• No results found

A case study on how an e-tailer can use a multiple criteria ABC analysis to identify risk in the selection of suppliers

N/A
N/A
Protected

Academic year: 2021

Share "A case study on how an e-tailer can use a multiple criteria ABC analysis to identify risk in the selection of suppliers"

Copied!
90
0
0

Loading.... (view fulltext now)

Full text

(1)

A case study on how an e-tailer can use a multiple

criteria ABC analysis to identify risk in the selection

of suppliers

Master thesis

Joel Strand and Louise Strandänger

Spring 2016

(2)

A case study on how an e-tailer can use a multiple

criteria ABC analysis to identify risk in the selection

of suppliers

Examensarbete i industriell ekonomi om 30 hp, vid utbildningen till civilingenjör i industriell ekonomi.

Ämne: Produktionsekonomi Presentationsdatum: 2016-06-07

Publiceringsdatum (elektronisk version): 2016-06-20

URL: http://urn.kb.se/resolve?urn=urn:nbn:se:liu:diva-02581 Språk: Engelska

Examinator: Veronica Lindström, iei Handledare: Helene Lidestam, iei ISRN: LIU-IEI-TEK-A--16/02581--SE

Tekniska högskola vid Linköpings universitet

Institutionen för ekonomisk och industriell utveckling Avdelningen för produktionsekonomi

Tekniska Högskolan 581 83 Linköping

(3)

Abstract

Purpose– The purpose of this master thesis is to explore how an e-tailer selling bulky items can use a multiple criteria ABC analysis to make its purchasing process more effective, while balancing richness and reach, with the performance measurements of profitability, total asset turnover and inventory turnover. The purpose will be accomplished through a single case study on an e-tailer active on the Swedish furniture and home furnishing market.

Methodology– This thesis applies a multiple criteria ABC-analysis to a single case study. The approach is semi-deductive as theory is combined with interviews on how to match and adapt theory about inventory control and purchasing with the specific requirements of an e-tailer selling bulky items.

Findings– This thesis has resulted in a set of recommendations that aim to make the purchasing process of an e-tailer more effective. That is, capital and inventory space will be better allocated to the e-tailer’s more profitable items. Among other things, this thesis shows how dead articles can be identified and how a purchaser can prioritize more profitable articles over less profitable ones when making procurement decisions. The other recommendations are for the e-tailer to investigate the possibilities of decoupling the supply chain by keeping stock at the suppliers’ premises, to match the supplier reliability with their importance in the supply chain, and lastly to explore possibilities of drop shipment. Further, the main finding is that a comparison between the A-, B-, and C-classes and the reliability of the suppliers, highlights a gap and a possible risk. Put differently, the importance of a specific item for the business should be reflected in the choice of supplier and the multiple criteria ABC analysis is the tool to illustrate the importance. Keywords – E-commerce, E-tailer, richness, reach, transaction cost, ABC analysis, multiple criteria ABC, MCABC, inventory turnover ratio, supplier selection, purchasing

(4)

Sammanfattning

Syfte – Syftet med detta examensarbete är att undersöka hur en e-handelsdetaljist som säljer skrymmande artiklar kan använda en flerdimensionell ABC-analys för att göra sin inköpsprocess mer effektiv och balansera richness och reach, med mätetal som lönsamhet, kapitalomsättnings-hastighet och lageromsättningskapitalomsättnings-hastighet. Syftet kommer att uppfyllas genom en fallstudie på en e-handelsdetaljist verksam på den svenska möbel- och heminredningsmarknaden.

Metod– Denna fallstudie använder sig av en flerdimensionell ABC-analys. Tillvägagångssättet är semi-deduktivt då intervjuer och teori om hur lagerstyrning och inköp kan matchas och anpassas till ett företags specifika behov.

Resultat – Den här uppsatsen har resulterat i en rad åtgärder som syftar till att göra en handlares inköpsprocess mer effektiv. På så vis att kapital och lageryta bättre allokeras till e-handlarens lönsamma artiklar. Bland annat visar den här uppsatsen hur döda artiklar kan iden-tifieras och hur inköparen kan prioritera mer lönsamma artiklar över olönsamma vid inköp. De andra åtgärdena handlar om att undersöka möjligheter att frikoppla försörjningskedjan genom att lagra produkter hos leverantören, att matcha leverantörernas pålitlighet och deras betydelse i försörjningskedjan, och slutligen att utforska möjligheter att utöka drop shipment. Det främsta bidraget är att eventuella felprioriteringar och risker blir tydliga genom en jämförelse mellan A-, B- och C-klasserna och leverantörernas pålitlighet. Med andra ord bör den affärsmässiga inverkan som respektive artikel har på e-handlarens resultat avspegla sig i valet av leverantör. En flerdi-mensionell ABC-analys kan användas för att påvisa respektive artikels affärsmässiga inverkan. Publikationstyp– Examensarbete för utbildning till civilingenjör (masteruppsats).

(5)

Preface

This thesis is written by two students, enrolled at different universities. Louise is a student at KTH and Joel is a student at Linköping University. The thesis will be published twice but with a different front page and International Standard Technical Report Number (ISRN) or löpnummer. Publication of KTH: 2016:84

Publication of LiU: LIU-IEI-TEK-A--16/02581--SE

Joel Strand and Louise Strandänger June 15, 2016, Stockholm

(6)

Terminology

Abbreviation Full name

AHP Analytical Hierarchy Process

CODP Customer Order Decoupling Point

EBIT Earnings before interests and taxes

EOQ Economic order quantity

ERP Enterprise Resource Planning

ITR Inventory turnover ratio

MCABC-analysis Multiple Criteria ABC Analysis

MTO Make to order

Order lead time The time which elapses between placing an order and the delivery of the goods

ROCE Return on capital employed

ROA Return on assets

ROE Return on equity

ROS Return on sales

SNI Swedish Standard Industrial Classification (swe. Svensk Närings-grensindelning)

(7)
(8)

Contents

Terminology i

1 Introduction 1

1.1 The Context – Retrospective and Contemporary Perspectives on Trade . . . 1

1.2 Introducing the Case . . . 2

1.3 Discussing the Problem . . . 3

1.4 Purpose . . . 3 1.4.1 Research Questions . . . 4 1.5 Limitations . . . 4 1.6 Delimitations . . . 4 1.7 Outline . . . 4 2 Frame of References 7 2.1 Reduced Arbitrage from Information Asymmetry . . . 7

2.1.1 Richness and Reach . . . 7

2.2 The Connection Between Profitability and Capital Investments . . . 9

2.3 Supply and Demand . . . 10

2.4 Forecasting . . . 10

2.5 Service Level and Safety Inventory . . . 12

2.6 Lot Sizing . . . 13

2.7 Order Point System . . . 14

2.8 ABC Analysis . . . 14

2.9 Multiple Criteria ABC Analysis . . . 15

2.9.1 Dollar Usage . . . 18

2.9.2 Replenishment Lead Time . . . 18

2.9.3 Other Criteria . . . 18

2.10 Value Proposition . . . 19

2.11 Purchasing in a Strategic Perspective . . . 19

2.11.1 Category Management . . . 20

2.11.2 Vertical Integration . . . 20

2.11.3 Limiting the Amount of Suppliers . . . 20

3 Methodology 23 3.1 Research Process . . . 23

3.2 Methodological Approach . . . 24

3.2.1 Deductive and Inductive Reasoning . . . 24

3.2.2 Research Paradigm . . . 25

3.2.3 Case Study . . . 25

3.2.4 Literature Review . . . 26

3.3 Methodology for Data Collection . . . 26

3.3.1 Transaction Data . . . 26

(9)

3.4 Methodology for Analysis . . . 28

3.4.1 Criticism of the Analysis Method . . . 28

3.5 Discussion of Method . . . 29 3.5.1 Construct Validity . . . 29 3.5.2 Internal Validity . . . 29 3.5.3 External Validity . . . 29 3.5.4 Reliability . . . 30 3.6 Source Criticism . . . 30 4 Empirical Findings 31 4.1 Competitive Advantage and Value Proposition . . . 31

4.2 Financial Position . . . 32

4.2.1 Sector Positioning . . . 32

4.2.2 Income Statement . . . 33

4.2.3 Balance Sheet . . . 34

4.3 Inventory Control in the Value Chain . . . 35

4.3.1 Forecasting . . . 35

4.3.2 Safety Inventory and Service Level . . . 35

4.3.3 Current Lot Sizing Method . . . 35

4.3.4 Purchasing Process . . . 35

4.3.5 Inventory Location and Warehouse Management . . . 36

4.3.6 Pricing . . . 37

4.3.7 Suppliers and Inventory . . . 37

4.3.8 Suppliers for Drop Shipment . . . 38

4.3.9 Supplier Grade . . . 38

4.4 Data Extraction for the ABC Analysis . . . 40

4.4.1 The Storage of Inventory Data at Auctus . . . 40

4.4.2 Data Structure and Characteristics . . . 41

4.4.3 An Example from the Multiple Criteria ABC File . . . 43

5 Analysis, Constructing and Performing an MCABC Analysis 45 5.1 Design of the Multiple Criteria ABC Analysis . . . 45

5.1.1 Criterion – Dollar Usage . . . 45

5.1.2 Criterion – Order Frequency . . . 46

5.1.3 Criterion – Volume . . . 46

5.1.4 Criterion – Replenishment Lead Time . . . 46

5.1.5 Imposed Restrictions on the Multiple Criteria ABC Analysis . . . 47

5.1.6 Answering Research Question 1 . . . 47

5.2 Summary of the Multiple Criteria ABC Analysis . . . 47

5.2.1 A-articles – Capital . . . 51

5.2.2 A-articles – Inventory Space . . . 52

5.2.3 B-articles – Capital . . . 53

5.2.4 B-articles – Inventory Space . . . 54

5.2.5 C-articles – Capital . . . 54

5.2.6 C-articles – Inventory Space . . . 57

6 Proposed Solutions 59 6.1 Staying on Top in the E-commerce Competition . . . 59

6.1.1 Make Prioritization Clear in the Purchasing Process . . . 59

6.1.2 Eliminate Dead Articles . . . 60

6.1.3 Decouple the Supply Chain . . . 60

6.1.4 Determine Safety Stock According to Product Importance and Supplier Ranking . . . 61

(10)

6.1.6 Explore Possibilities of Extended Drop Shipment . . . 62

6.1.7 Answering Research Question 2 . . . 63

7 Conclusions and Future Work 65 7.1 Conclusion . . . 65

7.1.1 Generalizability of the Results . . . 65

7.1.2 Empirical Contribution . . . 66

7.1.3 Theoretical Contribution . . . 66

7.2 Future Work . . . 66

Appendices 73

(11)
(12)

List of Figures

2.1 The trade-off between richness and reach. . . 8

2.2 Illustration of DuPont identity. . . 9

2.3 Normal distribution of replenishment lead time. . . 13

2.4 The theoretical Pareto distribution. . . 15

3.1 Illustration of the research process. . . 24

4.1 Illustration of value chain for Auctus. . . 32

4.2 The Purchasing Process at Auctus. . . 36

4.3 The Pareto distribution of Auctus’ suppliers. . . 37

4.4 The process from customer order to accounting. . . 41

5.1 Illustration of the supplier’s reliability and importance for A-articles. . . 52

5.2 Illustration of the supplier’s reliability and importance for B-articles. . . 53

(13)
(14)

List of Tables

2.1 Customer order decoupling point, for different production approaches. . . 11

2.2 Illustration of the MCABC, part I. . . 16

2.3 Illustration of the MCABC, part II. . . 17

2.4 Illustration of the MCABC, part III. . . 17

4.1 Sector key figures from Swedish statistics. . . 33

4.2 Income statement of Auctus. . . 33

4.3 Balance sheet of Auctus. . . 34

4.4 The top four most frequent suppliers to Auctus. . . 38

4.5 Drop shipment by top four suppliers. . . 39

4.6 The qualitative grading of supplier lead time reliability. . . 39

4.7 An example of raw transaction data from Auctus. . . 42

4.8 Snapshot from the purchasing support file. . . 42

4.9 Snapshot from the file which will be the foundation of the MCABC analysis. . . . 43

5.1 The result of the MCABC analysis. . . 49

5.2 The MCABC analysis, focusing on maximum variable. . . 50

5.3 The MCABC analysis of the A-article, separated into product categories. . . 50

5.4 The MCABC analysis of the B-article, separated into product categories. . . 50

5.5 The MCABC analysis of the C-article, separated into product categories. . . 50

(15)

Chapter 1

Introduction

This master thesis begins with a presention of the context of an e-commerce retail business (e-tailer) and the differences between a store with online presence only and a business with physical stores only.1 The chapter then continues with introduction of the case and the challenges that it

is facing. Finally, the purpose and the research questions with limitations and delimitations are presented.

1.1

The Context – Retrospective and Contemporary

Per-spectives on Trade

Through all times, humans have traded with each other to maximize their wealth. Specialization and comparative advantage (Britannica Academic 2016) among humans and entities make it more efficient to focus on one activity, and trade the outcome of this activity to gain possession of other essential goods and services. In ancient times, this was done through barter trade and later on, different monetary systems started to act as mediators and securities of value, lowering the transaction cost of trade. The most recent defiant of transaction costs is the industrial revolution of the Internet; transactions between humans and entities are more easily performed than ever before. Çetinkaya and Lee (2000) claim that e-commerce constitutes a paradigm shift and that it can be classified as a disruptive innovation due to its impact on how people do business today. Since the invention of hypertext and the World Wide Web, or just the web, transaction costs of trade have decreased more, especially in the case of online banking and online retailing (Gunasekaran, Marri, et al. 2002).2 Further, the web testifies to the falsity of traditional thoughts about the trade-off

between “Reach and Richness” (Evans and Wurster 1997, p. 73). Where richness is about the information and reach is about the amount of people that are exchanging the information (Evans and Wurster 1997).

The advent of the web and the reduced transaction costs have gotten more and more entities to integrate the web into their business models. But the demand, and ability to provide greater reach and richness, put the supply chain, and more specifically the inventory control and purchasing, to the test. More articles need to be kept in stock for the e-tailers to provide the demanded richness. The competitive advantage of a retailer is still to match suppliers and customers to reduce their search cost, which the Internet enables with a greater reach than a brick-and-mortar retailer.

1The term e-tailer is used in several publications, for example Burt and Sparks (2003), Grewal and Levy (2007),

and Beldad et al. (2010).

2Transaction cost is defined by Coase (1937), research for which he was awarded the Sveriges Riksbank Prize in

(16)

Theories about inventory control and purchasing management handle items with predictable de-mand well and use marketing to affect the customer behavior. This is useful when the store competes on location, and the customer faces an extra search cost if he or she is to find another store. Then, the customer might be more susceptible to the advertisement in the physical store. But on the Internet, where the customer might more easily compare different options to a neg-ligible search cost, the demand for a certain good is not managed in the same way. Evans and Wurster (2000) argue that an e-tailer will not experience the same advantage of intensively adver-tising only one or a few products; as the customers might see through the offer when they have the ability to compare the same product with several other e-tailers. Evans and Wurster continue by writing that the strategy of the e-tailer will then be to advertise a greater amount of products in a more neutral way, as the online comparison will make the customer choose the best alternative possible.

A more unpredictable demand and the greater number of unique articles facing an e-tailer has the potential of binding an unnecessary amount of capital, if the inventory control and the purchasing process is not adapted well enough to fit the situation. Moreover, the bulkiness of an article is not paid enough attention – unless this dimension is considered, an e-tailer might add unnecessary fixed costs in the form of extra inventory space. The challenge of holding the products that the customers want, when the customers want them is still the topical issue of inventory control and purchasing. Under a more unpredictable demand, and for a great number of items, this becomes an even more intricate matter.

1.2

Introducing the Case

This thesis will use a single case approach in order to deepen the knowledge about inventory control for e-taliers with operations encompassing bulky articles. The specific case firm operates as an e-tailer, selling furniture and home furnishings. The company wishes to remain anonymous, which is why the company is given the fictive name Auctus (growth in Latin). The reason for this precaution is that the thesis will contain authentic information about the company, using a fictive name will obstruct the identification, and thereby minimize the risk of harming the business while this thesis will still be able to fulfill its academic obligations.

Auctus has experienced a period of vigorous growth, since the establishment less than a decade ago. Scott and Bruce (1987) and Churchill and Lewis (1983) assert that, in an initial phase of a company’s existence, attaining a greater market share and a broader sales base is crucial for the its survival. Establishing a trustworthy value stream, and a unique value proposition – that is, the value that a firm creates for its customers – is the main goal in this early phase according to Ma-hadevan (2000). These phenomena can be observed in the case of Auctus. The hunt for increased sales might have an accelerating effect on the costs as the focus on growth and a generous value proposition neglects the perspective on controlling costs connected to business activities. Later on, once the business has proven its profitability, consolidating costs and improving cost efficiency become more vital as sales are sufficiently large for the firm’s existence not to be jeopardized in a near future, according to Scott and Bruce (1987) and Churchill and Lewis (1983). In the autumn of 2015 Auctus announced for students to help with cost control, which is in line with the theoretical prediction of a company’s growth.

Today, Auctus is in need for a more systematic approach to inventory control and purchasing. At present, there are three main symptoms that can be traced to the performance of the inventory control and purchasing. Firstly, supply delivery sometimes cannot unload goods at the Auctus warehouse due to the fact that the warehouse is out of available space. There is a lack of commu-nication between the purchasing management and the warehouse management, as the purchasing management procures and purchase whatever articles they think will answer to the customer de-mand, no matter what space that is currently available in the warehouse. Secondly, a large part

(17)

of Auctus’ sales are induced by discount campaigns. This tool is especially common for products that have not sold well in the past. Due to the independence of the purchasing department, and sometimes misleading information from the purchasing computer software, the purchase manage-ment are lured into initiating purchase orders for products running on a discount campaigns. The software then indicates an abnormally high demand and a dangerously low stock level, when the real reason for the discount campaign was to get rid of the product from the inventory. The result is that a product, which was about to be removed from the online store once the stock level was close to zero, is purchased once more. Finally, Auctus has a lower inventory turnover ratio than comparable companies in the same sector (SNI 47919).3 This finding is supported by Auctus’

in-come statement – the trend over the past three years indicates a greater increase in Raw material and consumables than in Net sales. The two income statement items are diverging when they – in the ideal case – should converge, or at least keep a constant relationship. The divergence of the income statement items indicates that Auctus has sold more at the expense of a larger, and possibly more inefficient designed inventory. A first action to understand why the symptoms arise would be to categorize the inventory according to a multiple criteria ABC analysis, which is an established tool for inventory management. The multiple criteria ABC analysis will be used as a base for a further investigation on how to come to terms with the symptoms.

1.3

Discussing the Problem

New sales channels and new ways of communicating have led to a broader scope of information being attainable to a greater amount of people, that is, a substantial extension of richness and reach. Apart from the benefits that these dimensions bring to the consumers, they also create challenges for online businesses. The reduced asymmetry of information due to the possibility for customers to easily compare products, assortments, delivery lead times and prices online, imply that a successful e-tailer needs to offer low prices and a broad range of products with high responsiveness. Greater pressure to provide extensive richness increases the pressure to keep more articles in stock, thereby tying up a larger portion of a company’s capital. Housing larger-sized products augment inventory costs even more, why a clever inventory management and purchasing strategy is required. Structures, systems, and strategic alliances, therefore, require a design that secures a proper execution of the value proposition. This thesis seeks to propose solutions for how to systematically balance richness and reach, by using a multiple criteria ABC analysis as a basis for decisions concerning the selection of suppliers and how to treat the relationships strategically. The proposed solutions will make the purchasing process more effective. Where effective refers the process’ exposure to unnecessary risk. The underlying idea is that businesses make money by taking on risk, for example in the supply chain, that the customer is willing to pay for, but there might exist other risks in the business that the customers are unwilling to pay for, so called unnecessary risk. When this unnecessary risk is reduced one can argue that the purchasing process operates in a more effective manner.

1.4

Purpose

The purpose is to explore how an e-tailer selling bulky articles can use a multiple criteria ABC analysis to make its purchasing process more effective while balancing richness and reach, with the performance measurements of profitability, total asset turnover, and inventory turnover. The purpose will be accomplished through a single case study on an e-tailer, active on the Swedish furniture and home furnishing market.

(18)

1.4.1

Research Questions

The following research questions are a deconstruction of the purpose. Research question 1 focuses on construct validity, hence, how the multiple criteria should be designed to measure the purchasing process and make it more effective. Research question 2 exists to investigate what the actual actions should be to facilitate a more effective purchasing process.

1. What should be the design of a multiple criteria ABC analysis to evaluate the purchasing process of bulky articles?

2. What will be the recommendations for the purchasing process of an e-talier to more effectively balancing richness and reach with the financial performance measurements?

1.5

Limitations

A limitation is an external restriction, a synonym would be the specific circumstances of this case which the authors cannot change. Here it is important to highlight that research question 1 asks how the multiple criteria analysis should be designed to evaluate the purchasing process of bulky articles. The design of the multiple criteria ABC analysis requires that yet unknown circumstances about data characteristics are considered. As a consequence, such circumstances about data will be communicated in chapter 5 ‘‘Analysis, Constructing and Performing an MCABC Analysis’’.

• Data that is available today goes back to May 1 of 2013, why sophisticated statistical predic-tions based on such a short time period will not be reliable. Instead Auctus uses a moving average as an approximation of the demand.

• Data available is of a varying quality, which will affect the working process of this thesis. In exactly what way will be answered through research question number one.

1.6

Delimitations

Delimitation are the restrictions made by the authors. This is different from the external restric-tions imposed on this thesis, presented above as limitarestric-tions.

• This thesis treats Auctus’ different stock point as one, but this will not affect the purpose while it will ease the authors’ analysis process. The reason for this is that the purchasing is a central function and the majority of articles are stores in a few locations geographically close to each other.

• The second symptom mentioned in the introduction of this thesis – the symptom about discount activities and perceived demand – will not be touched upon in this thesis. The symptom is more likely to be solved by investigating a different enterprise resource planning (ERP) system, than performing a multiple criteria ABC analysis.

1.7

Outline

The remaining part of this thesis consists of seven chapters, which are briefly explained be-low.

(19)

Chapter 2 – Frame of References Theories are presented about e-commerce along with the special implications that this type of context brings, and the identified gap in the literature. Literature concerning inventory control and purchasing is also described and compared.

Chapter 3 – Methodology The chapter begins with a presentation of the research process as an overview of all the incorporated steps, and where in the process that the research questions are answered. Later on, the choices of research methodologies are explained and motivated and finally the quality of the research is discussed.

Chapter 4 – Empirical findings The chapter includes a compilation of the empirical findings, beginning with a more detailed description of the case company, its context and operations. In this chapter, the case company’s reasoning and current situation regarding inventory control and purchasing are explained as well as the results from the multiple criteria ABC analysis.

Chapter 5 – Analysis, Constructing and Performing an MCABC Analysis The chapter connects the empirical findings to the theories from the literature review in a multiple criteria ABC analysis. The results from the multiple criteria ABC analysis are then evaluated together with the qualitative data.

Chapter 6 – Proposed Solutions The chapter discussion of the analysis made in Chapter 5 and proposals of what measures that an e-tailer in Auctus’ situation should take in order to tackle the issues highlighted in the analysis.

Chapter 7– Conclusions and Future Work The last chapter connects the proposed solu-tions to the purpose of the thesis and thereby presents the conclusions. Future research is also discussed.

(20)
(21)

Chapter 2

Frame of References

This chapter of the thesis will present the theoretical framework. It starts with an introduction of the e-tailer environment (Section 2.1) and compare it to the traditional retailers operating thruogh traditional physical shops. Next, the basics of firm profitability will be explained in Section 2.2. The remaining parts of this chapter will be used to discuss the traditional theories of inventory control and purchasing that are relevant to bridge the discussed problem.

2.1

Reduced Arbitrage from Information Asymmetry

Evans and Wurster (1997) write that customer relationships and supplier relationships – actually every kind of relationship in the value chain – in reality, is the possessed information about; for example, a customer, a company, a supplier or a product. Information asymmetry often determines the bargaining power in a buyer-seller relation. One of the most famous articles about information asymmetry is “The Market for ’Lemons’: Quality Uncertainty and the Market Mechanism” by Akerlof (1970). In the article, Akerlof shows that the market for buying a used car is very much determined by the asymmetry of information between the buyer and the seller. The seller can exploit the asymmetric information about the true condition of the car, and the buyer faces the option of trusting the seller or investigating every other option in the market. Where the last alternative has traditionally been a time-consuming task. Many businesses exploit this information asymmetry; Evans and Wurster argue that the asymmetry in many cases constitutes the competitive advantage of a company. This has been true as long as the information is restrained to flow where the information carrier goes, like a sales person or a direct email. Once actors in the value chain and the market are electronically connected, Evans and Wurster explain that information can travel by itself.

2.1.1

Richness and Reach

The definition of reach by Evans and Wurster (1997) is the number of people exchanging infor-mation with each other, while richness is separated into three categories:

• Bandwidth – the quantity of information transferable from sender to receiver in a given moment.

• Customized – How well adapted the information is to the receiver. Compare a TV commercial and a face-to-face sales pitch.

(22)

Grewal and Levy (2007) highlight other aspects of reach of businesses online. The Internet makes it possible for a company to reach more customers, and in turn, more customers are able to reach them. Not only does e-commerce imply that a larger number of people are able to access the information about a company’s offerings, but they are able to do so at any moment of the day or at any place. The advantages of richness and reach compensate for possible disadvantages of the online channels such as lack of personalized human contact, assessment the products in reality before acquiring them, and eventual additional costs for after-sales service, such as returns. The lack of pre-purchase trial affects standardized products, such as books and music less, than non-standardized products, for example, clothes, toys or furniture.

The definitions of richness and reach imply that ‘‘rich’’ information has to be presented in a special way, due to constraints of bandwidth, customization, and interactivity. Rich information has for example been presented at customer evenings or special advisory meeting. Figure 2.1 shows the trade-off between richness and reach. However, new ways of communicating, such as the Internet and the web make it possible to provide rich information and reach many people at the same time. (Evans and Wurster 1997)

Reach Richness

Traditional trade-off

New ways of combining richness and reach

Figure 2.1: The trade-off between richness and reach. The Figure is inspired by (Evans and Wurster 2000, p. NA, Figure 3-3).

While many customers are able to visit a brick-and-mortar store, it is hard to beat the twenty-four seven availability from a computer screen located anywhere around the globe. At the same time, information is getting richer as the customers accept ‘‘cookies’’ to be used for customization of offers. Schafer et al. (2001) state that greater richness does not have to mean that a company needs to necessarily develop more products to meet the many and diversified needs of customers, however, online sales have opened up an opportunity to provide customers with more choices. They compare this to a superstore selling thousands of books, as opposed to an online store where the customers may choose from millions of books. Grewal and Levy (2007) also emphasize the impact and importance of providing a vast range of articles as an e-tailer. The authors describe this in relation to the disadvantages that a buyer experiences when making their purchases via the Internet versus at a physical location.

Gunasekaran and Yusuf (2002) as well as Gunasekaran, Hung Lai, et al. (2008) have made similar observations about the increased customer influence that is becoming more present in the 21st

century. Richness and reach have driven companies to focus even more on customer demand, increasing the pressure on agile value streams. They state that in order to achieve agility in an organization, proactive planning is needed to meet changes in market and customer demands, maximize customer service level while minimizing the cost of goods. The objectives of this are to increase competitive advantages in a global market, and better the chances of long-term sur-vival.

(23)

2.2

The Connection Between Profitability and Capital

In-vestments

The main goal of a company is to create a return for its investors. For a firm to achieve long-term success and create the desired return, it needs a positive net profit, which is calculated as:

Net Profit = EBIT − Interest − Taxes (2.1)

Where EBIT is defined as:

EBIT = Revenue − (Operating Expenses + Non-Operating Income) (2.2) Skärvad and Olsson (2013) write that the net profit can be increased in two ways. On the one, hand the company may enhance its net sales, on the other hand the company can reduce its expenses. The focus might shift over the lifetime of a company.

It is natural for an average large company to achieve a bigger net profit than an average small company does. Because a large company can invest more resources, the output of it should be higher. This does not say anything about how efficiently the input is used. Hence, to broaden the financial analysis of a company, not only absolute figures, like the net profit, are important, but relative performance measurements too. A DuPont analysis provides such a tool of relative performance. The DuPont identity deconstructs the profitability of a company, where the net profit is one component. The DuPont identity (Figure 2.2), also shows the connection of inventory and profitability. That connection is especially important to this thesis.

Direct Labor Factory Overhead Material Finished Goods Work in Process Semi-finished Raw Materials General Expenses Admin. Expenses Selling Expenses Costs of Goods Cash Accounts Receivable Total Inventory

Sales Total Costs Fixed Assets

Current Assets

Profit Sales Sales Total

Assets

Profit

Margin Asset Turns

ROA

− +

÷ ÷

×

Figure 2.2: Illustration of DuPont identity, depicting how the income statement (left leg) and the balance sheet (right leg) construct the performance measurement return on asset (ROA). The Figure is inspired by (Sanderson 1997, p. 18.29, Figure 18.6).

(24)

The relative performance measurement, return on assets (ROA), indicates how well the assets are used to generate a return to the investors. The relationship between return on assets and investments in inventory is explained by the DuPont identity and the following equation, by Olhager (2000), will highlight it.

ROA = ProfitSales × Sales

Total Assets=Profit margin × Asset Turn (2.3) The return on asset can be deconstructed even further, following the DuPont identity upstream. Despite a company having an acceptable profit margin, ROA might be unsatisfactory due to a low asset turnover. In a retail business the asset turnover is mainly improved through focusing on the current assets and more specifically an increase of the inventory turnover ratio (ITR), equation by Olhager (2000):

ITR =(Net) Sales

Inventory (2.4)

The asset turnover is directly connected to the profitability of an entity, which has been shown with Equation (2.3). Hence, the purpose of this thesis is in line with the well spread and recognized the idea of profit maximization as the overall goal for a firm. While the main RQ of this thesis seeks to maximize the net profit through cost control by focusing on improving the inventory turnover rate while maintaining the current service level.

2.3

Supply and Demand

Managing material is about balancing requirements of supply and demand. If the demand is greater than the supply, then a manufacturing or purchasing order must be initiated to meet the demand. If the opposite situation occurs, planned and released orders will have to be delayed or the demand manipulated with a discount campaign, for example. By keeping an inventory, Jonsson (2008) state that it is possible to decouple supply and demand, hence, lower the probability or eliminate the unwanted and costly situation when supply and demand differ. Further, Sanderson (1997) describes six functions of inventory that all aim to decouple supply from demand, Silver et al. (1998) describe six functions too, but do not use the same terminology or meaning. Olhager (2000), on the other hand, describes seven types of functions for an inventory to fulfill. Olhager also makes a general comment about inventory – that there is an immense amount of reasons for keeping an inventory, and the name of the inventory is, therefore, multifold as well. This is why the authors of this thesis have made a selection of the inventory functions presented by Olhager and Sanderson. The selected inventory functions are supposed to be important to the specific circumstances of this thesis and the case company Auctus. This thesis will touch upon Buffer stock inventory (or safety stock) in Section 2.5, and lot size inventory in Section 2.6.

2.4

Forecasting

The aim of inventory control, according to Axsäter (1991), is to maintain an appropriately sized inventory. Further, Axsäter argues that the replenishment lead time is usually too long compared to the promised customer order lead time. The consequence – which has been touched upon in Table 2.1 on the facing page – is that the company needs to hold an inventory and make use of forecasts, in the attempt to predict the demand. De Lurgio et al. (1997) state that forecasting systems have the combined purpose of meeting short-term demand and evaluating the future customer requests for the development of new products. That is why it should be integrated into the daily operations, as well as the strategic planning.

Olhager (2012) says that the Customer Order Decoupling Point (CODP) marks the position in the value chain where the products are linked to a specific customer. Olhager also refer to the

(25)

researcher consensus view of the CODP marking the transition from anticipated demand to actual demand, presented in Table 2.1, for various production approaches. Hence, upstream the CODP position, forecasting is the tool used to estimate the demand. While downstream the CODP position, deterministic demand can be used to control the value creating processes in the value chain. In his article, Olhager (2012) states that retailers can be seen as applying Make-to-stock (MTS). Even though the production, in this case, equals the purchasing process. As can be seen in Table 2.1, the CODP for an MTS approach is positioned far down the value chain close to the customer. Therefore, forecasting of upstream demand is important, especially to a retailer.

Table 2.1: Customer order decoupling point, for different production approaches. The dotted lines indicate upstream activities, the solid lines indicate downstream activities. Inspired by (Olhager 2012, p. 38, Figure 1.).

Customer order

decoupling point

Engineer Fabricate Assemble Deliver

Make-to-stock . . . CODP Assemble-to-order . . . CODP Make-to-order . . . CODP Engineer-to-order CODP

De Lurgio et al. (1997) highlight some of the issues that often occur in forecasting systems. Firstly, it needs to be kept in mind that shipments and demand are not exactly the same, since the number of shipments can be affected by stockouts, delays or substituted products and so on. Thus, it is important to consider these factors when determining the demand. Secondly, promotions and discounts need to be registered in order to evaluate the influence it has had on the demand. Thirdly, the impact of promotions and outliers should be documented separately to facilitate the analysis of demand figures. Also, De Lurgio et al. (1997) say that all of the forecasts should be made independently of other articles. Lastly, the detection of the different phases in the product life cycle is important. Such a feature in a forecasting system is based on statistical methods, which means that accurate input data is required.

According to Axsäter (1991), many forecasting methods are based on an extrapolation of past demand, while this can lead to inaccurate results. Known factors, which not yet has had an influence on customer demand, might have a high impact in the future. Such factors are difficult to include in a computer-based system, why there sometimes is a need to manually determine future demand. Therefore, the system should be constructed in a way that allows manual adjustments of the forecasts. Some situations when this might be needed, is when there are changes in price, new rules and regulations take effect, or when new products are introduced to the market for which there is no historical data. However, De Lurgio et al. (1997) claim that methods using historical data are the most common when it comes to forecasting methods. The assumption then is that yesterday’s demand patterns paint a good picture of what the demand will look like tomorrow. De Lurgio et al. (1997) define some of the most customary methods, which are presented below. Simple moving averages are calculated several times over a varying time period. For example, one may wish to compute the average over a certain period of a year, for products that are particularly popular over that time period. Equation (2.5)

Ft+1= At= Yt+ Yt−1+ ... + Yt−n+2+ Yt−n+1 n = Pn i=t−n+1Yi n (2.5)

(26)

where

At=average through periodt

Ft+1=forecast for periodt+1

Yt=sales in periodt

Yt−1=sales in time periodt−1

As an alternative to the simple moving average, there is the weighted moving average, where different time periods are considered more important than others. A common separation is to give the latest time period more weight than previous ones, since the more recent figures could be assumed to be a better estimation of the current demand. The formula for the weighted moving average is presented below:

At= wtnYt+ wt−1nYt−1+ ...wt−n+1Yt−n+1= n X i=1 wt−i+1Yt−i+1 (2.6) where

wt=weighting factor for the time t,

1 = wt+ wt−1... + wt−i+1

2.5

Service Level and Safety Inventory

Inventory can have the function of a buffer, also known as safety stock. The safety stock should cover the differences between actual demand and forecasted demand. It might also aim to deal with the problem of differences between actual and planned lead time. In addition, a safety stock reduces the probability of stockout and helps the company to achieve a high service level. (Sanderson 1997)

To decide on a proper level of safety stock, it can be set in accordance with a service level or shortage cost model. The most common way is to use a specific service level (SL). The SL can be defined in various ways (cf., Olhager 2000; Sürie and Reuter 2015; Anupindi et al. 2014), but the SL most frequently occurring relates average lead time demand (LTD) and reorder point level (ROP). In text Olhager (2000) and Sürie and Reuter (2015) translates the service level into: SERV. 1 or Cycle service level, which indicates the probability that an order can be completely met by the inventory under an order cycle. A more formal expression of the SL is:

SL = P (LTD ≤ ROP ) (2.7)

Where LTD is the average lead time demand and ROP is defined as:

ROP =Average lead time demand + Safety Stock = LT D + ISaf ety (2.8)

Anupindi et al. (2014) write that LTD is commonly assumed to be normally distributed. An extension of Equation (2.7) result in the following expression:

SL = P (LTD ≤ ROP ) = P (Z ≤ z) (2.9)

In this case Z represents the average lead time demand, which is a stochastic variable with mean 0 and standard deviation 1. The safety stock is implicitly represented by z, via eq. (2.8). Figure 2.3

(27)

ROP LTD

fLTD(x)

LTD

Figure 2.3: Under the assumption that the replenishment lead time corresponds to a normal distribution; the shaded area is then the probability that the reordering point will be sufficient to avoid stock out in a desired ratio of the outcomes. The figure is inspired by (Anupindi et al. 2014, p. 180, Figure 2).

on the facing page shows a visual illustration of Equation (2.9). The shaded area equals the probability of the average lead time demand (LTD) being less than, or equal to, the reordering point.

The safety stock, or safety inventory is expressed by Anupindi et al. (2014), as: Isaf ety= z × σLT D⇔ z =

Isaf ety

σLT D

(2.10) These equations presented above, eqs. (2.8) to (2.10), can hence be used to calculate service level SL from safety inventory Isaf ety and vice versa, as long as average lead time demand LT D and

the standard deviation of average lead time demand σLT D is known.

Previously the replenishment lead time was assumed to be deterministic and not stochastic. But this is not always true and Anupindi et al. (2014) presents the equation, dealing with two inde-pendent and stochastic variables.1

σ2LT D= LσD2 + Dσ2L ⇔ σLT D= q Lσ2 D+ Dσ 2 L (2.11)

Where L is the replenishment lead time, σL is the standard deviation of the replenishment lead

time, D is the demand during the period, and σD is the standard deviation of the demand.

Calculating the service level and the safety inventory is no different once the standard deviation of the average lead time demand is calculated.

2.6

Lot Sizing

Lot size inventory or cycle stock connects to the quantity used between refilling the inventory. The goal of lot size inventory is to minimize the total costs of carrying the inventory and the cost of purchase for each order, to do this one has to take into account the trade-off between the two. By decreasing the cost of replenishment, it is possible to order a smaller lot size and reduce the inventory investment. (Sanderson 1997) In other words, increase or optimize the ITR. Some examples are presented below

1The authors of this thesis have replaced Anupindi et al.’s (2014) demand rate R with the more commonly used

(28)

Economic order quantity (EOQ) is a very useful formula when the correct data is available. (Ol-hager 2000) In this case and thesis, such data will not be attainable and the EOQ formula will thus be left outside this theoretical framework.

Estimated Order Quantity is one of the most unstructured methods of determining lot sizes is estimated order quantity. It is based on tacit knowledge and experience of what quantities that are suitable for the specific occasion. (Jonsson 2008)

The lot-for-lot method implies that there actually is no determined lot sizing in reality. Instead, quantities are assessed and re-sized to meet the required quantities each time an item is ordered. The lot-for-lot method is mostly used for customer order controlled material flows, costly products or components in contexts with small set-up costs. (Jonsson 2008)

In the case of using estimated run-out time, the lot size is determined so that the supply is sufficient to meet the demand for several planning periods, for example, weeks or days. As in the method for estimated order quantity, the decisions can be based on experienced estimations; nevertheless, they can also be economically calculated. Either way, the order quantity is defined as a run-out-time stated in a number of periods, and adjusted for each order according to the demands for the relevant periods. (Jonsson 2008)

2.7

Order Point System

Order point systems are the most common method for handling stocks of independent articles. (Olhager 2000) In a recent survey conducted by Jonsson and Mattsson (2014), it is stated that 75% of the distributing companies in Sweden use this type of tools for inventory control. The system’s function is to decide and signal when it is time to place a new order for an item based on stock level, demand, and associated costs. The stock-level not only refers to the physical inventory, but also encompasses outstanding orders – orders that have already been made, and back-orders – orders for items that are out of stock at the supplier. Aggregated quantity is usually called the stock position, calculated as in formula Equation (2.12). (Axsäter 1991)

stock position = physical stock level + outstanding orders − back orders (2.12) The stock position can be monitored either continuously or periodically, the former meaning that an order is placed as soon as the stock level has dropped to a certain point. The latter method means that the stock level for an item is reviewed only at certain times, and then possibly ordered if needed. The approaches are suitable for different cases. Periodical monitoring increases the need for safety stock while continuous inspection decreases the requirement, and is preferred when it is sought to coordinate orders of different products. In general, this alternative is better economically, since it yields less cost for the control, especially for items with high turnover rate. On the other hand, continuous inspection of an article with a low turnover rate only implicates higher needless costs. (Axsäter 1991)

2.8

ABC Analysis

Flores (1987) states that the foundation of ABC analysis can be traced back to at least the time when Pareto published his famous paper Cours d’économie politique (1896) where he observed inequality of the wealth distribution of economies, also known as the 80/20-rule, see Figure 2.4 on the facing page for a visual explanation. The Pareto ratio can also be found in a majority of companies, where approximately 20 percent of the products represent 80 percent of the total annual dollar usage and vice versa. This relationship proposes that the different items should be treated differently, and more attention should be put on the most revenue-generating articles.

(29)

(Silver 1991; Olhager 2000) An example of dividing the different classes is to identify A-items as the top 20% of the products, the next 30% as B-items, and the bottom 50% as C-items. However, the values of these boundaries and the numbers of them are not fixed and can be adjusted according to the context. As a final step, it needs to be assured whether the corresponding managerial implications are possible from an economic perspective. (Millstein et al. 2014)

20% 80% Number of Items Total V alue

Figure 2.4: This plot depicts the theoretical pareto distribution, in this fictive case, 80% of the value is provided by 20% of the items.

A simple classification can be made, usually concentrating only on dollar-usage or cost in an inventory context. However, in many cases, several criteria need to be included in order for management to make the right decisions. Flores and Clay Whybark’s (1986) Joint Criteria Matrix is an example of a tool assessing items based on two criteria. Lead time inaccuracies, the certainty of supply, and impact of stock out of the items may be critical factors to take into account (Flores 1987). Another concern may be the complex connections between the items, such as dependent articles, meaning for example that an A-article is always sold in conjunction with a C-article. Newly introduced products should be classified as A-articles to evaluate their progress. (Aronsson et al. 2004)

2.9

Multiple Criteria ABC Analysis

In his article from 2007, Ng presents a model for handling a multiple criteria inventory classifi-cation, which transforms all the criteria values into a single scalar score. This model can include more categories than for example the Joint Criteria Matrix introduced by Flores and Clay Why-bark (1986). Ng’s model is also claimed to be less time consuming when treating a vast number of articles than for example Ramanathan’s (2006) model, which requires the solving for the optimal weight through linear optimization for every article individually. Further, the simplicity of Ng’s (2007) model makes it easy to implement in a spreadsheet program.

Ng’s model classifies the number of products I, considering the number of categories J. The value of the item i under the criteria j is denoted as yij. All measures are presumed to relate positively

to the final score of a product. Therefore, negatively related measures need to be converted before using them in the model.

(30)

Before beginning the model process, proposed by Ng, it is required to turn all the measures into the same base so that they are comparable. This is done by using:

xij =

yij− mini=1,2,...,I(yij)

maxi=1,2,...,I(yij) − mini=1,2,...,I(yij) (2.13)

Equation (2.13) transforms all measures to a number between 0 and 1. Then, the analyst needs to make a ranking of the criteria. Ng admits that this step is not completely objective, but states that objectivity in this case is less important than in Analytic Hierarchy Process (AHP), for example. In Ng’s model, the decision maker only needs to decide if a criterion is more important than another one, i.e. the sequence. This is a simplification from many other multiple criteria ABC analyses (MCABC), which requires an exact weight assigned to each criterion.

The positive weight wij is defined as the contributing weight of performance for the ith article

under the jth criterion. Further, it is assumed that the different criteria are related as follows; wi1≥ wi2≥ . . . ≥ wiJ.

Ng’s model includes the following steps: 1. Compute all the partial averages, 1

j

Pj

k=1xik, j=1,2,...,J.

2. Localize the maximum among the partial averages. The matching value is the score Si of

the ith item.

3. Order the items according to their score.

4. Divide the items into groups in accordance with the principles of ABC analysis.

An example of Ng’s model is presented in Tables 2.2 to 2.4. Before starting any step of the model, the data needs transformation and Table 2.2 summarizes the initial manipulation of data, as it lists the items i of a pet shop with corresponding measures for each criterion yik. Their dollar

usage is denoted as yi1, order frequency as yi2 and lead time as yi3. Equation (2.13) is used to

transform these measures, which are presented as xikwhere k ∈ {1, .., 3}.

Table 2.2: A silly example of Ng’s model, yikare random numbers.

Pet store

item, i Dollar us-age, yi1

Order fre-quency, yi2 Lead time, yi3 Dollar us-age, xi1 Order fre-quency, xi2 Lead time, xi3

Transformed numbers, eq. (2.13)

Tame Impala 17 158 48 23 0,979 0,978 0,765 Vicuña 9 465 9 27 0,524 0,070 1,000 Lama 15 805 44 10 0,899 0,884 0,000 Hamster 15 200 38 15 0,863 0,744 0,294 Hippopotamus 13 888 18 23 0,786 0,279 0,765 Blackbird 627 32 16 0,000 0,605 0,353 Parrot 7 166 49 22 0,387 1,000 0,706 Zebra 5 691 25 19 0,300 0,442 0,529 Cow 15 953 6 26 0,908 0,000 0,941 Chimpanzee 17 509 30 22 1,000 0,674 0,706

In Table 2.3 on the facing page, the results of step 1 and 2 in Ng’s model are presented; the calculated partial averages of the criteria, and the maximum value, which is equal to the score Si.

(31)

Table 2.3: The partial averages and maximum, form the example in Table 2.2.

Pet store item 1 1xi1 1 2(xi1+ xi2) 1 3(xi1+ xi2+ xi3) Max( 1 j Pj k=1xik) Tame Impala 0,979 0,978 0,907 0,979 Vicuña 0,524 0,297 0,531 0,531 Lama 0,899 0,891 0,594 0,899 Hamster 0,863 0,804 0,634 0,863 Hippopotamus 0,786 0,532 0,610 0,786 Blackbird 0,000 0,302 0,319 0,319 Parrot 0,387 0,694 0,698 0,698 Zebra 0,300 0,371 0,424 0,424 Cow 0,908 0,454 0,616 0,908 Chimpanzee 1,000 0,837 0,793 1,000

Finally, step 3 and 4 are illustrated in Table 2.4, where the items are arranged according to their tallied scores. These are then divided into groups of A, B and C articles, based on suitable limits for the different classifications.

Table 2.4: A sorted list of the maximum values, A-items equals 20% of the top animals, B-items equals 30% of the animals, and C-items equals the last 50% of the animals.

Pet store item Max Classification

Chimpanzee 1,000 A Tame Impala 0,979 A Cow 0,908 B Lama 0,899 B Hamster 0,863 B Hippopotamus 0,786 C Parrot 0,698 C Vicuña 0,531 C Zebra 0,424 C Blackbird 0,319 C

As mentioned above, and suggested by theory, products from different classifications should be managed differently. The general idea is that A-items should receive the most amount of attention and be individually monitored, due to their high ranking in importance. (Silver 1991; Sanderson 1997) This means higher demand on accurate inventory documentation, and attempts should be made to decrease lot sizes, as well as shorten lead times. Sanderson (1997) and Aronsson et al. (2004) says that A-items should be ordered frequently and corresponding safety stocks should be kept low if they are expensive items. Strong relationships with the suppliers should be sought after, to assure that supplies are consistent and sufficient. (Aronsson et al. 2004) Due to their vitality, top management should be frequently informed about the performance of the A-items, and parameters should be closely monitored and re-assessed. Estimations of demand and attempts to influence it can also be beneficial from an inventory control perspective. Further, stockouts of A-items should be avoided, or at least, asses regarding the impact of a possible shortage. However, it needs to be kept in mind that although an article may receive a high score in a multiple criteria ABC, the underlying reasons need to be recognized. For example, article X may have a high demand and low value while article Y has low demand and high value, these two may need to be managed in very dissimilar ways. (Silver 1991)

(32)

B-items are the second most important articles, and therefore, need a reasonable but appreciable amount of resources spent on them. (Silver 1991) This is due to the nature of this item classification – not pertaining to either end of the extremes, there are not a lot of available literature on how to monitor and attend them, apart from that it should not be as detailed as the control for A-items. Nor should it be as loose and generic as for C-items.

Finally, the C-items – the lowest in the ranking, should be handled in a much simpler manner. For C-items, it is recommended to find as many common traits as possible, for example, usage rates, seasonal patterns, common suppliers and lead times. Thereby, decisions for several articles can be processed simultaneously. (Silver 1991) Aronsson et al. (2004) suggests that these items should be ordered less frequently, but have a larger safety stock than that of A-items. Further, no classifications are static and can change over time. It is therefore required to re-evaluate them from time to time. (Silver 1991)

The following paragraphs define and explain relevant criteria on which a multiple criteria ABC analysis can be based upon, more specifically the criteria that have been elected for this particular research problem. The criteria are described in order for the reader to better comprehend the impacts and implications that these choices might have.

2.9.1

Dollar Usage

Olhager (2000) defines the dollar usage for a product is defined as:

dollar usage = yearly usage × unit cost (2.14)

This measurement is commonly used to separate different articles since they can require different planning tools and managerial approaches. For example, a high volume often means that the volume also is more even. Another implication might be differences in the frequency of stocktaking. Articles with a high dollar usage should be audited more often since they represent a larger portion of the capital tied up in inventory. Stocktaking frequently also means that it can be made for different classifications at different times – so-called cyclic stocktaking, which an be a better way of using resources.

2.9.2

Replenishment Lead Time

There are several types of lead times. However, in the multiple criteria ABC analysis of this report, the focus lies on the replenishment lead time, which Silver (1991) define as the time from which the company places an order until the goods are delivered to the physical inventory. A mathematical definition of replenishment lead time variability is give in section 2.5.

2.9.3

Other Criteria

Based on the description of the purchasing process, presented later in Section 4.3.4 on page 35, bulkiness and order frequency are believed to be important criteria by Auctus. In this thesis, the definition of bulkiness is the size of a specific item. It can be measured by the pallet type that a product is stored on, or the volume of the product.

Order frequency is defined as the total number of orders of a specific item since the product was introduced, but it does not contain any information about what quantity each order holds. It is interesting to compare this variable to the total quantity demanded of a specific item. A low order frequency and a high demanded quantity differentiate a product from another article with

(33)

the equal amount of orders and quantity. In the first case, one order holds a high quantity; in the second case, every order holds only the quantity of one article.

2.10

Value Proposition

Mohr and Sengupta (2015) state that an important part of forming a firm’s strategy is to define what value it should create for its customers, also known as the value proposition. Not only is it important to concentrate on what customers want, but also what value that competitors offer. That should include companies that pose a threat both directly and indirectly since competition can arise outside of an industry’s boundaries.

Further, Mohr and Sengupta (2015) state that it needs to be secured that the strategy can be executed well. This puts pressure on acquiring the right competencies, suitable structures and systems, distribution decisions, pricing, and where to promote the products. However, the au-thors stress the importance of staying flexible and not locking oneself in an inflexible program. Fast changes in the market may redefine what value proposition that customers consider to be important, and thus the composition of execution requirements as well. Another component in this mix is effective management of alliances and partnerships, which also includes the vendor selection which is discussed in section 2.11.

2.11

Purchasing in a Strategic Perspective

Many researchers agree that purchasing has taken on an increasingly more important role in supply chain management (Chen et al. 2004; Knoppen and Sáenz 2015; Kraljic 1983). Knoppen and Sáenz (2015) highlight the purchasing function’s impact on a company’s long-term goals. Not only is it the key to attaining economically advantageous procurement deals, but it also contributes to the company strategy by its influence on new solutions and adaptions to customer needs. Depending on a firm’s profile and competitive advantage, the purchasing function can take on different roles. For instance, if a company competes through low prices, this needs to be reflected in the purchasing operations. Knoppen and Sáenz (2015) state that if strategic purchasing is not implemented in a satisfying manner, it may have a negative impact on performance due to decreased competitiveness, both in a short- and long-term perspective.

Kraljic (1983) explains that companies sourcing globally have to accept the uncertainties that come with it. Instead of passively dodging situations, firms must instead take a more active stance and learn how to turn challenges into opportunities. Therefore, Kraljic argues that purchasing should be treated as a strategic part of management instead of merely an operating function. Kraljic (1983) stresses the strategical view on purchasing by renaming it as supply management and says that the greater the uncertainty of for example supplier relationships and physical availability of items, the more important the supply management. Further, Kraljic claims that the importance of supply strategy depends on two factors, which are: (i) the importance of purchasing in terms of its impact on strategy, the total costs of materials and their impact on profitability, and so on; and (ii) the complexity of the supply market including scarcity of products, development pace of technology or new substitutions, entry barriers, logistics costs or complexity, or number of competitors. In a study made by Thompson (1996), it was concluded that effective purchasing should have a long-term cost focus, meaning that efforts should be directed at evaluating the total cost of acquisition rather than attempting to optimize every single transaction. Weber et al. (1991) state that it is impossible to provide low cost, quality products, without suppliers that fulfill the organization’s requirements. In addition, in the case of many firms, purchases from suppliers constitute a large portion of their total operating cost. Thereby, electing the right suppliers and

(34)

protecting the relationships with them is essential. However, the many criteria that need to be evaluated, make it a complex and difficult task.

2.11.1

Category Management

Monczka et al. (2015) write that category management is defined as the process of identifying stakeholder requirements and needs and comparing these to external supplier capabilities. The categories from the business point of view are products or services the business wishes to source, they are bundled in categories based on similar requirements or needs. Category management thus aligns internal requirements such as wanted supplier capacity and operational risk with sup-plier market conditions forming a category strategy. The category strategy contains a plan for negotiation contracts, how to evaluate and monitor the suppliers. Monczka et al. (2015) continues to explain that a strategy for category management seeks to reduce the risk for the purchasing business but also improve category performance in the dimensions important to the stakehold-ers. That is, as stated before, the categories are formed on the basis of a need. By identifying those requirements or needs and form them into a category which should be meet by external suppliers, the strategy has the potential of improving the purchasing along the dimension of the identified need. The category strategy shall strengthen the value proposition of the business as the category strategy will be in line with the overall business strategy. The forming of categories most likely requires top management’s involvement but once the categories are set, the category management, i.e. selection and evaluation of suppliers should be carried out by the purchasing department.

2.11.2

Vertical Integration

According to Buvik and John (2000), vertical integration is beneficial when acting in an envi-ronment of fast-changing or uncertain demand. Chen et al. (2004) stress three important fac-tors:

• Fostering close working relationships with a smaller amount of suppliers. • Promoting open communication between supply chain-partners.

• Developing long-term strategic relationships with suppliers in order to obtain mutual gains. The author claims that these factors together make a base for customer responsiveness, which is crucial in the markets of today with its rapidly changing customer demands. They state that open, informal sales channels are crucial for the development of tacit knowledge, which is vital from a strategic standpoint since it can help to better understand complex competitive subjects. Critical alliances with partners can also facilitate reaching strategic goals. In these alliances, accurate and relevant information is exchanged in a timely manner.

2.11.3

Limiting the Amount of Suppliers

Chen et al. (2004) also state that strategic purchasing is vital when establishing close relationships with a restricted amount of suppliers, which in turn has been shown to contribute to significant revenue gains and cost savings. Additionally, they discuss the greater risk that comes with limiting the number of suppliers, due to decreased flexibility and even supplier opportunism because of the high investments in those relationships. However, the authors highlight that working closely with suppliers contributes to greater trust among partners, dependability, and cooperation. If the suppliers are aware of the dependability, these factors might lead to suppliers not acting in an opportunistic way, and exploiting the relationship for example by increasing prices without no underlying reason. Further, Chen et al. (2004) claim that closer relationships with suppliers often

(35)

mean that they are long-term rather than “transactional based”, and highlight its contribution to greater cooperation, reduction of functional conflict and enhance decision making.

Cousins (1999) elaborates the reasoning of risks and opportunities that come with narrowing down the supplier base. Strategic decisions must be taken with caution so that the end result is not just giving suppliers more power, while in reality, the number of suppliers has not decreased at all. He underlines that a reduction has to take place at the same time as relationships are being further developed, in order to expand advantages to not only occur in a short-term perspective. However, Cousins (1999) states that the advantages may include facilitating management, spreading risks, and sharing resources. Thus, fewer and improved relationships may lead to using resources more efficiently, which leads to cost reductions.

An evaluation of these two factors can help determine what type of strategy that can be beneficial in order to use its purchasing power and suppliers, and to reduce risk. In addition, this assessment can help assess new sourcing opportunities in terms of vulnerability, threats strengths or new possibilities.

(36)
(37)

Chapter 3

Methodology

This chapter introduces the methodology of this thesis. In Section 3.1, the research process of this thesis is described. Next, in Section 3.2, the scientific position of this thesis is explained. Later on, the methodology for data collection and methodology for analysis will be introduced in Sections 3.3 to 3.4. The method for analysis is especially important, as it also specifies what outcome that can be expected from the analysis, given the scope and purpose of this thesis. Ultimately, the methodology chapter discusses the methodology and the sources from a scientific point of view, in sections 3.5 to 3.6.

3.1

Research Process

The work process of this thesis and wherein the process the different research questions are an-swered is illustrated by Figure 3.1 on the next page. The different boxes will be explained in the paragraphs below.

Identify Company Issues Initially, the case company Auctus, announced that it was in need of improving its cost control, but it had not yet determined exactly what should be examined to solve its issues. When the task of investigating this widely formulated problem was assigned to the authors, the first step in the research process was to conduct interviews with employees from the case company to specify their cost control problem. From the interviews, a clearer picture of the problem, as well as the underlying issues, was obtained. This represents the first box in Figure 3.1 called Identify Company Issue.

Theory Search 1 The theory search that was initiated during the identification of the company issue proceeds and forms a research strategy of using a multiple criteria ABC analysis to tackle the cost control issues connected to inventory control. This is called Theory Search 1 in the process figure.

Data Collection – Empirical Findings A compilation was made of the qualitative data from the interviews and financial information about the company. The empirical findings were structured in the same way as the analysis to facilitate the evaluation and comparison of data and theories. Based on the first theory collection and the data collection, it became clear how to design the multiple criteria ABC analysis to meet the purpose of this thesis. This is illustrated by the box called RQ 1.

References

Related documents

The increasing availability of data and attention to services has increased the understanding of the contribution of services to innovation and productivity in

Av tabellen framgår att det behövs utförlig information om de projekt som genomförs vid instituten. Då Tillväxtanalys ska föreslå en metod som kan visa hur institutens verksamhet

Generella styrmedel kan ha varit mindre verksamma än man har trott De generella styrmedlen, till skillnad från de specifika styrmedlen, har kommit att användas i större

Parallellmarknader innebär dock inte en drivkraft för en grön omställning Ökad andel direktförsäljning räddar många lokala producenter och kan tyckas utgöra en drivkraft

Närmare 90 procent av de statliga medlen (intäkter och utgifter) för näringslivets klimatomställning går till generella styrmedel, det vill säga styrmedel som påverkar

• Utbildningsnivåerna i Sveriges FA-regioner varierar kraftigt. I Stockholm har 46 procent av de sysselsatta eftergymnasial utbildning, medan samma andel i Dorotea endast

I dag uppgår denna del av befolkningen till knappt 4 200 personer och år 2030 beräknas det finnas drygt 4 800 personer i Gällivare kommun som är 65 år eller äldre i

Den förbättrade tillgängligheten berör framför allt boende i områden med en mycket hög eller hög tillgänglighet till tätorter, men även antalet personer med längre än