable BU’s and functions to capture “best total

value” through overall cost reduction, collaboration and procurement best practices’ (Cargill - European

procurement update 2008).

In Amsterdam the following (sub-) categories of indirect spend are centrally managed: Packaging, temporary labor, travel, lease cars, telecommunications and IT. Other categories that are centrally coordinated but from other locations include processing aids and energy and MRO/CaPex. The level of coverage varies a lot and can be business unit wide, national, European and global. The MRO/CaPex category is currently managed on a European level by a person that is located in Germany. Besides managing the category he has several other responsibilities so his time is not exclusively dedicated to MRO. However, in 2009 a new category manager will be appointed in Amsterdam, dedicated solely to MRO. He will report to the European procurement manager.

Despite central procurement efforts much of the MRO procurement is done locally by employees in the various plants and offices through Europe.

1.2 Problem situation

(8)

8 distribution products. To increase the amount of managed spend there should be a central effort to develop a strategy and best practices for this category.

Harm-Jan Meijer, Cargill’s senior buyer for indirect spend of two business units, namely Refined Oils Europe and Grains and Oil Seeds Supply Chain Europe, requests a study into the development of strategy and best practices of procurement for mechanical distribution products.

1.3 Product category

The category Maintenance, Repair and Operations (MRO) comprises all products that are used to keep the production site functioning and perform maintenance and repairs on production sites. Examples are safety supplies, equipment and mechanical items. The categorization at Cargill combines MRO with Capital expenditures in the category MRO/CapEx. CapEx stands for capital expenditures which are investments in the development and acquisition of physical assets.

Part of the MRO/CapEx is sub categorized based on the type of product and type of supply chain. Mechanical distribution products (MeDi) is a subcategory of MRO and concerns non-electronic products that are acquired through distributors. Major products groups in this category are:

 Bearings

 Power transmission gear  Hand Tools

 Nuts and bolts  (Mechanical) seals  Hoses and appendages

In this sense the product group is restricted to products and does not contain any services. However, the supply of the products can be accompanied by certain complementary services like technical advisory, installation efforts and education by the supplier. Some of these services can be seen as criteria to select a particular supplier.

Another consideration is whether a product is critical or noncritical. Putting a label on the mechanical distribution category if it is critical or noncritical is not a straightforward job. Many of the items are not critical and are consumed regularly during planned maintenance efforts. Examples of these items are nuts and bolts, seals, hand tools and bearings. However during emergency maintenance the availability of the products can be critical to prevent (further) downtime.

Cargills’ indirect spend in the Netherlands can be seen in table 1. Mechanical distribution accounts for over ‘confidential’ million euro’s and contributes ‘confidential’ to the total indirect costs.

‘confidential’

(9)

9

1.4 Problem Definition and research questions

1.4.1 Problem definition

A lack of strategy and cooperation regarding the procurement of mechanical distribution products within Cargill in The Netherlands.

1.4.2 Research objective

Develop a strategy and best practices for the organization of the procurement of mechanical distribution products in The Netherlands.

1.4.3 Main research question

With the research objective in mind the main research question can be formulated:

How can the organization of the procurement of mechanical products from distributors be improved?

1.4.4 Research framework

Before sub questions are formulated, the research framework (figure 2) is constructed to obtain insight in the knowledge that is needed for the research objective. It reflects the factors that affect the goal of the research, their mutual relations and the delimitations of the research. The starting point for the design of an improved situation of the procurement of mechanical distribution is the current situation. From a financial assessment of the current situation and theories available in (management) literature, the to be situation should take form. Finally, the implications and the implementation of the alternative design of the organization of purchasing will be looked into. The letters A,B,C and D reflect the way the research is ordered in this document.

Figure 2. Research framework

1.4.5 Sub questions:

From the research framework sub questions can be defined to help answer the research question.

1. What are the theories and best practices on the procurement of mechanical distribution products?

This sub question intends to provide the knowledge framework for the further research. It sets the perspective before looking at the current situation.

2. What is the current situation of the procurement process of mechanical distribution products at Cargill?

This sub question aims at gaining insight of the current situation. Here the research focuses on the organization of the procurement process and the stakeholders.

(10)

10 After describing the current situation it will be analyzed from a cost perspective so areas for improvement can be identified.

4. How could the identified cost components be reduced?

This part focuses on the search for improvements of the identified cost components in the organization of the procurement process.

5. How can the improvements be implemented?

This question intends to provide a solution to the problem of the transition of improved concepts in a workable and practical implementation trajectory.

1.4.6 Scope

The research focuses on what Kamann (Kamann, 2008) calls ‘Operational purchasing’. This is the part of the purchasing function that incorporates the actual purchasing and excludes the sourcing process. Figure 3 shows the 9 steps of the purchasing function and the part that is defined as operational.

(11)

11

2. METHODOLOGY

After the perspective that reflects the mental framework for the research is set, this chapter will discuss the Diagnosis, Design and Change model of De Leeuw (2002). This model represents the design process that is the basis for this research. Then the technical research design will be discussed. Doorewaard and Verschuren (2003) distinguish 3 different elements which are the sources that are used for the research, the strategy and the planning. Here, the research strategy will be elaborated on and the sources for data and information will be described. The planning is not included because of its limited relevance. Finally, several methods and techniques will be discussed that are used throughout the research.

2.1 Perspective

This thesis is concerned with the organization of the procurement process, therefore it is important that the concept of ‘organization’ is clearly defined and what it comprehends. Dictionaries define organization as ‘ the way different elements are put together systematically’. In this thesis the more comprehensive definition of De Leeuw (2002) will be used which states that organization is ‘Designing the structure, the coordination and the calibration of activities within this structure.

The three organizational elements are Policies, Organization and Processes (POP) (Kamann, 2006). They can be described for different levels of the organization. For corporate level but also for lower aggregation levels like the purchasing department. Both the purchasing function and the purchasing department should reflect the nature of the three elements on corporate level. When any of the three elements is changed this will require adjustments in the other two as well. An example would be changing an invoicing process. This cannot be successful without changing policies that make sure employees adapt to this new process and change the organization so the new processes can be implemented. The three organizational elements will compose the mental framework from which the research will be executed.

2.2 Design process

The Diagnosis, design and change model of De Leeuw (2002) is applicable to this research. It is based on the believe that problem solving exists of three phases, namely the diagnosis phase, the design phase and the change phase. Before starting the diagnosis phase the available literature on procurement in general and MRO in particular will be reviewed to provide an adequate basis.

Diagnosis

In the diagnosis phase the current situation of procurement of mechanical distribution products at Cargill will be explored. This will include the POP elements and a stakeholder analysis. De Leeuw mentions (De Leeuw, 2002) that this phase is more of a general phase and that after finishing this phase it is possible to incorporate the right aspects in the design. When the right aspects and the potential areas for improvement have been identified through Total Cost of Ownership analysis, the design phase can be commenced.

Design

The potential areas for improvement focus the search for alternative ways of organizing the procurement process. This will be based on literature and people involved. Subsequently, the impact of the alternative on the costs of the procurement process will be attended to. The new design should be judged on the aspects of viability. Johnson, Scholes and Whittington (2008) identify three key success criteria for the viability of a strategy.

(12)

12 - Acceptability, this refers to the expected outcomes and the conformance with the expectations

of different stakeholders

- Feasibility, make sure that it works in practice. This will be evaluated from human resources, technical and financial perspectives.

These success criteria will be reflected upon from the perspective of the relevant stakeholders that will be identified in the diagnosis phase.

Change

Finally the implications of the alternatives for Cargill will be described and a roadmap to the to be situation will be drawn.

2.3 Research strategy

The research strategy can be typified as a case study (Verschuren & Doorewaard, 2003). This is a research with the intention of obtaining thorough insight in one or a few objects or processes limited in time and space.

It is typified by e.g. a small number of research entities and in depth research instead of a broad approach. This often results in face-to-face interviews and the search for knowledge through a variety of sources instead of solely interviews for instance. This is called triangulation of sources and will be discussed in the next section. Another feature of a case study is that it intends to obtain an integral insight in the object or process. The holistic character is important when focusing on redesigning a process since changing is risk full when not all aspects are included in the reasoning.

The cases that will be subject of the research will be chosen according to the ‘snowball sampling’ principle which means that the cases are selected one after another. This method is appropriate when limited knowledge on the subject is available beforehand. The starting point of the research is the plant in the Botlek area in Rotterdam. In the course of the research the plants in Bergen op Zoom and Sas van Gent were added for investigation.

One of the advantages of a case study with regard to other strategies is that it is not necessary to structure and secure everything at the start of the research. During the research adjustments can be made which makes it a flexible approach. It is an important feature of this research since at the start little is known related to the final direction of the research.

2.4 Information sources

(13)

13 Research question 1 2 3 4 5 Persons x x x x Media x x Reality x x Documents x x x Literature x x x x

Table 2. Information sources

2.5 Methods and techniques

The methods and techniques that are used during the research will be briefly described. 2.5.1 Total Cost of Ownership

A thorough understanding of the costs associated with purchasing the mechanical distribution (MeDi) products is required for both the diagnosis and the design phase. Ferrin and Plank (2002) show in their research that ‘leading edge’ companies use the Total Cost of Ownership approach to value purchase opportunities. Since this research has a more comprehensive approach relating to the purchasing process the total cost of ownership can prove useful according to leading edge standards. Ferrin and Plank (2002) summarize other authors by listing three issues that are tackled by the TCO approach. The first is the need for a long term perspective and to consider not mere the initial purchase price. Secondly, other business functions besides the purchasing department also need to be included in the valuation of a purchase. Finally, all the activities associated with purchasing need to be understood and measured with respect to their cost impact.

Ellram and Siferd (1993) write that since 1928 authors have discussed the importance of ‘considering cost beyond purchase price in choosing a supplier’. In this research the search for a supplier is not the first priority but looking for the right way to organize the procurement. Reducing costs are central in this case and this is not limited to the purchase price. Ellram (Ellram, 1996) introduces Total Cost of Ownership as ‘A structured approach for determining the total costs associated with the acquisition process per se…’. This connects well with the nature of this research.

A general model that reflects the composition of the TCO for the purchasing process of Cargill will be constructed in chapter 4. The factors that make up the TCO are gathered from both external sources like (management) literature and interviews with Cargill employees.

2.5.2 The Pareto principle

In the diagnosis phase the TCO perspective will be used to get a good understanding of the cost of the purchasing process and the main contributors to the total cost. The Pareto principle is useful to set the focus for the design of improved ways to organize the procurement process.

(14)

14 This principle can be translated to the composition of the Total Cost of Ownership. If this principle is valid for the case of the procurement process at Cargill as well, then the majority of the TCO can be improved working on only 20% of the sources of those costs. This approach is used to focus the design phase on the 20% ‘that really matters’.

2.5.3 Stakeholder analysis

(15)

15

3. THEORY & BEST PRACTICES

This chapter will attend to the available theory and literature on the organization of procurement in general and MRO in particular. Leading in the exploration of theory and literature are the perspectives on policies, organization and processes with respect to procurement. The first part on purchasing portfolio focuses on appropriate policies and strategies for particular situations. The second part elaborates on perspectives on the organization of purchasing and how this should be incorporated in the organization. Next, the processes related to purchasing will be attended to. Finally the procurement of MRO products in particular will be explored.

3.1 Purchasing portfolio

The well-known matrix developed by Kraljic (figure 4) is a tool that provides purchasing strategies based on two dimensions, namely the profit impact of the product group and the supply risk (Kraljic, 1983).The matrix distinguishes four quadrants: leverage, routine, strategic and bottleneck. The profit impact can be defined in terms of volume, percentage of purchase spend and impact on product quality or business growth. The supply risk can be assessed in such terms as availability, number of suppliers, competitive demand, make-or-buy opportunities, storage risks and substitution possibilities.

Figure 4: Kraljic matrix (source: Kraljic, 2002)

The procurement of the different categories has to be approached in different ways based on their location in the matrix. For each of the four quadrants strategies can be defined.

Routine products do not have a big impact on the purchasing spend and there are many suppliers that can deliver. However, a lot of resources are used in the procurement process of all the relatively low value items in this quadrant. This unfavorable balance results in a focus on efficiency of the procurement process for the products in this quadrant. This is reflected by efforts to increase efficiency by e-systems or even outsourcing the procurement of this category (Case, Gilmour and Smithson, 2007).

Typical bottleneck items are not ordered with the same frequency and volume as products in the other quadrants. However, they are critical to the business in a certain way. The key objective for the procurement of this category is the assurance of the supply of these products.

(16)

16 Products that can be marked strategic have a high profit impact but the supply market can be complex with not many suitable suppliers that can offer the specific value that is needed. The right approach for these items is to seek collaboration with suppliers. In this way (joint) value creation can be increased and supply is assured.

Figure 5 shows some more characteristics of the suitable strategies for the different quadrants.

Figure 5: Purchasing sophistication (source: Kraljic, 1983)

3.2 Organization

Gelderman and Albronda (2007) define the next three major distinguished shapes of the purchasing organization:

- Central purchasing organization - Decentralized purchasing organization - Coordinated purchasing organization

(17)

17 purchasing organization. They argue that the choice for central vs. decentralized should be in accordance with the ‘company policy’, in particular the policy referring to the autonomy and independent responsibility of their profit and loss account. There are a number of considerations:

- A strong product commonality pleads for ‘central’ - Geographical distance pleads for ‘decentralized’ - Vulnerability to buying markets pleads for ‘central’ - More potential for cost savings pleads for ‘central’ - A higher need for expertise pleads for ‘central’ - Large scaled suppliers plead for ‘central’

- When customers’ influence is higher this pleads for ‘decentralized’

Gelderman and Albronda conclude that the considerations are based on creating leverage on the one side and independence of plants and business units on the other.

The matrix in figure 6 (M. Boodie, Berenschot consulting) also handles the central vs. local procurement issue but more significantly values the perception of the local ‘agents’. It shows the way purchasing should be organized as a result of two dimensions, namely financial importance and/or importance of the product group and the perceived importance by decentralized agents.

Figure 6. Financial importance vs. perceived importance matrix (source: M. Boodie)

The structures introduced in the lower and left quadrants are straightforward and require no further elaboration. The suggestion in the upper right quadrant however is more sophisticated. This would be a hybrid version with local purchasing organizations which are centrally coordinated. The coordination is the responsibility of central commodity teams.

3.3 Processes

(18)

18  The new-task situation, this is the case when an organization decides to purchase a new product

which has not been purchased before.

 Modified re-buy, This happens when a new product is bought from an existing supplier or a familiar product is bought from a new supplier.

 Direct re-buy, this means that a product is purchased before and the company is familiar with the concerning supplier. The supply conditions are usually registered in a contract.

MRO products are mostly bought in modified and direct re-buy situations. Since the great variety of MRO products the purchasing organization will spend much time on administrative activities. The repetitive and time consuming nature of the purchasing of MRO products pleads for a focus on efficiency and a reduction of overhead costs associated with the purchasing process.

Potential solutions can be found in standardization, restriction of the assortment, supply base reduction and the establishment of preferred vendors and framework agreements (Gelderman and Albronda 2007). More extensive solutions put more responsibility at the suppliers regarding supplies, inventory policies and control, also called vendor managed inventory. When suitable framework agreements are in place a much occurring problem is ‘maverick buying’. This is the term for buying products outside contracts that are in place. This non compliance has several consequences like fragmented buying power and less control over the supply conditions.

3.4 MRO procurement

Scientific literature on the subject of MRO purchasing is quite scarce. Bechtel and Patterson (1997) acknowledge the increased attention in MRO purchasing as a potential area for large savings. However they see the current developments like blanket contracts and procurement cards as too narrowly addressing only the ordering and processing costs. They argue that more significant savings can be obtained through optimizing inventory and improving receiving, distribution, and searching activities. Establishing partnerships with suppliers is a good catalyst for these improvements. Three steps to better MRO purchasing are suggested. Firstly companies need to cluster products into commodity groups and choose a suitable supplier for every group. Choosing a single source has several benefits like better prices due to increased volumes, reduced amounts of paperwork and dedicated on-site technical support. The next step would be to distinguish between spare parts and consumables. The spare parts can be managed in the traditional way, but the supply chain of consumables becomes a responsibility of the relevant supplier. They handle the inventory and supply through bar-coding and two-bin systems. Process efficiency is a major benefit in this situation. The final step is to get suppliers involved in the continuous improvement of the supply chain, processes, inventory levels and technical expertise.

(19)

19 Possibly less relevant for mechanical distribution, due to the low value nature of the items, is pooling of stock. In their case study, Le Sueur and Dale (1998) explored the potential for the pooling of stock between different locations regarding high value items. The logic was that critical products that had to be kept on stock but were used rarely could be kept on inventory in smaller numbers to reduce costs. However, in the case they researched there was a lack of common coding or description so it was impossible to discover any commonality. To continue the pooling of stock there needs to be a full technical review and extensive codification of all the items.

The journal of IIE Solutions features an article on best practices of MRO purchasing, based on a study by industrial supply distributor W.W. Grainger. Their conclusion is that reducing the number of suppliers is the major improvement in MRO purchasing and complexity. Furthermore they suggest to change inventory methods, change internal staffing and upgrade technology.

Besides the small quantity of peer reviewed literature on MRO procurement there is management literature on purchasing, engineering and supply chain management covering this subject. Most of their contributions are based on best practices of surveyed companies. Next, a review of their contributions. In the supply chain management review of November 2006 (V. Checketts and V. Patel) three elements are described. Firstly ad hoc purchases need to be controlled and orders need to be according to existing agreements with (preferred) vendors. Additionally, electronic payment needs to be increased, so transaction costs are lowered. Finally, visibility of MRO spend needs to be increased through spend analysis and intelligence.

S. Avery has several contributions to the MRO case in the journal of Purchasing (1995,2006). She encourages companies to increase departments’ efficiency, enhance corporate purchasing power, reduce inventory, implement ordering systems and standardize across the product category, make departments accountable for their MRO spending and improve the payment situation by consolidating invoices and negotiate discount terms. In a more recent edition she identifies three other best practices which are collaboration with suppliers, the centralization of the MRO procurement and reducing the supply base. In 2006, the journal called ‘Purchasing’ conducted a benchmark study on best practices of MRO procurement. Their study identified four characteristics that deliver the highest Return on Investment. These were: Measuring supplier performance, collaboration with suppliers, the implementation and integration of IT systems and attracting talented employees to become owner of the MRO purchasing process.

(20)

20

4. DIAGNOSIS

In this chapter the current situation will be explored. The supply market will be described and the Policies, Organization and Processes regarding procurement of mechanical distribution products will be attended to. Then there will be looked into the stakeholders involved in a stakeholder analysis. Finally, the current situation will be screened from a total cost of ownership perspective. This results in a thorough insight in the procurement process and its cost composition. From this point of departure, the next chapter explores the improvements to the purchasing process.

4.1 Supply base

‘confidential’

Typical MeDi suppliers have a broad assortment of brands and products. Although the main function of the suppliers is distribution of these products, some also have production facilities for part of their assortment. Since MeDi consists mainly of large quantity, low value orders suppliers are focusing on the acquisition process issues of their customers.

MeDi products are used during maintenance and repairs to keep operations running and it can happen that a product is critical to the production process and that time is an important issue in the supply chain. This results in a need for suppliers to have a local facility and some sort of an emergency service, when products are not kept on stock by Cargill.

During the research it was emphasized various times by local purchasers and maintenance employees that additional services of suppliers are crucial. Suppliers need to have technical expertise and be able to assist Cargill through technical challenges. Subsequently they can take a role in the (technological) education and development of knowledge and skills in the Cargill organization. Also, they need to cooperate with Cargill in the synchronization of their assortment and their stock levels to the demands of Cargill.

Finally, close cooperation results in process improvements regarding ordering, logistics and invoice processing. Currently, Cargill is joining up with suppliers to connect ordering systems and digitalize intensive invoice processes.

4.2 Policies, organization and processes

(21)

21 Different locations have their own supply base and although there is some commonality in suppliers used, there has been no effort to synchronize MeDi supply chains between locations. This results in fragmented spend over suppliers, and suppliers that have different agreements, terms of delivery and prices for the various Cargill locations. Analysis of spend data revealed that between Cargill locations there are differences in prices paid up to 49% for the same product, supplied by a single vendor. There is no central effort to look after the interests of Cargill as a whole regarding mechanical distribution.

There is no standardization across the different production sites of Cargill. Different locations can use different brands and product specifications for the same application. This is partly caused by differences in installed base. The installed base refers to the specifications of the equipment that is installed in the plant. This results in equipment that, although it has the same purpose, does not share the same components. Consequently spare parts are not interchangeable. These deviations are partly due to acquisitions of plants with a different installed base. Besides the lack of standardization there is also no coding on item level that is universal for all locations. Locations have their own coding method so it is hard to compare and combine information from different locations. Universal coding would be necessary to improve spend intelligence and transparency across business units.

MeDi items are kept on stock at warehouses on site or delivered on demand. The products are ordered by the warehouse manager when stock runs low. When products that are not kept on stock are required by the maintenance team it is ordered by the warehouse manager. Orders can be approved by the warehouse manager if the value of the items does not exceed ‘confidential’ . Orders for products exceeding this amount have to be approved by the maintenance manager. If the product price exceeds

‘confidential’ the plant manager needs to approve as well.

Deliveries are accepted and processed by the warehouse manager. Some suppliers deliver daily to the plants, others less frequent. The warehouse manager receives the products, inspects the contents, unpacks them and places the products in the warehouse. Then the delivery of the product is confirmed in the maintenance management system ‘Maximo’ which also contains an inventory management database. Maximo contains the data relating to the purchase order, e.g. vendor information, product description, quantity, price etc. The relevant data is send to the accounting system (J.D. Edwards).

The supplier sends the invoice to the Financial Service Centre (FSC) in Bulgaria . This is the Cargill shared service centre that is responsible for processing the invoices of Cargill locations in various countries in Europe. In the FSC the invoice is checked on legal requirements and digitalized by scanning. Then the data on the invoice is matched with the corresponding data in J.D. Edwards. If information regarding prices, P.O. numbers, plant description etc. do not match, the invoice is send to the concerning location. They are responsible for solving the error in either the invoice or the purchase order, which is a labor intensive process. It happens quiet often that invoices contain errors. Reasons for errors can be:

 Prices in Maximo are not updated and do not correspond with the prices suppliers put on the invoice.

 The supplier does not send the invoice to the FSC but to the plant location.

(22)

22 If the purchase order and the invoice matches, then the supplier can be paid. This is done by a third party that is responsible for the accounts payable.

There is a central effort to automate the invoice processing even further. Electronic Data Interchange (EDI) is used to interface between suppliers and Cargill’s accounting system. EDI makes sure that invoice data is always presented to Cargill in the right format and eliminates the need for manual scanning. This reduces the window for errors. Suppliers are persuaded to use the EDI step by step although they are not always satisfied with the change to EDI. This is due to the extra subscription fee they have to pay monthly, the increased administrative activities required, and the sensitivity of the EDI interface to the way the data is presented. For example, errors occur due to the units that are used. It needs to be synchronized if particular products are presented in pieces, (square) meters, kilograms etc. Some errors also occur due to deficiencies on behalf of Cargill. For instance in the case of emergency orders when Cargill employees quickly order a product without following the formal procedure that generates a P.O. number. When the supplier does not have a P.O. number it cannot charge Cargill for the goods supplied, because the EDI does not accept an invoice without a P.O. number generated by Cargill.

A detailed view on the ordering and invoicing process can be found in Appendix 1.

In some locations projects relating to Vendor Managed Inventory have started. This mostly relates to low value products that are used with high frequency and are stored in two-bin systems. The bins are monitored and replenished by the suppliers without interference of Cargill personnel. An example is nuts and bolts which are in some cases monitored and replenished if necessary by suppliers.

4.3 Stakeholders

From Cargills’ point of view, two main categories can be distinguished which are internal and external stakeholders. External stakeholders are relatively easy to find regarding the organization of the mechanical distribution purchasing. Firstly, the suppliers to Cargill that are further up the supply chain, form an important stakeholder. They are directly linked to Cargill through the mechanical distribution supply chain, so they can be affected by any changes in the process by Cargill. Also, Cargill will need the cooperation of suppliers for some changes in the process.

The other group of stakeholders that is external to Cargill are 3rd parties that have some kind of facilitating role in the process, i.e. companies offering EDI services like OB10 and companies offering logistical services.

The internal stakeholders are more abundant and diverse. For the identification of these parties the process flow charts will be looked at. From these charts the following stakeholders can be identified: Maintenance, warehouse managers, plant management, Financial Service Center and Plant Accounting. These are the parties directly involved in the purchasing activities of the plant. However, additional stakeholders are senior buyers and purchasers at plant locations, they focus on the sourcing process preceding the operational purchasing. This involves collecting quotations and negotiating contract terms. Furthermore there is the European central procurement department headquartered in Amsterdam, they are also strongly committed to the development of a vision and policy for procurement in general and MRO in particular. Finally there is the management of the business unit. They have the authority to decide on the implementation of any major changes in the procurement of mechanical distribution products within their BU.

(23)

23 becomes more of a strategic supplier, for instance after reducing the supply base and supply chain integration efforts.

In the case of Cargill there are not many facilitating parties. This is limited to occasionally an express delivery service for quick deliveries and OB10, the company that operates the digital portal through which in some cases invoices are exchanged from suppliers to Cargill. This group would have limited interest and limited power in the organization of purchasing of MeDi products.

The maintenance team is a stakeholder with a high level of interest because they are the end user of the product. They are powerful too because they can specify the attributes of the products and the boundary conditions the process has to comply with in terms of delivery times, additional services needed from the supplier and preferred brands.

Warehouse employees play an important role in the ordering process and the relationship management with suppliers. Their personal contact with the supplier on a regular basis can result in emotional motives and preferences. As a consequence, their level of interest is high but they have limited power.

Plant management usually consists of one person, a plant manager that is responsible for the (financial) results of the plant. This means he is concerned with the purchasing process from a cost perspective. On the other hand the plant managers’ greatest concern and responsibility is the up-time of the plant so from this point of view purchasing just has to make sure that the products required to keep the plant running are available from the right quality. It can be assumed that plant management is not interested in the way mechanical distribution procurement is organized as long as the products are available when needed and the continuity of the production process is not affected. When this changes the plant manager has power to influence the purchasing process.

The Financial Service Center is a so-called cost center purely occupied with the processing of the invoices. From their point of view they have a high level of interest in the way the process is organized because most steps in the purchasing process affect the invoicing in some way of another. In particular they are interested in their cost drivers which are the way invoices are delivered to them and their quantity and the correctness of the data on the invoices and in the Cargill systems. However, their power is limited to recommendations.

Plant accounting is involved in the process of repairing invoicing errors. Their interest of the purchasing process focuses on reducing errors in the invoicing and their power is limited.

Local purchasers are strongly committed to improving the purchasing process. Their power to do so is significant.

The central procurement organization is committed to the organization of the MRO purchasing process to increase the managed spend in this category and encourage leverage through cooperation between the BU’s of Cargill. Their power to influence the process is significant and resources are attributed to increase ‘managed and leveraged’ spend.

(24)

24 Figure 7. Power vs. interest matrix (adapted from Gardner, 1986)

Based on this classification a strategy can be defined for approaching and involving the different groups of stakeholders. According to Gardner (1986) the groups should be approached as follows:

Group A: Minimal effort

Group B: Keep them informed about the project, they might play an important role in the opinion of powerful stakeholders.

Group C: They are powerful but not that concerned about the project. They have to be kept satisfied to make sure they do not move to quadrant D.

Group D: Their high level of power and interest makes them the most important stakeholders in the process. They have to be involved in the project, their interests have to be heard and their support is crucial to the success of the project.

Based on the classification and the attributes of the 4 quadrants it is clear that suppliers, the maintenance team, local purchasers, plant management and the central procurement organization are the major stakeholders in this situation. They have to be ‘on board’ for the success of the project. The warehouse managers and the Financial Service Center need to be kept informed to minimize (negative) effects they could have on more powerful stakeholders. Additionally, they can play an important role in identifying errors and improvements in the procurement of mechanical distribution products. Finally, during the research it became clear that it is crucial to get support of the business unit management because they have the authority put changes into motion.

4.4 Total cost of ownership

In chapter 2, the Total Cost of Ownership approach was introduced for analyzing the current cost composition of the procurement process for mechanical distribution products. In this chapter the different factors that make up the TCO will be described. There are two sources to draw from, namely literature and experiences of Cargill employees. The literature is not unambiguous about the factors that comprise the TCO. One has to look at the specific activities related to external purchasing because this differs per company (DeGreave and Roodhooft 1999).

(25)

25 The factors that influence the TCO are derived from the flow charts of the ordering and the invoicing process which are based on the interviews with employees of Cargill.

Figuur 8 - Contributors to the Total Cost of Ownership

The literature on TCO includes cost components like maintenance, energy and disposal. However, these costs are incurred when the acquisition process is finished. For this reason it does not affect the considerations concerning the organization of the purchasing and supply chain and will not be included in the Total Cost of Ownership approach and TCO model.

The next paragraphs contain a description of the factors influencing the TCO and the components that comprise them. Additionally, the factors will be depicted in a model showing the relation of the different factors. In the model, the factors are colored:

Red: Based solely on literature Green: Based solely on interviews

Red/green: Based on literature and interviews

(26)

26 4.4.1 TCO model

Price

The price of MeDi products is one of the main contributors to the TCO. Traditionally, it is the main hallmark against which buyer performance has been measured, but in a supply chain setting it is just one of the relevant variables (Cavinato, 1992). The relevant price for the TCO results from two sub variables, namely the catalogue price and the rebate . The catalogue price is the base price that Cargill and its supplier agree upon and which will be on the invoice received by Cargill. Nevertheless, the rebate has to be incorporated in the price before one can conclude how much it contributes to the TCO. The rebate is the additional discount or payback the supplier offers in relation to the total spend of Cargill with the company. For example, Cargill could agree with a supplier that a 3% rebate will be redeposited when spending with that particular company exceeds Euro 100.000,- within a certain time frame.

Ordering

The costs associated with ordering are based on the activities executed to get the information at the supplier that matches a certain need for a product in the organization. Cavinato writes about direct transaction costs which include the detection of the need for a product, the requisitioning, the preparation and transmittal of documentation and the input of the receiving information to relevant systems. Information systems have played an important role in the evolution of the ordering process. Based on literature and interviews the following subvariables can be distinguished: Requisitioning, approving, assigning PO, Sending PO and confirm the delivery (appendix 1a). Degraeve, Roodhooft and Van Doveren (2005) add a sixth, which is the quotation activity. It involves the settling of the conditions like price and lead time associated with the transaction.

Supplier handling

(27)

27 supports the transactions. This consists of welcoming sales agents that visit the company, social activities, supplier education etc. The third factor is coping with contingencies, irregularities in the order process that have to be solved in cooperation with the supplier (Kamann, interview 20th February 2009).

Financial

The handling of the invoices by the company affects the TCO. The actor activity diagram of the invoicing process (appendix 1b) shows the different actors involved in the process. The costs associated with this are dependent on the way the process is organized. Information technology, in cooperation with the supplier, plays an increasing important role in decreasing invoice processing costs.

Errors in this process are common, therefore the number of errors is a cost driver as well. When something goes wrong in the invoicing process additional activities have to be executed to diagnose and correct the error before the invoice processing can be finished. Finally the payment conditions play a role in the financial variable as well (Degraeve etal. 1999). A longer term of payment gives the purchasing organization the option to obtain additional profits with these funds till the money has to be transferred.

Inventory

The costs of inventory, or inventory carrying costs, are influenced by a few factors. A number of authors have described these factors. Some of them are common and some differ. According to Robert Lamarre (2003), the costs are composed of the costs of capital, cost of space, handling costs, stock obsolescence, insurance and a combined category of spoilage, damage etc.

In a IOMA/Harding survey (2005) a list is created of factors used by companies to calculate the cost of inventory. They list the following, starting with the factor most mentioned by companies and ending with the least mentioned: cost of money, obsolescence, space, taxes, insurance, personnel, handling, storage, overhead and equipment.

(28)

28 Table 3: literature on inventory costs

Then there is Inventory management report(2004) which mentions cost of money, taxes, obsolescence, insurance, storage space, personnel, capital equipment, secondary inspections and computation costs. Table 4 contains an overview of the different factors the various information sources mention and their commonalities. Based on this the following factors will be added to the inventory cost model: cost of money, taxes, insurance, cost of space, handling costs and obsolescence. Combined with the findings within Cargill it can be concluded that adding overhead, equipment, spoilage and pilferage as separate cost drivers is not necessary because they are incorporated in the others in some way or not relevant for this specific situation.

Logistics

Logistical costs are composed of the regular costs of transporting the product between the supplier and the warehouse as well as the costs of express delivery for emergency orders.

(29)

-29 4.4.2 Current cost composition

This section explains the values used in the calculation of the TCO for the current situation and their origin.

Price: The total price paid for all mechanical distribution products used by the three plants are available.

In the current situation there are no rebate agreements in place.

Ordering: The costs associated with an order are derived from internal studies by Cargill into the time that

is associated with placing an order. This is approximately 30 minutes. Further internal investigation revealed that the costs of labor is approximately Euro 25,- per hour for the people involved in ordering. Combining this information comes down to Euro 12,50 per order. The number of orders per year is available so the total order costs can be calculated.

Supplier handling: Based on interviews it can be said that supplier costs associated with relation

management and communication are on average Euro 1000,- per year. Multiplied with the number of suppliers a figure for the cost of this cost category can be calculated.

Financial: Invoices are send directly to the financial service centre by the suppliers. Cargill monitors the

number of invoices that is processed by the financial service centre and how many of them contain an error (non compliancy). The financial service centre is a cost centre and allocates their costs to the various locations. This is done based on the type of invoice and whether it contains errors. When an invoice is non compliant it is sent back to the concerning location and they have to correct the error. Further investigation in the time local employees spend on correcting the error and the allocation rates of the FSC provides a cost rate of over ‘confidential’ per invoice that is non compliant. The payment terms affect this variable as well. It refers to the number of days Cargill can postpone payment of invoices. currently it is on average 30 days. It cannot be given a value, but is included as a cost factor since extending the 30 day period might induce extra savings.

Inventory: The costs associated with holding inventory have been determined as a percentage of the

inventory value. The value of the total inventory that represents mechanical distribution products has not been identified before. However, the suppliers for all the products in inventory were known and the value of every single product. From this information the total value of the items that are on stock and coming from mechanical distribution suppliers can be calculated. Taxes and insurance are zero since in The Netherlands inventory is not subject to tax laws and Cargill has no insurance coverage for their inventory.

Logistics: This cost factor is zero in the calculations since mechanical distribution suppliers deliver on a

(30)

30

Table 4 - Total Cost of Ownership current situation

(31)

31

5. IMPROVEMENTS

This chapter will discuss improved ways to organize the procurement of mechanical distribution products. Firstly, there will be a reflection on the current situation and where improvement efforts should be focused. This will form the basis for improvements which will be presented next. In the next chapter the improvements will be incorporated in a redesign of the procurement organization.

5.1 Reflection from TCO perspective

From table 5 it can be seen that the initial price (75,5%) of the product and the costs of keeping them in inventory (18,1%) accumulate to over 90% of the total costs. Since they are the biggest contributors to the total cost of ownership they will be a leading factor in the search for improvements and the new design. Improvements should reduce the initial purchase price and/or the inventory carrying costs to significantly contribute to lowering the total cost of ownership of the MeDi category (figure 10).

Figure 9. Improving the organization

5.2 Theory and best practices

5.2.1 Kraljic

To locate the position of the concerning product group in the Kraljic-matrix, it needs to be ranked on two criteria. With 1,5% of total indirect costs, the financial importance of the MeDi products category is quite low. The complexity of the supply market is not that easy to determine because it depends on what the exact value proposition is that Cargill is looking for. There are many suppliers of mechanical distribution products. Cargill uses ‘confidential’ in The Netherlands alone. However, if the value proposition Cargill demands includes a certain level of service in the form of installation, advisory and expertise as well then the complexity increases. Other factors that tighten the supply base are the demand that the supplier carries all the products needed, that they are willing to take certain products on stock and able to deliver to certain locations within a certain time window. If Cargill looks for a supplier that is capable of supplying all of Cargills’ MeDi product needs to all plants in Europe within two hours and without the plants required to carry safety stock, then the number of suppliers might be reduced to 1 or 2 or even none. This would mean a relatively complex supply market.

(32)

32 optimization, efficient processing and inventory optimization (Kraljic, 1983). Kraljic also suggests the establishment of local suppliers, decentralized decision authority and a focus on functional efficiency. 5.2.2 Policies, Organization and Processes

The review on theory and best practices in chapter 3 showed that most authors put the emphasis on process efficiency and cutting overhead costs associated with the MRO purchasing process. Cargill has put effort in process efficiency. IT systems are in place for monitoring inventory, ordering and invoice processing. Maximo, is a maintenance management system that monitors inventory and supports the order process functions well. However, it only monitors the stock levels and does not function as a tool to reduce them. This results in exceeding stock levels with accompanying costs.

The major driver for invoice processing costs are the errors that cause a mismatch. The reasons for the errors can often be allocated to communication deficiencies between Cargill and the supplier. Implementation of EDI can structure communication processes further to reduce the potential for errors. Also, EDI reduces the labor intensive processing of the invoices and accompanying errors. However, to ensure the benefits of EDI and the cooperation of suppliers, it is critical that Cargill collaborates with them to eliminate the increased administrative burden and agrees on the way the communication and ordering process is organized.

(33)

33

5.3 Price

Bundling spend as a way to reduce purchase price levels is best illustrated with the Kraljic matrix. Kamann (2003) suggests an active approach in the localization of suppliers in the Kraljic matrix. Gelderman and Albronda (2007) mention ‘conditional dynamics’ and Gelderman and Van Weele (2003) mention ‘Strategic directions’, which both refer to actions focused on moving from one quadrant to another (figure 11). Both Kamann (2003) and Gelderman etal. (2003, 2007) suggest that if the procurement of a product group is fragmented it can be bundled so it moves up in the direction of the leverage quadrant. Combining the spend creates leverage so better prices can be negotiated. A better price is the major driver for bundling spend and reduction of the supply base. Besides that, there are other benefits as well.

According to Gelderman and Albronda purchasers should always ask themselves if it is possible to standardize and bundle routine products to improve supply possibilities. They also recommend supply base reduction, increasing purchased volumes and framework agreements. Delivery conditions can be set and simplification/improvements to ordering processes, logistical processes and invoicing processes can be developed.

Goffin, Szwejczewski, and New (1997) also mention advantages of supply base reduction like more time available to manage close relationships with suppliers. In collaboration with suppliers improvements can be developed regarding inventory levels, standardization, administrative costs and maverick buying.

Figure 10. Strategic directions (source: Gelderman & Van Weele, 2003)

5.4 Inventory

(34)

34 stocked. This calls for a thorough insight in the nature of the items kept on stock before cutting them. They need to be classified and for each category different policies need to be designed according to the characteristics of the products (Chakravarty, 1981). Bailey and Helms (2007) suggest that a criticality classification is usually the first step in inventory understanding and reducing spare parts inventory. The classification of the products in categories is called ABC analysis (Lung Ng, 2005). Classical ABC analysis assigns products to categories based on the annual usage in for instance dollars or euro’s. However, Braglia, Grassis and Montanari (2004) point out that this one-dimensional analysis does not offer the potential to differentiate between other parameters like criticality, price of the item, availability of the item, lead times, and the volume they take in inventory. Braglia etal. (2004) sum up a number of methods that are developed for multi criteria ABC analysis. It differs per method which criteria are used as parameters. Bailey and Helms (2007) suggest the construction of a team that decides on the criteria used in the classification.

Since the nature of mechanical MRO it makes sense to let the factor ‘criticality’ play an important role in the classification in addition to annual usage in euro’s. If Cargill decides to classify the items according to these criteria, a suitable categorization and corresponding policies would be:

A.  Critical spare parts

This category should always be kept on stock. Policy: Keep on stock at sufficient levels.

B.  Spare parts Used annually but not critical.

Policy: Reduce stock levels as much as possible. Collaborate with suppliers for consignment stock. C.  Consumables

Frequently used but of low value.

Policy: This category can be optimized through process efficiency, Vendor Managed Inventory, Vendor Managed and Owned Inventory and outsourcing.

Consignment stock means that suppliers keep the products in inventory while guaranteeing a minimum inventory level and delivery conditions (Valentini and Zavanella, 2003). This secures the supply of the product within a certain amount of time when needed. This is a suitable policy for the spare parts that are critical to a less extend. It could be applied to critical spare parts as well on the condition that delivery times are sufficiently short to prevent extended down time.

(35)

35

6. CONCLUSION AND RECOMMENDATIONS

Firstly, the redesigned organization and the effects on the procurement of mechanical distribution products will be discussed. The next part will elaborate on the suitability of the concept, the acceptability by the stakeholders and the feasibility of the concept. The last section will discuss the realization of the redesign, the so-called roadmap to transition.

6.1 Redesign

The focus areas for the redesign of the purchasing of mechanical MRO are threefold. Supply base reduction and the bundling of spend is an important element because of its positive effects on the purchase price. The second major element will be inventory mapping and the accompanying policies to take on the inventory costs. Finally, process efficiency will also be leading in redesigning purchasing. When Cargill decides to reduce the supply base there needs to be leadership on a national level to get the various locations to cooperate and hand in some of their autonomy. It can be done by a team that is composed of members from different business units or a single person that has support of the locations. There has to be consensus on two things: The specific value proposition that is required and the objective of reducing the supply base. The value proposition needs to be made clear in consultation with the key stakeholders. The value proposition should incorporate certain levels of service, quality, process requirements and delivery conditions.

There also needs to be agreed upon the objective of reducing the supply base. Is it for the mechanical distribution category as a whole, then it means that the suppliers remaining in the future have a very broad assortment, perhaps even adjusted to the wishes of Cargill. In this situation the number of suppliers can be reduced to a few or even one. The other option that was introduced in the chapter on theory and best practices is to announce a specific supplier for each product group. For instance, the product groups of bearings, seals, transmission gear and consumables would be supplied by separate parties. This scenario would come down to more suppliers left in the optimal situation, but has potential for increased service and technical expertise offered by the suppliers. In both situations framework agreements should come into place with remaining suppliers.

A smaller supply base can be managed more successfully, and the potential for intensifying collaboration with suppliers increases. Collaboration with suppliers is a prerequisite for process improvements and inventory optimization as could be seen earlier in this chapter and the chapter on theory and best practices. Inventory mapping can reduce inventories. Teams composed of employees of both Cargill and suppliers should decide on the criteria used for classification. After mechanical distribution products are classified, Cargill and suppliers should agree on consignment stock and VMI. This should result in less inventory carried by Cargill while the supply is guaranteed by consignment stock and supply conditions on behalf of the suppliers. The supplier should guarantee that over time their inventory level will be between a set minimum and maximum value (Valentino and Zavanella, 2003). Standardization of products and their coding is important for the mapping exercise and suppliers can support this too.

In the redesigned purchasing situation the supply base would be reduced to 2 or 3 suppliers for the MeDi category in The Netherlands. The leverage effect reduces prices and collaboration with the suppliers intensifies. Table 6 reflects the changes in the cost composition after reducing the supply base.

‘confidential’

Procurement of mechanical distribution products at Cargill July 2009 Author: Olivier Vegter Principal company: Cargill Supervisor: Harm-Jan Meijer University of Groningen 1