Master thesis Technology Management A business case for the improvement of the industrial performance of Draka Amsterdam Ing. E.M. Gardebroek

(1)

Master thesis

Technology Management

A business case for the improvement of the industrial

performance of Draka Amsterdam

Ing. E.M. Gardebroek

University of Groningen

Faculty of Economics and Business

(2)

Introduction

Page 1

Master Thesis Ing. E.M. Gardebroek, Januari 2013

Master Thesis

A business case for the improvement of the industrial performance of Draka Amsterdam.

Company: Prysmian Draka Amsterdam

Author: Ing. E.M. Gardebroek

Student number: s2073102

Study: MSc. Technology Management

Faculty: Economics and Business

University: Rijksuniversiteit Groningen

Company supervisor: C. Rossenaar (Plant Director Prysmian Draka Amsterdam) University supervisor: Drs. Ing. H.L. Faber

University Co-assessor: Prof. Dr. I.F.A. Vis

(3)

Introduction

Page 2

(4)

Introduction

Page 3

V

OORWOORD

Op maandag 23 juli 2012 werd ik wat eerder wakker dan normaal. Ik weet nog dat ik een erg leuk weekend had gehad en dat ik er op dat moment veel zin in had om weer aan de slag te gaan bij Draka Amsterdam. Even later belde mijn moeder mij op met het onbegrijpelijke feit dat mijn vader was overleden… Even stond de tijd stil. Even later stortte mijn wereld in.

Het schrijven van mijn Master scriptie zag ik voorheen als “de kroon op mijn lange studieloopbaan” waar ik erg naar uit keek, maar helaas zal ik er altijd met veel verdriet op terugkijken. Het feit dat ik mijn vader – in de “fysieke wereld” – nooit meer trots kon maken deed afbraak aan mijn motivatie en voor een paar maanden had mijn verdriet de overhand op ieder geluk in mijn leven. Het schrijven van mijn onderzoek verliep moeizaam, maar dankzij de steun, het medeleven en de vriendelijkheid van al mijn collega’s binnen Draka Amsterdam en dhr. Faber van de Rijksuniversiteit Groningen heb ik het uiteindelijk toch tot een goed einde kunnen brengen.

Graag wil ik al mijn collega’s binnen Draka Amsterdam bedanken voor hun steun en hulp gedurende mijn onderzoek. Een speciaal dankwoord gaat uit naar dhr. Rossenaar. De manier waarop hij Draka Amsterdam in deze huidige crisis aanstuurt is voorbeeldig en de manier waarop hij mij heeft gesteund tijdens mijn tijd van verdriet was uniek.

Tevens wil ik dhr. Faber en mevr. Vis bedanken voor het feit dat zij vanuit de Rijksuniversiteit Groningen mijn onderzoek wilden begeleiden.

Dhr. Faber heeft naast zijn uitmuntende begeleiding tevens getoond een zeer meelevend en zeer vriendelijk mens te zijn. Daarom gaat ook naar hem een speciaal dankwoord uit.

Naast de betrokkenen bij dit onderzoek, wil ik ook mijn familie en vrienden bedanken voor al hun steun en liefde. Het verwerken van dit verlies hoef ik niet alleen te doen, maar is een proces wat wij gezamenlijk zullen doorstaan.

Tot slot wil ik mijn vader bedanken voor alles wat ik heb bereikt en zal bereiken in mijn leven. Hij was mijn bron van motivatie en zijn trots was mijn grootste beloning voor al mijn prestaties. Pappa, ik zal jou en je trotsheid altijd blijven missen, maar ik zal blijven presteren.

(5)

Introduction

Page 4

CONFIDENTIAL NOTICE

(6)

Introduction

Page 5

M

ANAGEMENT SUMMARY

This Master Thesis Research is done within Draka Amsterdam, a cable manufacturing factory which is since 2011 part of The Prysmian Group. The research is done in the period from May 2012 to December 2012. Draka Amsterdam suffers from the current economic crisis because the Dutch building sector, to which Draka Amsterdam supplies many of its products, stagnates. Besides this smaller market, also the competition has become fiercer over the last years. This difficult situation currently results in a too low amount of production orders to cover the costs leading to underfunding. Because The Prysmian Group has announced that it might close some of its Dutch factories after March 2014, there currently is much uncertainty about the existence of Draka Amsterdam after this date. Within Draka Amsterdam there is a great need for more production orders and, according to the sales department, the only way to achieve this is by reducing both

lead times and production costs.

In order to deal with the current problematic situation, the Management Team of Draka Amsterdam has come up with the idea to serve the market from an additional Customer Order Decoupling Point – abbreviated as CODP – filled with intermediates in the form of uncoloured cores. This idea – further referred to as the Improvement Initiative – seemed to have much potential for Draka Amsterdam since it was expected to result in less working capital in combination with short lead times. Since the Improvement Initiative seemed very promising and time is precious for Draka Amsterdam these days, this Master Thesis has been fully focused on this idea only. The Main Research Question has been defined as:

“What will be the effects on the lead times and production costs for Draka Amsterdam when implementing an Additional CODP with neutrally coloured cores, which can be given their specific colour before the

core-stranding process, and how should the production process be optimized after the implementation of this CODP?”

The results of the research showed that the situation after the implementation of the Improvement Initiative – further referred to as the future state – would lead to a situation in which approximately 18 different

uncoloured cores would be stocked at an additional CODP in front of the core-stranding machines. With

these 18 cores in stock, the feasible aim for Draka Amsterdam is to be able to produce 1190 different cables within the required lead time from the market which is established at five working days.

Within the current state, Draka Amsterdam serves the market from two CODPs; one stock with raw material and one stock with finished goods. Since the current lead times are on average three weeks – 15 working days –, which is too long, Draka Amsterdam has on average around € 5.900.000 on finished goods in stock to deal with the short lead time requirements. Filtering this stock with finished goods on the 1190 different cables which can be produced from the additional CODP in the future state, revealed that approximately 60% consists out of these cables. Whenever Draka Amsterdam would reduce – within the future state – these 60% of cables to a length of 500 meters – to cut off short lengths –, it could decrease its stock with finished goods with approximately € 2.630.000. The amount of work in process will, according to the calculations, remain almost equal despite the extra amount of required stock at the additional CODP. This extra stock with working capital – 18 uncoloured cores – is compensated by shorter overall lead times resulting in almost equal work in process.

Implementing the Improvement Initiative will give Draka Amsterdam the opportunity to deal with the demanded customization – Draka Amsterdam offers approximately 3100 different products – and lead times from the market. Process optimization within the future state should be according to the principle of

(7)

Introduction

Page 6

state according to this principle will require a Lean approach upstream of the additional CODP and an Agile approach downstream of the additional CODP. Within the production upstream of the additional CODP, 80% of the production exists of the 18 mostly used cores and therefore this part can be seen as the mass production part. Customization within the production process of Draka Amsterdam comes downstream the additional CODP in which the cores can be stranded into many different cables. In order to serve the market with the required customization and lead times, an Agile process downstream of the additional CODP is required.

During the research, the following findings came to light which are applicable for the eventual successfulness of the Improvement Initiative.

- Storage space at the additional CODP; The total stock with the 18 mostly common cores at the

additional CODP before the core-stranding machines will require much space. Draka Amsterdam should do additional research in order to determine the best place for stocking these cores.

- Splitting up the Universal Strander; The Universal Strander – abbreviated as US – is capable of doing

both core-stranding and armouring tasks. However, it is not capable of doing these tasks synchronously. Detaching the armouring part from the US will make it possible to do both tasks and will also create additional space for stocking the 18 most common cores

- Stock management software; Continuously measuring the available stock of all three CODPs is

required for securing the supply of products and maintaining an optimal level of working capital. However, the current automated system within Draka Amsterdam is not able to do this.

- Maintain effective communication; In order to jointly achieve a successful implementation of the

Improvement Initiative, effective communication on every level is required. Within the current state, there is no good communication between the sales department and the factory. Since the sales department will eventually have to ensure an increasing amount of production orders, it is important that they understand the principle of the Improvement Initiative. Furthermore, effective communication within the factory and evolving people within the project will result in a better implementation.

The Improvement Initiative will, however, also bring along some disadvantages. Besides the need for extra storing space for the 18 most common cores and the need for a stock measuring system, it will also result in extra waste – € 100.000 per year – and the probable need for extra operators. Both disadvantages concerning this extra waste and need for more manpower come due to the stoppage of clustering orders downstream the additional CODP resulting in extra required setup procedures.

(8)

Introduction

Page 7

1. I

NTRODUCTION

This introduction chapter will start with background information about Draka and the Prysmian Group followed by a description about the “problematic” situation within the factory in Amsterdam – further referred to a “Draka Amsterdam”. In accordance with this situation, the initial management question, which has eventually been formulated, led to a possible solution on which this Master Thesis is based. This introduction chapter will describe this possible solution which formed the basis for the Main Research Question. Eventually, this chapter will end with the report organization and reading structure.

In this Master Thesis, all important concepts are in bold whenever they are mentioned for the first time. All these concepts in bold are defined in the Glossary at the back of this Master Thesis.

§1.1 P

ROFILE OF

D

RAKA

Draka is a worldwide company which develops, produces and sells all kinds of cable solutions for a wide diversity of customers from different industries, varying from aircrafts, trains and cars to homes, offices and wind turbines. The company has been founded on the 20th of April 1910 by Jan Teeuwis Duyvis and Francis Howe by the name of “De Hollandsche Draad- en Kabelfabrieken”. In 1970, this name changed to “Draka Kabel” at the moment that it was taken over by Royal Philips NV. By the year 1986, Philips dissociated itself from the company and Draka Kabel changed its name to ‘Draka Holding’, but today it is mainly referred to as “Draka”. In 2010, Draka counted nearly 9.400 employees divided over factories within 31 different countries. By the beginning of 2011, the French Nexans, the Chinese Xinmao and the Italian Prysmian were interested in taking over Draka. Eventually, Prysmian has acquired Draka in February 2011. Since this takeover of Draka, Prysmian has become the biggest cable manufacturer in the world.

This Master Thesis in Technology Management will focus on Draka Amsterdam, the factory where Draka initially was founded more than a century ago. These days, the factory is capable of producing around 3100 different products which are sold to wholesales, sold to companies for specific projects or produced as intercompany volume for other factories within the Prysmian Group. The Prysmian Group comprises all Draka and Prysmian factories together. An overview of the organization within Draka Amsterdam is given in figure 1.1.

(11)

Introduction

Page 10

Within Draka Amsterdam there are 83 people working of whom 50 are “blue collars”, 20 indirect “blue collars” and 13 “white collars”. In total there are 45 cable production machines of which only a few is used to produce the majority of the products for the market. The production process for an average cable comprises eight production steps which are shown in figure 1.2. Note that there are four points at which a product can be sold: after core-isolating, twice after sheathing and after armouring. These four points are indicated in figure 1.2 as “exit”.

Definitions about these production steps are given in the glossary at the end of this thesis. A more detailed explanation of each production step including photographs of all intermediate goods is given in appendix A. Uncommon production steps such as “lead pressing” (providing cables with lead protection for the petrochemical industry) and “braiding” (an armouring method for thin cables) will further not be mentioned in this thesis. This is because these processes are used only for a small amount of the total production volume and therefore they were not relevant for the research.

(12)

Introduction

Page 11

§1.2 P

ROBLEM SITUATION WITHIN

D

RAKA

A

MSTERDAM

Draka Amsterdam is facing difficult times due to the current economic crisis. Since 2008, the cable market has shrunk and the competitiveness of the factory has, more than ever, become an important aspect of keeping the factory profitable. Since 2011, the building sector, to which Draka Amsterdam supplies many of its products, stagnates and therefore the factory has to cope with an even smaller market than it had during the first three years of the crisis. Also the amount of intercompany volume, which includes outsourced production orders from other factories within the Prysmian Group, has decreased. In 2011, Draka Amsterdam produced many orders for factories in Norway, France and the Czech Republic. During 2012, the only production orders for intercompany volume came from Norway.

Prysmian has announced that it only guarantees the currently available intercompany volume for Draka Amsterdam until March 2014. After this date, Draka Amsterdam might have to deal with another setback in available production orders. Forecasts state that it will be difficult to gather new production orders because the competition is fierce and the size of the cable market is decreasing. In the Netherlands, the Prysmian Group – including Draka Amsterdam – still has 55% of the cable market but, at the same time, Draka Amsterdam is losing orders because the competition is up to 30% cheaper. All in all, these developments have caused a coverage loss of approximately € 3,5 million in 2011 and for 2012, a coverage loss of approximately € 2,5 million is predicted. Accurate solutions to deal with this current state have been difficult to define and creativity, which was recently not very needed in this traditional factory, has become a very useful competence. The different factors causing the current on-going unprofitable situation are listed together in figure 1.3.

Eventually the factors listed in figure 1.3 led to the following initial Management Question, which formed the start of this Master Thesis research:

“How can Draka Amsterdam strengthen its competitiveness to gain more production orders?”

(13)

Introduction

Page 12

In order to improve the competitiveness of Draka Amsterdam, there are many possibilities such as focusing on exclusivity or improving the quality. However, a more thorough examination within and outside of the factory, which is described in detail in appendix B, led to the following exact Functional Problem:

“Draka Amsterdam suffers from the current economic crisis because the Dutch building sector stagnates. Also the competition has become fiercer over the last years. This resulted in a too low amount of production orders to cover the costs, resulting in loss for Draka Amsterdam. Production costs and lead times are proven to be both too high and therefore a bigger share on this shrinking cable market may only

be captured by improving these two parameters”

This functional problem is very serious and threatens the future of Draka Amsterdam. This is because Prysmian has announced that it might close some of its six factories in the Netherlands after March 2014 whenever they are not profitable enough. Since Draka Amsterdam is currently suffering from high coverage losses, there is much uncertainty about the existence of this factory after this date.

§1.3 T

HE

I

MPROVEMENT

I

NITIATIVE AS INPUT FOR THE RESEARCH DEFINITION

Having the fact in mind that an answer for the initial Management Question would lie in a reduction of the lead times and production costs made the Management Team – further abbreviated as MT – considered some possible solutions. Of all Improvement Initiatives one promises to have, by far, the most potential. This

solution by the MT – which will further be referred to as the “Improvement Initiative” – forms the basis of this Master Thesis research. Figure 1.4 shows the process at the beginning of this Master Thesis research

whereby the Management Question has been converted into the Functional Problem which eventually led to the Improvement Initiative. This paragraph will describe the Improvement Initiative and will end with the Main Research Question.

This Improvement Initiative comprises the altering of the colouring process of the cores within the cables. Whenever a cable is cut open, the different cores can be distinguished from each other due to their different colours. Instead of giving these cores their colour during the extrusion process, a colour can be added just before the cores are combined into a cable during the core-stranding process.

The isolation of each core would be given a neutral colour and just before the core-stranding process, the cores will be given their specific colours with special colouring devices. Colouring the cores during a later stadium within the production process can lead to the implementation of an additional customer order

decoupling point – further abbreviated as CODP – within the production process and may eventually lead to

(14)

Introduction

Page 13

shorter lead times towards the customer. The CODP identifies the point in the material flow where the product is linked to a specific customer (Olhager, 2012). Figure 1.5 explains the implementation of this Additional CODP in a visual way.

As stated before, Draka Amsterdam is capable of producing around 3100 different types of end products – both cables and separate cores. The production of an average cable usually takes eight production steps and after each step, many different intermediates are produced leading to much ramification within the production process. Figure 1.6 visualizes this ramification by showing that each cable starts from one type of raw material – 8mm copper – which eventually will be constructed into one of the 3100 different types of end products being available for customers.

Reducing the amount of ramification after the core isolation process, by making the isolation only available in a neutral colour, will decrease the amount of different cores from 339 to 107. Research based on empirical data of 2011 showed that 18 of these 107 different cores are covering 80% of all products which are eventually stranded on the Universal Strander – further abbreviated as US – and the Lesmo. Whenever

Fig. 1.5, Creation of an Additional CODP after implementing the proposed colouring process

(15)

Introduction

Page 14

Draka Amsterdam would create a CODP with these 18 different cores, it would be able to produce 1190 different cables within a shorter time frame than currently.

According to Mr. Diemel, stock manager within Draka Amsterdam, creating an Additional CODP before the core-stranding process may decrease the average lead time significantly. He also states that it will have a positive effect on the total amount of working capital which is the sum of all raw material, work in process – further abbreviated as WIP – and finished goods. Within the current state, there is a great need for finished goods due to the short lead time requirements – minimum of five working days – which are demanded by the market. The current lead times are often too long to serve the market and many products are therefore stocked as finished goods. After the creation of an Additional CODP before the core-stranding process, the amount of finished goods needed to serve the market may decrease due to the shorter lead times from this CODP.

The amount of WIP, in the form of neutrally coloured cores, will increase in front of the core-stranding machines – the US and Lesmo. Therefore, the exact impact on the total amount of working capital was difficult to define prior to the research. Over the total amount of working capital, which was on average € 11.316.000 during 2012 until week 45, an annual percentage of 11% needs to be paid. This percentage is based on costs for insurance, space to stock and interest. A significant reduction of the working capital will therefore lead to lower production costs for the products as well.

Within the current state, Draka Amsterdam has two different CODPs; one before the production process and one after the production process – see figure 1.5. The CODP before the production process includes the raw material – 8mm copper – needed for production. Production orders which are given from this point are “make to order” – further abbreviated as MTO. The production orders needed to maintain the level of the stocks filled with finished goods are “make to stock” – further abbreviated as MTS. The stocks with MTS products are forecast-driven using “Material Requirements Planning" – further abbreviated as MRP. The basic function of MRP is to plan material requirements and deal with two basic dimensions of production control; quantities and requirements (Hopp & Spearman, 2008). By changing the colouring process it might be possible to replace the current CODPs – both MTS and MTO – with one “assemble to order” – further abbreviated as ATO – CODP which is capable to serve the market with an overall shorter lead time for a wide range of different cables. However, according to Towill (2005) it should be clear that it is highly unlikely that a business will discover that only one CODP is needed to sufficiently serve the market in an efficient way. This assumption will probably be applicable for Draka Amsterdam as well resulting in more than one CODP after implementing the Improvement Initiative.

During the entire research, there was no need to investigate the technical feasibility. It has been assumed that Draka Amsterdam will eventually be able to “master” a technology by which cores can be coloured just before the core-stranding process and that possible technical obstacles are non-existing. Therefore the technical aspects of the Improvement Initiative are outside the scope of this research. During this research an additional research has been initiated by Draka Amsterdam to determine which technology will be used to colour the cores.

(16)

Introduction

Page 15

The Main Research Question of this Master Thesis research – which will further be discussed in chapter 3 – is defined as:

“What will be the effects on the lead times and production costs for Draka Amsterdam when implementing an Additional CODP with neutrally coloured cores, which can be given their specific colour

before the core-stranding process, and how should the production process be optimized after the implementation of this CODP?”

§1.4 R

EPORT ORGANIZATION AND READING STRUCTURE

Figure 1.7 summarizes the report organization. The Management Question which was initially established in the beginning led to the Improvement Initiative. Eventually, the course of the research was adjusted and the focus became on this particular possible solution for the current problematic situation within Draka Amsterdam. Since the predicted advantages of the Improvement Initiative were unfounded, research had to be done before making an investment for implementation. For this reason a theoretical framework – chapter 2 – has been made up on which the recommendations of this research could be based. In accordance with the theoretical framework, the methodology – chapter 3 – has been made up. The methodology contains the research outline and information about the Diagnosis and Design phase of which the eventual results are respectively described in chapter 4 and 5. Additional information concerning the eventual implementation process of the Improvement Initiative is described in chapter 6.

(17)

Theoretical framework

Page 16

2. T

HEORETICAL FRAMEWORK

The theoretical framework for this research starts with relevant information about the CODP. This is because the creation of an Additional CODP is the main aspect of the Improvement Initiative. The creation of an Additional CODP also requires knowledge about the correct positioning of it. Therefore, the important factors for positioning a CODP are described as well.

Literature states that the recommended approach for production optimization upstream and downstream of the CODP is fundamentally different. Therefore, both managerial recommendations and optimization methods for upstream and downstream of the CODP will also be described in this theoretical framework. The theoretical framework will form a basis of the eventual decision about whether or not the Improvement Initiative should be implemented. Furthermore, the theoretical framework will give insight about how the production process should be managed after the creation of an Additional CODP before the core-stranding process.

§2.1 C

OSTUMER ORDER DECOUPLING POINT

The CODP separates the part of the supply chain that responds directly to the customer from the part of the supply chain that uses forward planning and a strategic stock to buffer against the variability in the demand of the supply chain (Naylor, Naim, & Berry, 1999). In the literature, the CODP is also called “Order

Penetration Point” (OPP), “Decoupling Point” (DP) and “Customer Order Point” (COP). In Dutch it is referred

to as the “Klantorder Ontkoppelpunt” (KOOP). For the sake of clarity this research only uses the term “Costumer Order Decoupling Point”.

In the literature, the CODP is mentioned within different industries for different kind of products such as food (Soman, van Donk, & Gaalman, 2004), electronic products (Towill, 2005) and cars (Van Hoek, 2001). The principle of a CODP was first mentioned in a logistic context by Sharman (1984) who indicated the following five different locations within a supply chain where the CODP can be positioned. These are before:

- Engineering;

- Fabrication;

- Assembly;

- Delivery;

- Installation.

These locations within a supply chain, except the CODP before installation, are visually given in figure 2.1. Different manufacturing strategies must be developed for pre-CODP operations (i.e. upstream; forecast-driven) vs. post-CODP operations (i.e. downstream, customer-order-forecast-driven), since these two stages are fundamentally different (Olhager, J., 2003).

(18)

Page 17

A CODP divides the material flow activities in a customer order part and a part which is controlled based on

predictions and stocks (Bokhorst & Slomp, 2011). According to Sharman (1984) it is the last point at which

inventory is held. The inventory at the CODP is a strategic stock-point since delivery promises are based on the stock availability at the CODP and the lead times and capacity availability for the customer order-driven activities downstream the CODP (Olhager, 2012). In other words, a downstream movement of the CODP will lead to shorter lead times as long as there is enough inventory in stock at the CODP to serve the market. Associated with the positioning of the CODP is the cognate issue of postponement. Postponement means delaying activities in the supply chain until customer orders are received with the intention of customizing products, as opposed to performing those activities in anticipation of future orders (Van Hoek, 2001). The aim of postponement is to increase the efficiency of the supply chain by moving product differentiation at the CODP closer to the end user (Naylor, Naim, & Berry, 1999). If the customization offered by a factory is wide and enters the product at early production stages, an MTO policy is necessary, whereas if customization enters at a very late production stage ATO may be more appropriate (Olhager, J., 2003). This means that the possibilities of postponing the CODP are depending on the amount of customization given to the products. Lampel and Minztberg (1996) divided the degree of customization in the following five levels:

- Pure standardization;

- Segmented standardization;

- Customized standardization;

- Tailed customization;

- Pure customization.

Yang & Burns (2003) emphasized the relationship between postponement and customization by connecting six stages in the supply chain with the continuum of customization by Lampel and Minztberg. This resulted in the overview of figure 2.2 in which the dotted line indicates the location of the CODP. The CODP divides the

postponement activities with the customization activities.

(19)

Page 18

Table 2.1, Market-related factors for positioning the CODP

§2.1.1 P

OSITIONING OF THE CUSTOMER ORDER DECOUPLING POINT

Towill (2005) states that locating a CODP of the supply chain correctly can open a wealth of opportunities while a badly positioned CODP will jeopardize the efficiency of the entire manufacturing process. Since it is clear that a CODP can be positioned on many places within the supply chain, it should also be clear that the optimal position may be determinable by certain factors. According to Olhager (2003) the most important factors can be divided into three categories, related to “market-”, “product-” and

“production-characteristics”. The following sections will explain all underlying factors behind these three categories.

MARKET-RELATED FACTORS

According to Olhager (2003), there are six market-related factors for determining the optimal locations of the CODP:

- Delivery lead time requirements;

- Product demand volatility;

- Product volume;

- Product range and product customization requirements;

- Customer order size and frequency;

- Seasonal demand.

Moving the CODP as close to the end consumer as possible will ensure the shortest lead time for the consumer (Mason-Jones & Towill, 1999). Therefore, a shorter lead time delivery requirement will demand a downstream movement of the CODP. Low product demand volatility means that an item can be forecast-driven (Olhager, J., 2003) and thereby requires a downstream position of the CODP. Product volume is related to demand volatility in that the relative volatility is lower for high-volume items (Olhager, J., 2003). According to Yang and Burns (2003), a high product range and much product customization requirements results in an upstream movement of the CODP. High frequency leads to repetitive demand making forecasting easier and large customer order sizes are typically associated with high demand volumes (Olhager, J., 2003). Changes in seasonal demand may require a product to shift between MTS and MTO or ATO depending upon the season (Olhager, J., 2003).

Table 2.1 summarizes the market-related factors and indicates when these factors are suggesting an upstream or downstream movement of the CODP.

Movement of the CODP

Market related factors Downstream Upstream

1. Delivery lead time requirements Sho rt lead-time Lo ng lead-time

2. Product demand volatility Lo w vo tality High vo tality

3. Product volume High vo lume Lo w vo lume

4. Product range and product customization requirements Lo w range High range and and custo mizatio n and custo mizatio n

5. Customer order size and frequency Lo w o rder sizes High o rder sizes and high frequency and lo w frequency

(20)

Page 19

Table 2.2, Product-related factors for positioning the CODP PRODUCT-RELATED FACTORS

According to Olhager (2003), the product-related factors affecting the position of a CODP are: - Modular product design;

- Customization opportunities;

- Material profile;

- Product structure.

With modularization, the exact collection of pre-produced modules is assembled into the final system so as to rapidly produce the precise product ordered by the customer (Towill, 2005). A high rate of modularization will lead to a downstream movement of the CODP. Whenever the customization which is offered is wide and enters the product at early production stages, an MTO policy is necessary, whereas if customization enters at a very late stage ATO may be more appropriate (Olhager, J., 2003). Much customization will therefore requisite a downstream movement of the CODP. The so called “material profile” refers to the ‘V-A-T logical structure’ which is a framework to describe a complex production system in general (Spencer & Lockamy, 2009). V is the typical profile for a process firm with a divergent material flow from raw materials to finished goods. A and T profiles represent assembly situations (Olhager, J., 2003). Product structure indicates the product complexity (Olhager, J., 2003). Olhager refers to Wang and von Tunzelmann (2000) who asses complexity in terms of the dimensions of “depth” and “breadth“. In the context of product complexity, depth means to the cognitive complexity embodied in the components and breadth arises out of the number of components and sub-assemblies involved (Wang & von Tunzelmann, 2000). Since a greater breadth refers to a higher number of components and sub-assemblies it also refers to more customization and can therefore be linked to an upstream movement of the CODP.

Table 2.2 summarizes the product-related factors and indicates when these factors are suggesting an upstream or downstream movement of the CODP. Note that within this table the material profile is a non-measurable factor since it represents a specific designation of an observed production system.

PRODUCTION-RELATED FACTORS

According to Olhager (2003), the production-related factors for positioning of the CODP are: - Production lead time;

- Planning points;

- Flexibility;

- Bottleneck;

- Sequence-dependent set-up times.

Product-related factors Downstream Upstream

7. Modular product design High mo dularizatio n Lo w mo dularizatio n

8. customization opportunities High custo mizatio n Lo w custo mizatio n

9. material profile Is a no n-measurable facto r

(21)

Page 20

A direct result of moving the CODP within the supply chain will be the changing lead time. The relationship between production lead time and delivery lead time requirements poses a major constraint on the CODP position (Olhager, J., 2003). According to Olhager (2003), a planning point is a manufacturing resource or a set of manufacturing resources such as a work centre or a work cell that can be regarded as one entity from a production and capacity planning point of view. He also states that the number of planning points in a manufacturing process restricts the number of potential CODP position. Flexibility, in general, can be defined as the ease by which changes can be realized (Bokhorst & Slomp, 2011). By positioning the CODP further upstream a company can achieve a higher degree of flexibility (Rudberg & Wikner, 2004). According to Hopp and Spearman (2008), the bottleneck can be referred to as the station in the line with the highest long-term utilization. From a resource optimisation point of view, it is advantageous to have the bottleneck upstream the CODP, so the bottleneck does not have to deal with volatile demand and a variety of different products. With respect to the just-in-time principle of elimination of waste, it would be best to have the bottleneck downstream the CODP so that the bottleneck only needs to work on products for which the firm has customer orders (Olhager, J., 2003). Resources with sequence-dependent set-up times are best positioned upstream the CODP. Such resources can easily turn into bottlenecks without proper sequencing, which is a likely course of action for downstream operations (Olhager, J., 2003).

Table 2.3 summarizes the production-related factors and indicates when these factors are suggesting an upstream or downstream movement of the CODP. Note that within this table the planning points do not determine whether a CODP has to move upstream or downstream.

Altogether, the previous sections can be summarized as follows:

“The optimal position for a CODP is determined by 16 different factors which are categorised into market-, product- and production- characteristics. One should consider the applicability of each of these factors for the

specific situation for which a person wants to determine the optimal position for a CODP.”

§2.2 L

EAN VERSUS

A

GILITY

According to Rudberg and Wikner (2004), it can be argued that customers (end consumers as well as industrial customers) nowadays put two major pressures on many companies. Customers want products that fit their specific needs but are on the same time not willing to pay high premiums for these customized products compared to competing standard products in the market. Skinner (1974) emphasized the fact that there are many ways to compete besides of producing at low costs, but a factory cannot perform well on

Production-related factors Downstream Upstream

11. production lead time Sho rt lead-time Lo ng lead-time

12. planning points A certain amo unt o f planning po int do es no t give a suggestio n fo r a specific CODP po sitio n.

13. flexibility Lo w flexibility High flexibility

14. bottleneck High mo dularizatio n reso urce

o ptimisatio n

15. sequence-dependent set-up times Little sequence- M uch sequence-dependent set-up times dependent set-up times

(22)

Page 21

“every yardstick”. As companies choose among process alternatives, they need a clear understanding of the changing alignment between manufacturing and the needs of their markets (Hill, Menda, & Dilts, 1998). Olhager (2012) mentions that standard products with very narrow range “win” orders on price, wherefore the key manufacturing task is to provide low-cost production – is applicable to MTS operations upstream a CODP. On the other hand he mentions that special products in a wide range win orders based on delivery speed and unique design capability, wherefore manufacturing has to meet specifications and delivery schedules, which requires high flexibility – is applicable to MTO operations and downstream a CODP. This statement by Olhager is in alignment with the equilibrium between productivity upstream of the CODP and flexibility downstream of the CODP which has been established by Wikner and Rudberg (2004) – see figure 2.3. A cost-leadership strategy is well aligned with Lean manufacturing operations capabilities and cost performance, while a differentiation strategy is well aligned with Agile manufacturing operations capabilities and flexibility performance (Hallgren & Olhager, 2009). Paragraph 2.2.1 and 2.2.2 will describe Lean and

Agility separately and paragraph 2.2.3 will give a description of Leagility, which is a combination of Lean and

Agility together in one production process.

§2.2.1 L

EAN

The term “Lean”, which origins from the Toyota Production System (TPS), has been established by Womack and Jones in 1996. Lean focuses on the elimination of “muda” which mean waste. Waste comprises any human activity which absorbs resources but creates no value (Womack, J.; Jones, D., 2003). Lean focuses on efficiency, aiming to produce products and services at the lowest cost and as fast as possible (Antony, 2011). According to Womack and Jones (2003), Lean can be summarized in the following five principles:

1.) Value; Value refers to typical changes in the form, fit, function, or information content. These steps

tangibly contribute value in the customer's eyes (McCarty, Daniels, Bremer, & Gupta, 2004). The first step in becoming Lean is to specify all value within a product.

2.) Value stream; The second step is to identify the Value stream. A value stream comprises all activities, both “value added” as well as “non-value added”, currently needed to bring a product from raw material into the hands of the customer (Goubergen, 2012). The aim is to reduce the non-value added activities to a minimum.

3.) Flow; Once value has been precisely specified, the value stream for a specific product fully mapped by the Lean enterprise, and obviously wasteful steps eliminated, the next step is to make the remaining steps flow (Womack, J.; Jones, D., 2003).

4.) Pull; If an upstream operator delivers work to the downstream process, but does so in response to status changes in the downstream process, then this is pull (Hopp & Spearman, 2008). Creating a pull

(23)

Page 22

environment will reduce the amount of unwanted products being produced. In a pull system, the production is demand driven, i.e., only the quantity required by the customer is produced, whereas in traditional push systems parts are pushed through the delivery channel from the production (Matzka, Di Mascolo, & Furmans, 2009)

5.) Perfection; The process of improvement is endless. When an organization begins to accurately specify value, identify the entire value stream, make the value-creating steps for specific products flow continuously, and let customers pull value from the enterprise it will, according to Womack and Jones (2003), eventually need to start with continuous improvement which refers to the endless pursuit of “perfection”.

§2.2.2 A

GILITY

Present time competitiveness requires, instead of only a low product’s price, also quality, delivery time, and customer choice or in a more exact way, customer satisfaction (Zharifi & Zhang, 1999). In order to respond to these requirements, a company needs to become agile. The origins of Agility as a business concept lies in “flexible manufacturing systems” hence that a key characteristic of an agile organisation is flexibility (Christopher, 2000). According to Sharifi and Zhang (1999), Agility in concept comprises two main factors which are:

- Responding to change (anticipated or unexpected) in proper ways and due time. - Exploiting changes and taking advantage of them as opportunities.

An Agile manufacturer, in this way is an organisation with a broad vision on the new order of the business world, and with a handful of capabilities and abilities to deal with turbulence and capture the advantageous side of the business (Zharifi & Zhang, 1999). According to Christopher (2000), to be truly Agile a company must possess the following four distinguishing characteristics:

1.) Market sensitive; Being market sensitive means that the supply chain is capable of reading and

responding to real demand (Christopher, 2000) – e.g. capture emerging trends and listen to customers. According to Agarwal et al. (2006), market sensitiveness is characterized by six measures; delivery speed, delivery reliability, new product introduction, new product development time, manufacturing lead time and customer responsiveness.

2.) Virtual supply chain; Virtual supply chains supply real time data across companies boundaries and

aim to reduce inventory levels by using more effective use of information – particularly information considering customer demand (Sweeney, 2008). Virtual supply chains are information based rather than inventory based (Christopher, 2000).

3.) Process integration; Process integration comprises a collaboration between buyers and suppliers,

joint development, common systems and shared information (Christopher, 2000). In an Agile supply chain there are high levels of integration between processes within the firm and between the firms upstream and downstream in the external supply chain (Sweeney, 2008). This alters the fragmentation which often is a characteristic of many traditional managed supply chains (Sweeney, 2008).

4.) Network based; Organisations being capable of good structuring and good co-ordination of

(24)

Page 23

§2.2.3 L

EAGILITY

The difference between the upstream and the downstream pertaining to the position of the CODP was initially mentioned by Naylor et al. (1999) who state that the upstream supply chain should be Lean and the downstream supply should be Agile. In this context, the demand for variety of products and the demand for variability in production are determining whether an approach needs to be Lean or Agile. Agility means using market knowledge and a virtual corporation to exploit profitable opportunities in a volatile market place (Naylor, Naim, & Berry, 1999). Leanness means developing a value stream to eliminate all waste, including time, and to ensure a level schedule (Naylor, Naim, & Berry, 1999). The graphs in figure 2.4 show the difference between the demand upstream and downstream of the CODP in which the high variable demand downstream requires Agility and the constant demand upstream requires Lean. Within these graphs, the vertical axis indicates the demand over a period of time.

According to Naylor et al. (1999), manufacturers should not be looking at operations in isolation from the rest of the supply chain. Whether to develop and Agile capability or a Lean manufacturing structure depends on which part supply chain is considered for optimization. The combination of Lean upstream and Agility downstream of the CODP is the core idea of the “Leagility approach”. In focusing on the CODP, a so-called Leagile supply chain concept is a cornerstone to enabling competitiveness in many market sectors (Towill, 2005). Olhager (2012) and Towill (2005) both made an overview of the important aspects on both sides of the CODP which have been combined in one overview – see table 2.4. In case of a Leagile approach the goal is to combine these aspects optimally to maximize profit by satisfying the customer with a quick delivery of specified products.

Aspects Upstream the CODP Downstream the CODP

Lean versus agile Lean Agile

Delivery Philosophy Make to stock Make to order

Order volatility Small Large

Order variaty Small Large

Value added Low High

Product level Generalized modules Customer specific

Business objective for this stage Driven by cost Driven by availabily Supply chain design Physically efficient Market responsive

Order winners Price Delivery speed, flexibility

Key properties Productivity Flexibility

Fig. 2.4, Effects of the CODP (extracted from Naylor et al., 1999)

(25)

Page 24

Considering the differences between the process upstream and downstream of the CODP, the following statement is defined:

“Manufacturing strategies are fundamentally different upstream and downstream of a CODP whereas the process upstream of the CODP demands a Lean strategy while the process downstream of the CODP demands an Agile strategy. Combining Lean upstream and Agility downstream of the CODP is referred to as a

Leagile approach.”

§2.3 M

AKE TO STOCK

,

A

SSEMBLE TO ORDER

,

M

AKE TO ORDER

Wikner and Rudberg (2004), state that four CODPs are most frequently used; engineer-to-order – abbreviated as ETO –, MTO, ATO and MTS. However, in the context of this Master Thesis research, ETO will further not be described since ETO orders hardly happen within Draka Amsterdam. MTS includes all options regarding keeping inventory in the distribution system; either at distributors, wholesalers or retailers (Olhager, 2012). In all these environments, the product is produced to stock with respect to the form; however, they may differ in terms of time and space relative the ultimate customer (Olhager, 2012). Whenever the CODP for a specific product lies "within” the production process and is made up from intermediates/parts, which are MTS, the further production is ATO. Different products being delivered in an ATO fashion do not necessarily have to have the CODP in the same position (Olhager, 2012). In case of MTO, production orders are started at the beginning of the production process and an order has to run though the whole production process. Olhager (2003), states that whenever a required lead time is short, while the manufacturing lead time is long, MTO will never be a choice. The only difference between MTO and ATO lies within the point from which the order will start to produce, whereas MTO starts from raw material and ATO starts from intermediates. There are two fundamental sections in a material flow: MTS and MTO. The choice of MTS versus MTO is typically a natural and clear-cut one in practice, and the differences and consequences are usually well understood by manufacturing and supply chain managers (Olhager, 2012).

Olhager (2003), states that out of the 15 factors being mentioned in §2.1.1, two factors are the most important ones for the positioning of the CODP; the production to delivery lead time ratio (P/D ratio) and the relative demand volatility (RDV). The RDV is defined as the coefficient of variation, i.e. the standard deviation of demand relative to the average demand (Olhager, J., 2003). Both factors identify four different situations – MTO, ATO, MTS or a combination of all – which are mentioned in the matrix in figure 2.5.

(26)

Page 25

In summary, Olhager (2003) gives the following indications about when a specific CODP is suitable: - If the P/D ratio is less than one, a MTO product delivery strategy is possible;

- When the P/D ratio is more than one, part of the internal supply chain must be forecast driven leading to MTO and ATO;

- A high RDV will naturally lead to a MTO policy;

- A low RDV indicates that some parts can be produced to stock, potentially leading to ATO;

- A very low RDV may, even when the P/D ratio would allow MTO, lead to a MTS policy with the purpose to increase productivity;

According to Olhager (2003), a balance must be made between on the one hand minimizing the number of forecast items and, on the other hand, maximizing the opportunities to take advantage of economies of scale. In accordance with the different manufacturing strategies, the following statement has been defined:

“The determination of a manufacturing strategy – MTS, MTO or ATO – for a specific product depends in particular on the “production to delivery lead time ratio” and the “relative demand volatility”. Therefore, different products with different demands will probably result in different manufacturing strategies which

eventually will lead to different CODPs within a production process.”

§2.4 I

MPROVEMENT METHODS

The creation of an Additional CODP within the production process asks for a Lean approach upstream of the CODP and an Agile approach downstream of the CODP. In order to achieve this, an organization can use different improvement methods. This paragraph will describe some important Lean improvement methods. Many Lean improvements will eventually result in flexibility leading to a more Agile production process. Both Agility and flexibility imply “range” and “response dimensions” (Baker, 1996). According to Baker (1996), Agility places greater focus on the strategic levels, whilst flexibility is most often associated with the operational levels.

VALUE STREAM MAPPING

In order to investigate the efficiency of the current state of a production process, a clear overview of every step is needed. Getting a clear overview of the process can be achieved by using Value Stream Mapping” – further abbreviated as VSM. Rother and Shook (1999) describe this methodology as essential since it helps to visualize more than just the single-process level. The value stream map – further abbreviated as VSM – permits someone to identify every process in the flow, pull them out from the background clutter of the organization, and build an entire value stream according to Lean principles (Womack, J.; Jones, D., 1999). Kaizen efforts, or any Lean manufacturing technique, are most effective when applied strategically within the context of building a Lean value stream (Womack, J.; Jones, D., 1999). Kaizen events can drastically compress the throughput time for the product, eliminate wasted steps, and rectify quality, flexibility – Agility

–, availability, and adequacy problems (Womack, J.; Jones, D., 2003). According to Goubergen (2012), the

value stream can be defined as all activities, both “Value Added” as well as “Non Value Added”, currently needed to bring a product from raw material into the hands of the customer.

(27)

Page 26

TOTAL PRODUCTIVE MAINTENANCE

Total Productive Maintenance – further abbreviated as TPM – is designed to maximize equipment

effectiveness by establishing a comprehensive productive-maintenance system covering the entire life of the equipment, spanning all equipment-related fields – e.g. planning, use and maintenance – and, with the participation of all employees from top management down to shop-floor workers, to promote productive maintenance through motivation management or voluntary small-group activities (McKone, Schroeder, & Cua, 2001). According to El-Haik and Al-aomar (2006), attempts to minimize common equipment related wastes comprise:

- Setup and calibration;

- Breakdowns and failures – short and catastrophic; - Starving (idling) and blockage – stoppage;

- Reduced speed – low performance; - Startup – warm-up;

- Defects – scrap, rework, rejects.

According to El-Haik and Al-aomar (2006), properly implemented TPM also eliminates many machine-related bottlenecks through the following:

- Improved machine reliability; - Extended machine life;

- Increased capacity without purchasing additional resources. WORK STANDARDIZATION

According to El-Haik and Al-aomar (2006), work standardization is often sought as the best method for minimizing excess capacity, achieving production excellence, and laying the foundation for continuous improvement. Standardized work refers to the systematic determination and documentation of the work element sequence and process for each operation. The objective is to communicate clearly to an operator exactly how the job should be performed so that variability is minimized (El-Haik & Al-Aomar, 2006). Standardized work will bring more consistency and better quality to the shop floor.

POKA-YOKE

Within a Lean process, a Poka-yoke helps an operator to avoid mistakes. Saurin et al. (2012) describe poka-yoke as a device that either prevents or detects abnormalities, which might be detrimental either to product quality or to employees’ health and safety. According to Saurin et al. (2012), poka-yoke can be considered as the following three types of devices:

- Physical; if they block the flow of mass, energy or information, and do not depend on users

interpreting them – e.g. a wall.

- Functional; if they might be turned on or turned off due to an event – e.g. a lock or a password –,

without depending on user interpretation

- Symbolic; if they require interpretation, yet are physically present at the moment they are necessary

– e.g. a safety sign.

SINGLE-MINUTE EXCHANGE OF DIE(S)

Single-Minute Exchange of Die(s) – further abbreviated as SMED – is a theory and a set of techniques that

(28)

Page 27

2009). SMED improves the setup process and provides a setup time reduction up to 90% with moderate investments (Cakmakci, 2009). According to Shingo (1985), setup procedures can be divided in two parts:

- Internal setup; work that has to be done while a machine is off

- External setup; work which can be done while a machine is running

According to Cakmakci (2009), the SMED system includes the following three steps: - Separating the internal and the external setup;

- Converting internal setup to external setup; - Streamlining all aspects of the setup operations;

An overview of these steps and practical techniques is given in figure 2.6.

Reducing the time which is needed for the internal setup will result in more flexibility at the specific step for which SMED has been adopted. Making the productions steps more flexible will eventually result in a more

Agile production process. The fact that the SMED methodology, which is well known as a Lean methodology,

contributes to a more Agile process indicates that Agility is an element of Lean. However, in the context of a CODP, an often mentioned difference between Lean and Agility is that Lean focusses more on cost reduction and Agility focuses more on flexibility towards the market.

HEIJUNKA/PRODUCTION SMOOTHING

Heijunka is a key-element of the Toyota productionsystem which levels the release of production Kanbansin order to achieve an even production flow over all possible types of products (Matzka, Di Mascolo, & Furmans, 2009). The goal of heijunka is to supply one or more customer processes with a constant flow of small lots of different parts and at the same time generating a constant demand of parts for upstream processes, thus reducing or eliminating the need for spare capacity or stocks to cope with peaks of demand (Matzka, Di Mascolo, & Furmans, 2009). A second goal is to reduce the bullwhip effect (Matzka, Di Mascolo, & Furmans, 2009).

(29)

Page 28

5S

According to Womack and Jones (2003), 5S derives from the Japanese words for the following five practices leading to a clean and manageable work area:

- Seiri – organization; - Seiton – tidiness; - Seiso – purity;

- Seiketsu – cleanliness; - Shitsuke – discipline.

According to El-Haik and Al-aomar (2006), the 5s method, can be translated into the following English words: - Sort through and sort out: clean out the work area, keep in the work area what is necessary to do

the work, and relocate or discard what is not actually used or needed (El-Haik & Al-Aomar, 2006). - Set in order and set workstation limits; arrange needed items so that they are easy to find, easy to

use, and easy to return. Principles of workstation design and motion economics are applied to arrange work items properly. The goal is to streamline production and eliminate search time and delays (El-Haik & Al-Aomar, 2006).

- Shine and inspect through cleaning; clean work tools and equipment and make them available at

their points of use. While cleaning, inspect tools and equipment and provide the care needed to sustain performance. This may include inspection, calibration, lubrication, and other care and preventive maintenance actions (El-Haik & Al-Aomar, 2006).

- Standardize; Make all work areas similar so that procedures are obvious and instinctual and defects

stand out. Standard signs, marks, colors, and shapes are used to standardize the workplace (El-Haik & Al-Aomar, 2006).

- Sustain; Make the preceding four rules an integral attribute of the business or production system.

Slowly but surely, an effort should be made to train employees and use 5S audits to sustain the continuity of their use. As 5S practices become natural habits, over time the total benefits of 5S will materialize and become transparent (El-Haik & Al-Aomar, 2006).

Within the context of Lean, implementation of 5s will help to organize the shop floor and maintain the discipline which is needed for this.

IMPROVEMENT METHODS SUMMARIZED

The above-described improvement methods comprise only a part of the total amount of methods for making a production process more Lean and Agile. Because the underlying thoughts of most Lean optimization methods are fundamentally different from each other, it is important to first find out for what optimization purpose it will eventually be used – e.g. “more Lean” or “more Agile”. Altogether, with regards to the considered optimization methods, the following statement has defined:

“Preparatory to the optimization of a specific part of the production process – e.g. upstream or downstream of the CODP – an appropriate improvement method should be chosen. Since most of the optimization

(30)

Methodology

Page 29

3. M

ETHODOLOGY

According to Prins (2008) and De Leeuw (2005), the process of improving a business – or solving a problem – can be divided in three phases; Diagnosis, Design and Change. The Diagnosis aims for finding an explanation for the undesired output. The Design can be considered as an advice towards Draka Amsterdam in the form of the identification of certain consequences and opportunities with regards to the Improvement Initiative. Therefore, the design phase will not comprise a new idea for improvement, but an elaboration of the Improvement Initiative which was prior to this research only known as a possible way for improvement. Every predicted advantage by the MT about the Improvement Initiative was unfounded and needed to be researched.

Implementation of the Design will not be part of the research since the aim of this research is to come up with a prediction of the effects which the implementation of the Improvement Initiative will have on Draka Amsterdam. This research is an advice towards Draka Amsterdam about whether or not it should implement

the Improvement Initiative. This methodology chapter contains the research outline provided with

explanation and underlying information about both the Diagnosis and the Design phase.

§3.1 R

ESEARCH

O

UTLINE

Figure 3.1 shows the Research Outline in which the first two phases of Prins and De Leeuw are listed;

Diagnose and Design. Whether or not the Improvement Initiative will be implemented, depends on the

outcomes of the Diagnosis and the Design phase. Relevant underlying information about the diagnose phase is given in the paragraph 3.2 and the eventual outcome of this phase is described in chapter 4. The Design phase comprises predictions and recommendations about the future state after the Improvement Initiative has been implemented. Relevant underlying information about the Design phase is given in paragraph 3.3 and the eventual outcome of this phase is described in chapter 5. Chapter 6 – Issues for implementation – will eventually describe how the employees within Draka Amsterdam were involved within the research and will give some recommendations about the implementation process.

Master thesis Technology Management A business case for the improvement of the industrial performance of Draka Amsterdam Ing. E.M. Gardebroek

Master thesis

Technology Management

A business case for the improvement of the industrial

performance of Draka Amsterdam

Ing. E.M. Gardebroek

V

OORWOORD

M

ANAGEMENT SUMMARY

TABLE OF CONTENTS

1. I

NTRODUCTION

§1.1

P

D

§1.2

P

D

A

§1.3

T

I

I

§1.4

R

2. T

HEORETICAL FRAMEWORK

§2.1

C

§2.1.1

P

§2.2

L

A

§2.2.1

L

§2.2.2

A

§2.2.3

L

§2.3

M

,

A

,

M

§2.4

I

3. M

ETHODOLOGY

§3.1

R

O