Supply Chain Optimization

(1)

Supply Chain Optimization

---a practical research for Gallagher Europe

May 2010-October 2010

Author: Yiwen Yang

Student number: S1938819

Faculty of Economics & Business

MSc Business Administration, specialization Operations & Supply Chains

University supervisors: Dr. X. Zhu, Prof. Dr. Ir. J. Slomp

Company: Gallagher Europe B.V.

(2)

Preface

This is my graduation thesis for the Master of Science in Business Administration, specialization in Operations and Supply Chains, at the University of Groningen. From May 2010 till the end of September 2010, I did the internship of doing a research to the supply chain of Gallagher Europe and its Chinese suppliers. This thesis was written during the internship of five months at Gallagher Europe. Here, I would like to thank all the colleagues who have helped me during the time I was in the company. Especially I would like to thank Ronald Wassink, Harry Boerstra, and Gerard v/d Bij. It is honored to have Ronald as my company supervisor, who led my thesis to a higher level with his critical view. Thank Harry and Gerard for taking their precious time to answering my questions and providing me valuable information. It was an unpredicted challenge to combine what you have learned from the book with real practice in the company. And I realize that this opportunity is really a unique experience which will be of use to my future career.

Further, I would like to thank my supervisors from the university: Xiang Zhu and Jannes Slomp. Thank Jannes introducing me to the company and giving me this great opportunity. Thank Xiang for the feedback meetings as very useful and helpful. Your both insights and guidance are indispensable for this research.

(3)

Management Summary

Gallagher Europe was established in the 70’s, and now has grown into an international commercial enterprise, with headquarter in Netherlands, two subsidiaries in United Kingdom and Switzerland, and sales offices in Belgium, France, Sweden, Austria, and Germany. The company has the authorization of European distribution of products with worldwide A-brand name Gallagher. In 2003, company A, of which core activities are a complete range of insulators and accessories for field electric fencing, was acquired by Gallagher Europe. From then, the company started their step of globalization and put their foot onto the land of Asia. The company purchases electronic fencing accessories in China. However, because of long geographic distance between Holland and China, it brings difficulties to make sure the supply chain in good control, compared with other European suppliers. How to keep the supply chain cost-efficient and effective becomes the problem for the management of the company.

Therefore, the purpose of this research is as follows:

Have a scan of company’s current supply chain, and see if there are opportunities for the optimization.

(4)

List of Abbreviations

CBM Cubic meter

ERP Enterprise Resource Planning

ESI Early Supplier Involvement

ETA Expected arrival date

ETD Expected delivery date

FCL Full-container loaded

GE Gallagher Europe

GGL Gallagher Group Ltd.

HR Human Resources

ICT Information and Communication Technologies

LCL Less-than container loaded

LP Loading port

PQCAM Procurement and Quality Control Assistant

Manager

PFR Product failure report

PT Process time QQ Query Quota RW Regional warehouse S Single supplier TT Transition time UP Unloading port

VSA Value stream analysis

VSM Value Stream Mapping

w/m Weight or measurement

(7)

1. Gallagher Europe

In this chapter, the company in which the project is conducted will be introduced. The paper will give a brief introduction about the company history, products, and organization structure.

1.1 The company profile

Veldman & Dijkstra, rooted in the 70’s, has grown into an international commercial enterprise Gallagher Europe (GE), with headquarter in Netherlands,two subsidiaries in United Kingdom and Switzerland, and sales offices in Belgium, France, Sweden, Austria, and Germany. The company has over 100 employees. Gallagher Europe is responsible for the European distribution of the Gallagher products. The worldwide A-brand Gallagher is core business of Gallagher Group Ltd. (GGL), which is established in Hamilton in New Zealand in 1938 and currently has 41% shareholder of Gallagher Europe.

Company A, which was established in 1947, initially manufactured ceramic insulators. In the ‘60s, the company was expanded to include the production of plastic fencing insulators. Present-day core activities of A are a complete range of insulators and accessories for field electric fencing. In 2003, A was acquired by GE. A supplies commodity accessories for low price and private label as requested by customers. Also, A supplies B-brand products.

1.2 The product range

Except the core products Gallagher, the leading quality brand in electric fencing, the products in B-brand with less quality and low price, and private brand, the company has other complimentary products, as shown in figure 1. The company has high product variety, even in the same product category, in order to meet customers’ different demand. The markets of all products are in Europe.

# Products

1333 941

169 667

(8)

1.3 Role and organization

There are two central warehouses in Netherlands, respectively for Gallagher products and A products. Two subsidiaries in UK and Switzerland have their own decentralized warehouse for Gallagher products. The company purchases the products worldwide and distribute to four kinds of dealers: specialist dealers, free trade dealers, cooperation dealers, and E-business dealers. Therefore, in the external supply chain context, the role of the company acts as the distributor, as shown in figure 2. The vision of the company for the future is to distribute related productlines through the same trade channel, and realize maximum availability through traditional and new distribution channels.

Figure 2: Overview of the role of the company

The organization chart of Gallagher Europe is illustrated by figure 3. The management team of the company consists of: two chief executive officers, Sales and Marketing Managers, Finance Manager, Sourcing Manager and Operations Manager. The Information and Communication Technologies (ICT) department and Human Resources (HR) department support the whole company. Purchase division and warehouse division are managed by the operations manager. Export means sales in Italy, Spain, Poland, Norway, Denmark, Island, Hungry, etc., who are in the charge of internal sales in Netherlands.

(9)

CEOs

ICT HR

Sales & Marketing

Finance Operations

Marketing Export A Gallagher Gallagher

(10)

2. Research design

In this chapter, research context, project intention and research objective will be given in the first place. Then the main research question is determined based on these three dimensions. Sub-questions are derived from the main question. At last, the conceptual research model will be defined and structure of the paper will be presented.

2.1 Project intention and problem statement

After defining the research context, the project intention will be explained. Afterwards, research objective and research questions will be determined.

2.1.1 Research context

The company has multiple suppliers who are located in different regions, countries, or even states. But the core business is electronic fencing products. And the main suppliers of the products are in China and New Zealand. Figure 4 illustrated the supply chain. However, China and New Zealand supply different products, although for the same market. Moreover, in the supply chain between New Zealand and Holland, all the production and transportation are scheduled and arranged by GGL. The whole supply chain is in charge of GGL. Therefore, the focus of this project would be on the supply chain between China and Holland, in which GE acts as the focal player.

Figure 4: The main supply chains of the company

In the current situation, GE purchases electronic fencing accessories in China and all products are shipped in 20- or 40-feet fully-loaded container from Shanghai or

Gallagher Europe (Distribution for Europe Market)

Sourcing in China _{GGL New Zealand}Sourcing from

Dealers Material flow

(11)

Ningbo port, and Qingdao port to Rotterdam port. After arriving in Rotterdam, the full container is transported to the central warehouse in Groningen. The warehouse arranges the shipment from inventory to customers. The overview of the supply chain is illustrated in the following figure 5.

Figure 5: Overview of the supply chain 2.1.2 Project intention

In the supply chain between China and Holland, GE is the focal company and responsible for demand planning, quality control and shipments arrangements. Because of the geographical location of Chinese suppliers, it brings difficulties in controlling the operation at some level. Therefore, the management of the company intends to have a scan of current supply chain, and see if there are opportunities for the optimization.

2.1.3 Research objective

As the project goal stated in last section, the objective of this research is to determine and analyze the processes that are involved in the supply chain. Furthermore it will explore the opportunities for process improvements, in order to optimize the supply chain.

2.1.4 Research questions

(12)

processes. Therefore, during this phase, the second sub question is formulated as: 2. What are the valuable factors that influence current processes?

In the final re-design phase, possible improvements on current situation are recommended. And the influences are expected. Hence, the third and the fourth sub questions comprise of:

3. What are the possible improvements on the processes?

4. What the impacts that the improvements would bring about?

2.2 Conceptual research model

After first-round interviews, relevant activities and the person who is in charge of are identified, as figure 6 illustrated. Thus, it is concluded that four key processes are incorporated in the supply chain: demand planning, ordering, quality management and transportation. Nevertheless, these processes are influenced by a couple of factors. In the diagnosis phase, details will be discussed based on conceptual model as shown in figure 7.

Figure 6: Activities in the supply chain

Time for demand planning, forecasting and ordering Time for confirming purchasing orders Time for quality control over production time Oversea shipping time Planner in NL Staff in China Staff in China Staff in China Shipper Time for managing the goods pick-up at the port and transportation to Groningen

(13)

Figure 7: Conceptual Research Model

2.3 Research Methodology

In this section, the methodology of how to answer the research questions will be introduced.

2.3.1 Data collection

According to Yin (2003), there are six distinguished data sources: direct observations, documentation, archival records, interviews, participant observation, and physical artifact. In this research paper, three research data collection methods will be used. All interviews are conducted on semi-structured basis and recorded. The data collected in this paper is characterized by both qualitative nature and quantitative nature.

 Sub Question 1: What processes are incorporated in the supply chain?

In answering this question, process flow chart was applied to describe the supply chain and processes incorporated. The mapping was based on the first-round interviews that were conducted with the participants who work at the management level, including operations manager, sourcing manager, purchasing manager, warehouse manager, and product manager. The question lists can be found in

Supply chain performance Demand Planning

Ordering Transportation arrangement Planner’s experience & knowledge Forecasting Product design Supplier quality management Market strategy

Shipping policy Shipping policy

Shipping cost Demand planning

(14)

second-round interviews at the operational level were conducted. The interviews involved two participants, the planner of purchasing department in Netherlands, and purchasing and quality control assistant manager (PQCAM) of sourcing department in China. The question lists can be found in Appendix 2. In addition, a number of company documents are used as a source of information, collected from such as spreadsheets, files, and catalogs.

 Sub Question 3: What are the possible improvements on the processes?

Based on the data, that are collected from company’s reports, and method suggested from academic background, ABC classification is applied. Accordingly, the improvements are recommended.

 Sub Question 4: What the impacts that the improvements would bring about? In order to answer this question, Value Stream Mapping (VSM) is used. The mapping was built based on the interviews with two key people: the planner and the PQCAM. The question list could be seen in Appendix 2. Because the mapping clearly illustrates how the supply chain be processed, for example, the time that the employee spends on individual process, it could be obviously come up with the conclusion of the impact that the improvements would bring on staff’s workflow. Hence, in section 5.4, the paper would discuss the influences of improvements on employees.

2.3.2 Data Processing

The most important step in analyzing qualitative data is labeling. Each part of the document or notes is labeled according to the relevance for further analysis. The interviews are processed into word transcripts and sent back to the interviewees for checking if any biases exist.

2.4 Report Structure

(15)

Figure 8: Research model (based on model of de Leeuw, 1996) Pre-Orientation Phase Orientation phase Diagnosis phase Redesign phase Conclusion Chapter 1 Company profile Chapter 3 Theoretical Background Chapter 4 Current processes within supply chain Chapter 5

Recommendations & Impacts

(16)

3. Theoretical background

In this chapter, the literature, concerning the supply chain, will be reviewed, which provides theoretical background for further empirical analysis. Firstly, in section 3.1, the definition of supply chain management will be given, and then four distinct supply chain business processes will be introduced. At this phase, the first sub question “What processes are incorporated in the supply chain?” would be answered from academic perspective. In order to answer the second sub question “What are the valuable factors that affect current processes?” objectively in Chapter 4, section 3.2 defined theoretical measurements for the supply chain performance.

3.1 Supply Chain Management

Given the definition of supply chain management, the business processes within the supply chain are identified and introduced.

3.1.1 Definition of supply chain management

In a typical supply chain, raw materials are procured at suppliers, and items are manufactured at one or more plants and shipped to the distribution center or directly to customers. In this respective, the definition of supply chain management by Simchi-Levi et al. (2008) is used as follows:

Supply chain management is a set of approaches utilized to efficiently integrate suppliers, manufacturers, warehouses, and stores, so that merchandise is produced and distributed at the right quantities, to the right locations, and at the right time, in order to minimize systemwide costs while satisfying service level requirements.

3.1.2 Supply Chain business processes  Value Stream Mapping (VSM)

(17)

5.4.

 Business Processes

Successful supply chain management now requires integrating activities into key supply chain process rather than managing individual functions (Serve et al., 2002). Stock & Lambert (2001) defined a process as “a structure of activities designed for action with a focus on end customers and on the dynamic management of flows involving products, information, cash, knowledge, and ideas.” In a supply chain, the number of business processes varies between respective parties. Therefore, it is important to identify the key processes in specific business operating context, and then manage them in order to achieve superior supply chain performance.

In this paper, considering the operations within the company and based on the information got from interviews, four key business processes are determined as demand planning, purchasing/ordering, quality management, and transportation. Thus, in the following, the literature concerning these four processes will be reviewed in order to get clear understanding from theoretical background.

Demand Planning

(18)

Figure 9: Linking forecasting and planning (Adapted from Makridakis et al., 1978) Wallace & Stahl (2008) defined five steps, as shown in figure 10, in the demand planning phase. Step A involves reviewing and investigating the causes of forecast variance, and finalizing the aggregate statistical forecasts. Since forecasting for new products are different from existing products, the new products forecasts are updated separately and early in the demand planning process. The third step projects future business demand. When external factors, such as price changes, promotions, new product launch plans and others, are considered, the related assumptions, for example, the expected behavior of customers or reaction of the market, are needed to be taken into account. In the final step, the forecast is determined and validates. The third step mentioned here could be thought of planner’s experience and knowledge, which is one of key factors that influence planning process and will be discussed in section 4.2.1.

Figure 10: The demand planning phase (Source: Wallace & Stahl, 2008) Purchasing/Ordering

Purchasing generally refers to the actual procuring materials, packing, finished-goods and those activities associated with the buying process (Stock & Lambert, 2001).

A. Forecast variance review and aggregation

C. Forecasting of New Order Volumes

D. Application of External Factors and Assumptions

E. Execution B. Addition/Updating of

(19)

Therefore, activities in purchasing phase are more process-oriented. Stock & Lambert (2001) concluded primary objectives of purchasing, which are summarized in the following figure 11. These objectives provide selective criteria for assessing current purchasing practice in section 4.2.2.

Figure 11: Purchasing Objectives

According to the objectives, a number of key performance measures are used to evaluate and manage purchasing activities. Supplier delivery, measured in terms of early, on-time, or late, guarantees the flow of materials. Procurement planning, including availability and price forecasting, is crucial for keeping and developing relationship with suppliers and also helps to achieve productive working relationship with other departments within the company. Inventory, in terms of inventory turnover and stock level, measures inventory cost and service level.

Quality Management

B. B. Flynn et al. (1994) defined quality management as an integrated approach to achieve and sustain high quality output, focusing on the maintenance and continuous improvement of processes and defect prevention at all levels and in all functions of the organization, in order to meet or exceed customer expectations. Quality management involves directing and controlling some factors. In the paper of P. G. Benson et al. (1991), eight critical factors of quality management, that are defined by Saraph, Benson, and Schroeder (1989), are deployed: top management support, role of the quality department, training, product/service design, supplier quality management, process management, quality data and reporting, and

Purchasing objective

(20)

Critical factors of quality management

Explanation of the Factor

Top management

support

The degree of participation by management in quality improvement efforts; Specificity of quality goals.

Training Quality-related training for all employees.

Product/Service design In-depth understanding of customer requirements; Clarity of specifications; Avoidance of frequent redesigns.

Supplier quality

management

Purchasing policy emphasizing quality rather than price; Supplier participation in quality control and product design.

Quality data and

reporting

Feedback of quality data employees and managers for problem solving; Timely quality measurement, Customer perceptions of quality.

Employee relations Implementation of employee involvement and quality circles; Responsibility of employees for quality; Effectiveness of supervision in handling quality issues. Table 1 Quality Management (Adapte from P. G. Benson et al., 1991)

Currently, with the desire to expand the market worldwide, many organizations tend to find the resource outside their country of origin, due to consideration of cost, speed to market, development strategy, etc.The globalization of production systems has added significantly to the challenges and risks of managing quality (Enderwick, 2008). In addition, it enhances the possibilities of process interrupts, which might due to the occurrence of product defect so that the processing time would be increased in order to fix the problem. Therefore, improved quality performance is a major factor in shrinking the profile of a process, because the poor quality directly represents wasted time---time spent in waiting, in rework, in extra inspections to prevent recurrence, in fixing problems, and the like (Bayazit & Karpak, 2007).

Transportation

(21)

Planning Procurement Manufacturing Distribution

Network and

asset

rationalization

Landed costs Interplant

movements

Load plans

Lead times Inbound

in-transit inventory management

JIT and other

specialized services

Pick lists

Vendor sourcing Reduced raw

material and work-in-process inventories Shipping documentation preparation Economic order quantity Dock scheduling Outbound shipment management Mode/carrier selection

Table 2 Areas in the supply chain affected by transportation (Stock & Lambert, 2001) Holter et al.(2008) defined several vital parameters in the process of transportation, including transport rates and related charges, transit time, transport visibility, on-time delivery, and cost of transport management. Roorda et al. (2010) advocated that transportation starts with mode selection, then consolidation, and next goes to loading or unloading. Shipment consolidation allows for exploiting the possibilities of reducing transportation cost due to economies of scale. It is suggested that the larger the overall shipment and longer the distance it is transported, the lower the transportation cost achieved (Bowersox & Closs, 1996).

(22)

to resource and expertise, companies tend to outsource logistics functions to more professional parties. The major parties are freight forwarder and third-party logistics service providers (3PL). The freight forwarder can provide faster and more complete service than a carrier, such as documentation services, which is especially vital for firms who have little international experiences (Stock & Lambert, 2001). 3PL can provide even more services than freight forwarder. In addition to transportation and distribution services, 3PL tend to provide more integrated logistics services. The typical services are inventory management, IT services, such as tracking and tracing, value added activities, such as secondary assembly and installation of products (GÖl & Çatay, 2007).

3.2 Supply Chain performance

In this section, the conflicts, which would encounter during optimizing supply chain performance, are explained. Next, from the academic background, the indicators to measure the performance are introduced.

3.2.1 Trade-offs within the supply chain

As the definition of supply chain management indicated in section 3.1.1, the supply chain management now moves toward global optimization, and, as a result, system wide cost would be reduced. Therefore, the major challenge in supply chain management is to coordinate supply chain activities so as to maximize supply chain performance (Simchi-Levi et al., 2008). However, there exist conflicting objectives for different stages in the supply chain. Moreover, these conflicts are inevitable when coordinate different activities. Simchi-Levi et al. (2008) summarized five trade-offs when managing the supply chain. The paper in the following chapters will discuss one of trade-offs under circumstance of GE.

 The lot size-inventory trade-off

Concerning per-unit setup costs and control of processes, manufacturers like to produce in large lot size, while in the meantime, the inventory level would be increased.

 The inventory-transportation cost trade-off

The effect of economies of scale emerges when the delivery is always made in full truckloads. However, transportation cost reduction might be overwhelmed by higher inventory costs. On manufacture side, items have to wait till the full truckloads delivery could be made. Buyers have to hold high stock in order to cover the demand during the longer periods of delivery. On the other hand, buyers might order more items than needed in order to reduce transportation cost reduction due to economies of scale.

 The lead time-transportation cost trade-off

(23)

 The product variety-inventory trade-off

High product variety increases the complexity of manufacturing, leading to manufacturing cost rises and efficiency decreases. On the other hand, product variety brings about the difficulty on accurately forecasting the demand for each item, more stock need to be hold in order to ensure the customer demand could be satisfied and maintained.

 The cost-customer service trade-off

All the trade-offs mentioned above are examples of the cost-customer service trade-off. Evidently, reduction on inventories, manufacturing costs, and transportation costs is achieved at the expense of customer service.

3.2.2 Performance measurements

Stock & Lambert (2001) classified decision-making in the supply chain into three levels; strategic level; tactical level and operational level. Similarly, Gunasekaran et al. (2004) defined the metrics for supply chain performance at strategic, tactical and operational to clarify the appropriate level of management authority and responsibility for performance. In his framework, as showed in the following table 3, four key processes are considered: plan, source, make and deliver. The framework provides criteria for assessing the processes in GE case in Chapter 5.

Key processes

Strategic Tactical Operational

Plan Level of customer

perceived value of product; Variances against budget; Order lead time

Accuracy of

forecasting

techniques; Human resource productivity

Order entry methods;

Human resource

productivity

Source Supplier delivery

performance

Procurement planning

Make Range of products

and services Percentage of defects Percentage of Defects; Human resource productivity index Deliver Flexibility of service

system to meet customer needs

Delivery reliability performance

(24)

(25)

4. Current processes within the supply chain

In this chapter, the key processes in current supply chain will be defined and described. In section 4.1, the mapping of overall processes will be conducted in order to have an overview of the processes. The first sub question “What processes are incorporated in the supply chain?” is answered. In section 4.2, detailed analyses are given, in order to answer the second sub question “What are valuable factors that affect current processes?” which would be summarized in section 4.3.

4.1 Process mapping

Process mapping, also known as value stream analysis (VSA), makes business processes transparent. With “swim lane process mapping”, a type of cross-functional process mapping, and defined symbols, business processes can be clearly mapped and viewed on how the processes actually run can be achieved (Wiegand & Nutz, 2008).

The supply chain between GE and China consists of four key business processes: demand planning, purchasing/ordering, quality control over production time, and transportation. Figure 12 illustrates the flowchart of overall processes within the supply chain, where processes are separated by dotted lines and each process includes one or more activities.

4.2 In-depth Analysis

In this section, the processes in the supply chain of GE and China would be analyzed in depth.

4.2.1 Demand planning

(26)

Demand Planning

Placing purchasing orders

Approval from Purchasing Manager

Send orders to PQCAM in China Quality Control on raw materilas Quality Control on production Quality Control on packing

Update the forecast

File By system Yes Confirmation of Orders with suppliers On site On site On site Make transportation plan Yes By Mail No Planning & forecasting Ordering Production Prepare documentation Supervisor loading onto the

container Inform planner in NL by sending copies of documentation Yes On site By Mail Handling to & in the port of departure in China Send original documentation to planner in NL By express Transport to Rotterdam Check warehouse schedule

Contact broker for pick-up at the port

Handling & transport to Groningen Receiving in warehouse Yes Yes No No Transportation By mail

(27)

Figure 13: Implementation of supply chain related information system

Figure 14: Supply chain related information system used in the company Once a month, the system Slim4 refreshes the data base and comes up with new calculation, since there is one more month historical sales data available. The planner

BAAN

Warehouse Finance Sales &

Marketing

Slim4

Purchasing

Planner Slim4 BAAN Internal

sales Customer order Sales Data Stock level; Demand pattern; Average demand; Ordering advice (when & how many).

(28)

check. If the planner confirms that they unfit with the demand which is expected before and cannot be covered by current open orders, the replenishment ordering will be planned.

Therefore, at this phase, a number of factors influence the process of demand planning, as figure 15 indicated. The decision of planning ordering is based on forecast and planner’s own experience and knowledge. Since the operation of the company is under the principle of make-to-stock, the forecast is made based on current stock level and projected future demand. Besides the forecast that is produced by the system, judgmental adjustments made by planner’s experience and knowledge also take importance in the planning process. In order to incorporate the effects of information at future predictions, such as new product launches, promotions, etc., which is neglected when using statistical forecasting in the planning process, the forecaster is required to judgmentally adjust statistical forecasts (Eroglu & knemeyer, 2010).

Figure 15: Factors that impact the demand planning process

As introduced in section which discussed theoretical background, making forecast is one key activity in planning process. Forecast is one of major reasons in creating stock because the orders are placed according to the forecast. Additionally, forecast is one of effective tools to help suppliers get prepared for the coming production. So, if the accuracy of the forecast is low, it is likely to bring unexpected inventory and disturbance to the production. Currently, the company has two figures about the stock: estimated stock and realized stock level.

The company sets the formula for calculating estimated stock end month and realized stock level as follow:

Estimated stock end month= stock level at estimating time + open purchase order + forecast purchase order + unexpected - estimated turnover

Demand Planning Stock Level Planner’s experience & knowledge Forecasting

(29)

Realized stock level end month= stock level at estimating time + realized purchase – realized turnover

In which,

Open purchase order means the order being placed but goods not arrived yet; Forecast purchase order means ordering advices that are given by the system; Unexpected is the quantity estimated by the manager to cover the quantity that the planner made as manual adjustments; so, the amount that forecast purchase order plus unexpected is supposed to equal the amount that the planner made on the forecast;

Estimated turnover means expected sales;

Realized purchase means open purchase order at estimating time plus actual orders placed;

Realized turnover means actual sales.

If subtract Estimated stock end month from Realized stock level, the formula goes to = (actual orders – forecast purchase orders – unexpected) – (realized turnover – estimated turnover), indicating the difference between forecasted stock and real stock.

In current practice, forecast determines the amount of (actual orders – forecast purchase orders – unexpected), and market demand decides the turnover. Therefore, referring to above formula, it could be concluded that there are two factors---forecast and market demand---that determine the stock level. However, forecast outweighs market demand due to unpredictability of demand. Furthermore, in order to avoid stock-out which might result in sales lost, it is likely that the planner always orders more than that suggested by the system. If the planner exaggerates the demand much more, the stock would be significantly increased. The figure 16 and 17 presents the inventory level against demand in the year of 2009-2010, indicating that there are much more products than needed.

(30)

Demand-Inventory (A) 0 500000 1000000 1500000 2000000 Sep-09 Oct-09 Nov-09 Dec-09 Jan-10 Feb-10 Mar-10 Apr-10 May-10 Jun-10 Time A m ou nt _Demand Inventory

Figure 17: Demand for A products against stock in the warehouse

Basically, the forecast is sent to two key suppliers four times a year and twice a year for other suppliers. However, in real practice, it depends on if there are big changes happen, which are called exceptions. When the exceptions emerge, the planner needs to check if the forecast needs to be updated, regarding to the predicted demand pattern that is calculated by the system. Otherwise, the planner would not make new forecast and send to PQCAM in China again.

The following three figures 18, 19, and 20 demonstrate the differences between actual orders and forecasts of two brand products from two key suppliers, in which big variance could be observed, i.e. circles in the figures. In other words, the forecast accuracy is relatively low. There are three reasons which could explain this issue on some level:

1) Sudden market demand increased or decreased; 2) Planner overestimated or underestimated demand;

3) In practice, the planner might aggregate the amount, which shows in the forecast is the amount for several months, and place the order at one time, in order to reduce the ordering cost. In other words, the way of the planner placing orders is not in line with what the forecast showed.

0 30000 60000 90000 120000 150000 180000 210000 240000 270000 300000 330000 360000 390000 420000 450000 Jul-09 Aug-09 Sep-09 Oct-09 Nov-09 Dec-09 Jan-10 Feb-10 Mar-10 Apr-10 May-10 Jun-10 Actual orders

Forecast made in October Forecast made in February

(31)

0 2000 4000 6000 8000 10000 12000 14000 16000 18000 20000 Jul-09 Aug-09 Sep-09 Oct-09 Nov-09 Dec-09 Jan-10 Feb-10 Mar-10 Apr-10 May-10 Jun-10 Actual orders

Forecast made in August Forecast made in October Forecast made in December Forecast made in Febrary Forecast made in April

Figure 19: GE products from supplier B

0 100000 200000 300000 400000 500000 600000 700000 800000 900000 1000000 1100000 1200000 1300000 1400000 Jul-09 Aug-09 Sep-09 Oct-09 Nov-09 Dec-09 Jan-10 Feb-10 Mar-10 Apr-10 May-10 Jun-10

Actual orders for all products Forecast made in August Forecast made in October

Figure 20: A products from supplier A 4.2.2 Purchasing/ Ordering

Since in the planning phase, the schedule of the ordering has been made, it is sort of easy for the planner to place purchasing orders. The planner places the orders through the BAAN. The orders will be printed out and handed to purchasing manager for the approval. After permission, the orders would be sent to PQCAM in China by e-mail.

(32)

ordering:

I. Placing orders onto Chinese suppliers

After checking historical purchasing orders, it is found that, sometimes, the orders would be revised once or twice due to incorrect price, quantity, and etc. At the most, it took more than one month to make the final agreed orders.

II. Decision on consolidation shipment

At this phase, the schedule of the coming full container load shipment will be decided. The policy of consolidation shipment is time-based, which accumulated loads according to the requested delivery date on the purchasing orders. After receiving the orders, the shipping time and the size of the container are determined. The exact goods to be shipped are identified under the discussion between the planner and PQCAM. There are certain rules for the decision-making:

A. With respect to the requested receiving date on the orders, the shipping time is scheduled accordingly.

B. If the goods, that are required on the orders, could fill up a full container, 20ft or 40ft, the shipment is scheduled accordingly.

C. If not, there would have two situations:

a) If the goods are more than filling up full 20ft container, but far away from the volume for 40ft, some orders or some goods need to be delayed to next shipment. The planner would check the stock level again, find the product whose stock level could cover the demand until next shipment, and delay the delivery of these products this time.

b) If the products are less than the volume for 20ft container, planner would add orders for the products with the earliest advising ordering date.

D. Sometimes, after loading the required products, there is limited volume left for other products. In this case, when the planner decides which products to be added, the volume of the product becomes the only consideration.

Figure 21: Factors influence ordering process

(33)

4.2.3 Quality control over production time

Stitt (2004) concluded that quality control depends on the acquisition of consistent ingredients, environmental controls in the plant, recipe adherence, and machine operation. Therefore, quality control is of importance within the supply chain. In order to avoid product defect, GE starts quality control along with the production, ranging from raw material purchasing, semi-finished products production, finished products production to the final stage of packing. The control consists of mould supervision, machine maintenance, additive selection and floor shop operations. Due to specific requirement for distinct production process, suppliers outsource certain production processes, i.e. gilding, punching, bending, etc., to other outer partner manufacturers. Therefore, it requires extra effort and time for PQCAM to check the quality of semi-finished products on these 2nd suppliers’ sites. Moreover, there is lack of good quality management within suppliers’ internal organization, which brings more workload of quality control onto PQCAM.

In Table 1 in Chapter 3, six critical factors, that impact on quality management, are listed: top management support, training, product design, supplier quality management, quality data and reporting, and employee relations. In the case of GE, in addition to two factors, product design and supplier quality management, there is another factor that influences quality management, as summarized by figure 22. In light of marketing strategy, there is strict requirement on packing, i.e. labeling. Not only does PQCAM need to give the training and the instructions to the shop floor operators, but also supervision on site is required. In Chinese supply chain, more than 100 different types of products are manufactured. High product variety complex product designs and further affects quality control. Whether product specification is clarified in product design stage decides the operation of quality management.

(34)

2. The supplier side has poor quality management. That means PQCAM has to take over the responsibility and spent time on supervising shop-floor work. 4.2.4 Transportation

I) Transportation out of China

Due to long distance and time spent, it is vital to arrange the transportation efficiently and cost-effectively. At this phase, transportation planning is very important. From the cost perspective, mode of transportation is decided as ocean shipping. And full-container loading (FCL) is required due to economic scale effect. If the products, which are produced in different suppliers, are of little amount and cannot fill in individual full container, they will be shipped to the supplier who has larger amount of outgoing goods, usually the key suppliers in Ningbo. The suppliers arrange the shipments to the consolidation site and then the products are combined together to fill in the container. On the other hand, transportation means determines the time when to arrange the transportation. The following figure 23 presents factors that influence the process of transportation arrangement.

Figure 23: Factors that affect transportation planning

Currently, there are three forwarders who are responsible for exporting products from China to Holland. One is in charge of the shipments of suppliers in Qingdao, and the other two are responsible for the shipments out of Ningbo, including one is in charge of booking ship space and the other one is responsible for customs clearance and truck transportation to the departure port.

The suppliers prepare the documentation for customs clearance and oversea shipping. However, it takes quite amount of time for PQCAM to check these documentation because of poor job done by the suppliers. The documentation needed to be adjusted again and again until the information is correct and complete. Additionally, it is crucial to have the documentation ready on time. The documentation, such as customs declaration, need to be submitted by requested date. Otherwise, the goods cannot be shipped as scheduled, which will bring about devastating consequence.

Transportation arrangements

Shipping policy

Shipping time Shipping cost

(35)

Moreover, PQCAM spent quite a lot time in coordinating with the forwarders. The selection of forwarder and shipping company is based on price consideration.When booking the ship space, the shipping company would be chosen based on the shipping cost and shipping date. Then, the truck transportation is arranged to ship the goods from the factory to the port of departure. At this step, PQCAM monitors the goods loading onto the container on site. It is important to load the goods in a good way in order to avoid any damage to the package during the shipment.

II) Transportation from Rotterdam to Groningen

Right after goods are loaded onto the container, copies of documentation are firstly sent to the planner and inform the expected delivery date (ETD) and expected arrival date (ETA). The original files are sent to Holland by express, three weeks prior to the shipment arrival date. One week before the ship reaches the Rotterdam port, the planner sends the original documentation, requested modes of transportation and arrival date in the warehouse to the logistics forwarder in Holland. Within 2 or 3 days, the forwarder would send back the confirmation of the exact picking-up date and arrival time in the warehouse. Regarding the selection of modes of transportation, it all depends on the quantity, time and cost. There are two choices: truck and train. However, truck is always a better choice because the train takes longer transportation time and requires more unloading cost, although the transportation fee is cheaper.

In addition, there is a schedule of receiving goods, which is shared between the warehouse and the purchasing department, but only for the schedule of large volume products. The warehouse could always handle the small shipment without any notice in advance. For the large shipments from China, after receive the copies of documentation, the planner firstly needs to check the schedule to see if it is possible for the warehouse to receive the goods in expected arrival date, and then contacts the forwarder to confirm the exact arrival time. If everything is fine, the planner would fill in the schedule and block this period of time for warehouse to get prepared and hire extra labor for unloading and receiving goods. In this way, other planners could not book any other warehouse activities.

4.3 Summary

(36)

ordering and quality control, would be given as well.

Figure 24: Cause-Effect Diagram

(37)

5. Recommendations & Impacts

With regard to the analysis in last chapter, the opportunities for improvements on each process would be recommended in this chapter. The third sub question “What are possible improvements on the processes?” and the fourth sub question “What the impacts that the improvements would bring about?” would be answered accordingly.

5.1 Recommendations on inventory management

As summarized in section 4.3, in the case of GE, stock is one of key factors that influence the operation of supply chain. San-Jose et al., (2009) advocated that inventory is required since in real-world situation of the supply and demand are never perfect, but it must be determined very carefully for two reasons: to meet the demand to maximize sales and to minimize the total inventory cost. ABC classification system, as a tool to identify the importance for different items, is effective inventory management to optimize the stock without lowering the service level. As analyzed in section 4.2.1, forecast affects stock level, which impacts the total inventory cost in the end as well. Therefore, the recommendations are given on implementation of ABC classification and improving forecasting.

 ABC classification

The system classifies the item into three categories: A, B, and C (Paul, 2007)

 A class: composed of 20% of the items, which account for 75% of total turnover. This group of items is considered as outstandingly important.  B class: made up of 30% to 40% of the items, which comprise approximately

15% of total turnover. This category is thought as average importance.

 C class: the remaining items, which make up a large part of the product range, 40% to 50% of the items, but only account for approximately 10% of total turnover. This class is ranked as relatively unimportant.

(38)

calculated by referring to current order frequency that is determined by supplier basis. It is obviously seen that in current situation, much more items are classified in A class.

ABC classfication (A products)

0 10 20 30 40 50 A B C N um ber o f it em s Current Academic

Figure 25: A products classified as A, B, and C

ABC classification (GE products)

0 10 20 30 40 50 A B C N um ber o f it em s Current Academic

Figure 26: GE products classified as A, B, and C

(39)

current ordering cost €2300 (2300*81.02%).

Weighing of stock value of products from Chinese suppliers in total stock value (A products)

€ 0.00 € 100,000.00 € 200,000.00 € 300,000.00 € 400,000.00 € 500,000.00 € 600,000.00 € 700,000.00 € 800,000.00 Sep-09 Oct-09 Nov-09 Dec-09 Jan-10 Feb-10 Mar-10 Apr-10 May-10 Jun-10 Weighing of stock value of products from Chinese

suppliers in total stock value (A products) Time 0.00% 10.00% 20.00% 30.00% 40.00% 50.00% 60.00% 70.00%

Value of stock from Chinese suppliers Value of total stock

weighing

Figure 27: Weighing of stock value (A products)  Accuracy of forecasting

Given the circumstance that the operation in GE is make-to-stock, forecast is very important. This is already being explained by the formula that is used in GE’s stock report to calculate the stock level, which is introduced in section 4.2.1. Therefore, forecast accuracy has significant effect on the inventory. Increasing the accuracy helps to control the stock level. However, current accuracy of forecasting for Chinese suppliers are relatively low, and the reasons are given in section 4.2.1 as well. Although it is difficult to predict the market demand precisely, and there is little thing to do with statistical forecast that is done by the computer system, there are some measures to help the planner make more accurate manual adjustments.

I) Timely communication with sales persons

(40)

increase the accuracy of forecasting. III) Frequency of making forecast

From the records of last year, even for the key suppliers, the forecast is not made on fixed schedule. First of all, it is not helpful in assessing the accuracy of forecasting objectively. Secondly, it is not useful for suppliers to make good preparation for the coming production, if there is long time and even big variance between each forecast. After applying ABC classification to identify the importance of the items, the frequency of making forecast could be determined accordingly. For A items, since it is suggested to make frequent orders, the forecast should be made more frequently than for the items in other two classes.

IV) Using performance indicators of forecast accuracy

In the planning process, accuracy of forecasting is an important performance indicator. But firstly, it is suggested to maintain the consistency of the way of making forecast and placing actual orders. As explained in section 4.2.1, if the planner plans to aggregate order amount together, it should be indicated in the forecast as well. Currently, GGL calculates GE’s accuracy of forecasting for the products they buy. However, considering different products and different market demand, the accuracy got from GGL cannot be applied to assess the forecast for Chinese suppliers. It is better to have separate record of accuracy of forecasting, but with the same calculation method. Keep track of accuracy of forecasting helps to check if the manual adjustments are in rational range, which influence the inventory level extensively.

5.2 Recommendations on Transportation planning

Transportation is another key factor that brings about considerable influences on current processes. On the one hand, shipping policy of full-container loading (FCL) impacts the transportation cost and inventory level. On the other hand, PQCAM spends certain level of time on arranging transportation, such as checking documentation, communicating with forwarders, supervising loading onto the truck, etc. Although the situation of current transportation is satisfying that shipments could meet the requested delivery time, the possibilities of improvements are explored in this section.

 Shipping policy

Within the five logistics network configurations that are advocated by Creazza et al. (2010), two of them could be applied in the case of GE:

(41)

Figure 28: Logistics network configuration 1

2. Configuration 2 (Figure 29) ---Direct shipment with groupage containerized (less than container load (LCL)): non containerized goods are shipped by means of road haulage to freight forwarder’s facility located near the loading port (LP) for container consolidation; after the ocean shipping leg, the container is deconsolidated at the unloading port (UP) and non containerized goods are hauled to one or more regional warehouses (RW).

Figure 29: Logistics network configuration 2

Currently, in order to pursue economic scale, GE ships all the goods under FCL. However, in section 5.1, the products are classified into A, B, and C categories. Since it is suggested that C items should be ordered in fewer frequency but in large quantity at one time, it is economic to ship C items in FCL. But for items in A class, order frequently with small batches is recommended. Therefore, in order to reduce the stock level due to the large amount resulted from the policy of FCL, the possibility of shipping A items or B items under LCL is taken into account, especially when carrier provides discount for LCL shipment.

(42)

is based on the volume of products). If shipped from Ningbo or Qingdao, it is priced by per CBM (cubic meter). And the road transportation in Netherlands is priced concerning both weight and distance. The parameters are listed in the following table 4, 5, and 6. The size of a general 20ft container is 26cbm, weighing 17.5t. However, the biggest volume for LCL shipment from Qingdao or Ningbo is allowed at 20cbm. So, under the first scenario of 20ft container shipment, just cost of LCL shipment from Shanghai port is calculated. The approximate result under scenario I could be seen in table 7 (The detailed calculation is in Appendix 3). The exchange rate between USD and Euro is 1 USD = 0.7753 Euro.

Type of container & Departure port Rate (approximate USD/ container)

20ft (departure port Shanghai) 1845

20ft (departure port Qingdao) 1870

20ft (departure port Ningbo) 1895

Table 4: FCL Seafreight International Security Port Surcharge Document per Bill of Landing Terminal handling cost Road transportation to Groningen Decla-ration rate Customs formaliti es 20ft 30 32.5 165 335 52.5 5

Table 5: Other expenses for FCL shipment (€/ container) Discharge rate International

Security Port Surcharge Other Port charges Terminal handling cost (w/m) Declaration rate Administr ation & communic ation tariff 60 (up to 2.5 x (400 kg p / cbm) per 1000kg) 6.5 56 17.5 50 25 24 (measuring 2.5 x (-400kg p / cbm) per cbm) 6.5 56 17.5 50 25

Table 6: Other expenses for LCL shipment (€)

Modes 20ft container(€)

FCL (Departure port at Shanghai) 2050.43

FCL (Departure port at Qingdao) 2069.81

FCL (Departure port at Ningbo) 2089.19

LCL (Departure port at Shanghai) 2456.34

Table 7: Scenario I---Shipping cost of 20ft container in FCL and LCL

(43)

shipping way of FCL is much cheaper than LCL. In addition, although FCL shipped from Shanghai port is cheaper than from Ningbo port, the transportation cost from supplier’s factory to Shanghai port is more expensive than to Ningbo port, because the suppliers are located in Ningbo.

If the products could fill with one 20ft container, but far away from filling up a 40ft container, there comes another two scenarios. (Notes: one 20ft container maximally could hold 20 pallets. And each pallet is roughly considered as 1.1*1.1*1.1 cbm.) II) Assume just one more pallet of items left (The detailed calculation is in

Appendix 3).

Modes One pallet (€)

LCL (Departure port at Qingdao) 400.13

LCL (Departure port at Ningbo) 397.47

Table 8: Shipping cost of one pallet under LCL

III) Assume ten pallets of items left (The detailed calculation is in Appendix 3).

Modes Ten pallets (€)

LCL (Departure port at Qingdao) 957.74

LCL (Departure port at Ningbo) 931.94

Table 9: Shipping cost of ten pallets under LCL

From table 8, it could be seen that it is uneconomical to ship one pallet by LCL. So, the products for loaded onto the pallet are suggested to wait for next shipment, together with other goods. In addition, concluded from table 8 & 9, along with the number of pallets increased, the cost of shipping one pallet is decreased. Seen from the price sheet provided by the forwarder, with departure port at Qingdao or Ningbo, the larger CBM the products have, the bigger discount provided. But the shipment from Shanghai port does not have such discount policy. Therefore, in the last scenario, the shipping costs, departed from Qingdao port and Ningbo port, are calculated based on the biggest discount. The result is presented in table 10.

(44)

with table 7, if extra charges amount to €500, FCL shipment is still the best way of shipping. Otherwise, if the amount of products is in the range between 10 pallets and 15 pallets, LCL shipment could be a choice.

To be concluded, the shipment decision between FCL and LCL needs careful consideration. On the one hand, FCL could reduce the transportation cost, but bring the risk of increasing inventory. On the other hand, LCL could relieve the stock level at some level, but might add up the transportation cost.

 The third-party Logistics Provider (3PL)

Considering that PQCAM spent a lot of time on communicating with forwarders and checking the documentation, it might be an option to turn to a 3PL, who takes over all logistical work. In addition, when cooperate with a 3PL, it is likely to have reliable and long-term relationship with the carrier. It is not good to switch forwarder frequently since it takes a while for both the staff and the forwarder to adapt each other’s way of working. Moreover, keeping the relationship as a partner encourages the carrier to provide more customer-oriented services. In addition, if LCL shipment is taken into account in the future, it is more convenient to let 3PL take charge of everything. In light of these reasons, finding 3PL becomes good choice. The time that the staff spent on arranging transportation could be used to do other activities, such as finding new suppliers, exploring Chinese market opportunities, and etc.

However, 3PL selection is of importance and involves multiple criteria. GÖl & Çatay (2007) used an analytic hierarchy process to select a logistics service provider for an automotive company. In the paper, total criteria are depicted in the decision hierarchy. Referring to the case of GE, 3PL selection criteria are adjusted in the following table 11.

Criteria Explanations

Price Competitive pricing compared with counterparts

General reputation Image and general reputation in 3PL industry Health financial situation Liquidity of company’s cash flow and profitability Experiences Provider’s faming products industry expertise, and

business in an international scope

Location The location of offices, or branches, or warehouses, or

distribution centers

Capabilities The percentage of the asset ownership, availability of qualified talent, customer-oriented service, and supply chain vision

Information systems Computer systems used for tracking and tracing

Responsiveness Ability to take actions to unforeseen problems and unexpected events

Key performance indicator measurement

Delivery reliability rate

(45)

5.3 Recommendations on ordering and quality control

In addition to controlling inventory and managing transportation that helps to optimize the supply chain, there are other aspects that could improve the process operation. In this section, opportunities of improving the other two processes---ordering and quality control---are suggested.

 Ordering

As mentioned in section 4.2.2, sometimes, it took more than two times to the final correct purchasing orders being placed. Much time is spent on discussion by emails between PQCAM and the planner about the price. There are two reasons:

1) The incorrect information in the system

The price of the products would be negotiated and might adjusted twice a year. If the planner is not informed to update the price in the system timely, the orders generated by the system are still with old price.

2) Foreign exchange rate

The company buys certain products in RMB. But in the system, the price is calculated in euro. That means there exist foreign exchange rate and price becomes flexible. But now, decision is made that the exchange rate should refer to the figure on the date that the orders are placed.

So, the possibility of improving ordering process is in strengthening correctness of information in the system. As long as the new price is confirmed, the agreed price sheet should be shared immediately and the system is required to update timely. In this way, for both the planner and PQCAM, the time spent on re-placing orders and queries due to incorrect information could be eliminated.

 Quality Control

Huang & Nguyen (2009) advocated that attention to sound quality inspection practices are among the key ingredients to sourcing success in China. So far, there are two ways for GE to confirm the quality of products, internal inspection and external customer feedback. One day before the goods are ready to be shipped, PQCAM took random quality inspection. Product Failure Report (PFR), based on customers’ complaints, is used as a way of externally product quality control. The customer complaints on products sourced from China are summarized in table 12.

(46)

Although the number of complaints for Chinese-supplied products is decreased in the year 2008-2009 and 2009-2010, the percentage of total number of complaints is reduced only by 0.04%. This means that quality management still should be paid attention. As the problem description issued in the report, most causes are due to the design of products. The solution taken is to change the module or solve with suppliers. From the interviews, it is told that the product design is completely in the control of GE. The suppliers are just responsible for job of manufacturing. At the product development stage, suppliers make the sample under the instruction of bill of materials and drawings of products, which are designed by GE. Until the sample passed testing, the products are allowed to put into mass production. However, sometimes, there are conflicts between real manufacturing and design engineering. The person, who works on the shop-floor, knows more about material, machine, and production technique. As a result, one of recommendations on quality control is Early Supplier Involvement (ESI) from the beginning of product development process. One of major advantages of ESI is early identification of potential problems, thus improving the quality of the product, eliminating rework and reducing costs (McIvor & Humphreys, 2004). The product department has expertise in designing, while the suppliers have experiences in real manufacturing. Working with suppliers can speed up product development time significantly. Wynstra et al.(2001), support that supplier involvement improve product development in terms of effectiveness, by mobilizing and leveraging supplier expertise regarding Design for Manufacturing (DFM), the quality and reliability of component designs, alternative materials and possibilities for component standardization. It is suggested that from the very beginning of design phase, before the sample is made, the supplier offers the opinions from the perspective of shop-floor operation. Clear instruction of product profile, function, production, and quality-checking criteria should be settled in the hands of both GE and suppliers.

(47)

Responsibilities Importance

Administration 25%

Communication between the company and Chinese partners

25%

Quality Management 25%

Supplier relationship development and management 25%

Table 13 Job-description of PQCAM

5.4 Impacts of improvements on staff’s workflow

In this section, value stream mapping is applied to respectively depict the workflow of two key staff in the supply chain of Holland and China. In conclusion, if recommendations, that are mentioned above, are taken, the time highlighted in the circle in the following figures 30 and 31 could be reduced significantly or eliminated. As shown in figure 30, within the supply chain, the planner in Groningen is involved in three processes: demand planning, ordering, and arrangement of transportation from Rotterdam to Groningen. Originally, the total workflow time is the time that transition time (TT) added by process time (PT) and Query Quota (QQ), 43.528 days. However, if the system could generate an order with correct information, such as correct price and quantity, the time of the planner spent on placing repetitive orders could be eliminated. The time could be reduced by 3 days.

Supply Chain Optimization