Transport Cartons' Impact on Supply Chain Efficiency

(1)

IN THE FIELD OF TECHNOLOGY DEGREE PROJECT

INDUSTRIAL ENGINEERING AND MANAGEMENT AND THE MAIN FIELD OF STUDY

INDUSTRIAL MANAGEMENT, SECOND CYCLE, 30 CREDITS

,

STOCKHOLM SWEDEN 2019

Transport Cartons' Impact

on Supply Chain Efficiency

GABRIELLA ANDERSSON

EVA JONSSON

(2)

(3)

Transport Cartons’ Impact on Supply Chain

Efficiency

by

Gabriella Andersson

Eva Jonsson

Master of Science Thesis TRITA-ITM-EX 2019:279

KTH Industrial Engineering and Management

Industrial Management

SE-100 44 STOCKHOLM

(4)

(5)

Transportkartongers påverkan på

försörjningskedjans effektivitet

av

Gabriella Andersson

Eva Jonsson

Examensarbete TRITA-ITM-EX 2019:279

KTH Industriell teknik och management

Industriell ekonomi och organisation

SE-100 44 STOCKHOLM

(6)

(7)

Master of Science Thesis TRITA-ITM-EX 2019:279

Transport Cartons’ Impact on Supply Chain

Efficiency

Gabriella Andersson

Eva Jonsson

Approved

2019-06-11

Examiner

Anna Jerbrant

Supervisor

Bo Karlson

Commissioner

Hennes & Mauritz

Contact person

Jens Fröhlich

Abstract

Large product assortments commonly lead to that fast fashion retailers have a large number of

transport cartons, with varying sizes, to distribute products from manufacturing sites to

distribution centers. Ultimately, the transport cartons have a large impact on the supply chain

efficiency, as they affect several nodes in the supply chain network. Previous research has

considered transport cartons’ impact on individual supply chain nodes, but little research has

been devoted to evaluate the transport cartons impact on the supply chain network as a whole.

Thus, this thesis aims to contribute with knowledge to bridge the gap in literature by

investigating how a company’s transport cartons affect the overall supply chain efficiency, by

applying a holistic perspective.

The thesis was conducted as a case study in collaboration with Hennes & Mauritz (H&M), a

world-leading fast fashion retailer. The research identifies opportunities and challenges related to

transport carton usage, and how the cartons impact supply chain efficiency. Moreover, the

research evaluates the costs associated with different sets of transport carton sizes, based on a

cost estimation model constructed on data provided by H&M. It is concluded that a small set of

transport carton sizes could increase supply chain efficiency and generate potential cost savings

due to mainly two reasons. First, a small set of transport cartons could lead to efficient container

loading. Second, it could lead to efficient carton handling and optimal space utilization in the

distribution centers.

Keywords: Supply chain efficiency, transport cartons, standard cartons, packing, packaging,

(8)

(9)

Examensarbete TRITA-ITM-EX 2019:279

Transportkartongers påverkan på

försörjningskedjans effektivitet

Gabriella Andersson

Eva Jonsson

Godkänt

2019-06-11

Examinator

Anna Jerbrant

Handledare

Bo Karlson

Uppdragsgivare

Hennes & Mauritz

Kontaktperson

Jens Fröhlich

Sammanfattning

Till följd av stora produktsortiment har snabbmodeaktörer ofta ett stort set av transportkartonger,

med varierande storlekar, för att distribuera produkterna från tillverkning till varuhus. Eftersom

transportkartongerna påverkar fler noder i försörjningskedjan, har de följaktligen stor påverkan

på hur effektiv försörjningskedjan är. Tidigare forskning har utvärderat hur transportkartongerna

kan påverka enskilda noder i försörjningskedjan, men lite fokus har ägnats åt att utvärdera

kartongernas effekt på försörjningskedjan, sett ur ett helhetsperspektiv. Detta examensarbete

syftar till att bidra med kunskap för att överbrygga detta gap genom att undersöka hur ett

företags transportkartonger påverkar den totala försörjningskedjans effektivitet genom att

tillämpa ett helhetsperspektiv.

Examensarbetet genomfördes som en fallstudie i samarbete med Hennes & Mauritz (H&M), som

är en av de världsledande aktörerna inom snabbmode. Studien identifierar möjligheter och

utmaningar relaterade till transportkartonganvändning, och hur kartongerna påverkar

försörjningskedjans effektivitet. Vidare utvärderades kostnaden av att ha olika uppsättningar av

transportkartonger, baserat på en beräkningsmodell grundad på data från H&M. Resultaten

visade att ett litet set av transportkartonger kan ha positiv påverkan på försörjningskedjans

effektivitet och generera kostnadsbesparingar till följd av två huvudsakliga anledningar. Den

första anledningen är att ett litet set förenklar effektiv containerladdning. Den andra anledningen

är att det kan leda till mer effektiv kartonghantering på lager och bättre utnyttjande av

lagerutrymme.

Nyckelord: Flödeskedje-effektivitet, transportkartong, standardkartong, packning, paketering,

(10)

(11)

Acknowledgements

This master thesis represents the last part of our five-year MSc degree in Industrial Engineering and Management at KTH, the Royal Institute of Technology. The thesis corresponds to 30 ECTS credits and has been conducted at the division of Industrial Management at KTH in cooperation with H&M. We would like to express our sincere gratitude to all persons who have supported us during this spring while writing our master thesis.

First, we would like to dedicate a special thanks to our supervisors at H&M, Jens Fr¨ohlich and

Anders Urhed, for giving us the opportunity to write this thesis and for their support throughout

the entire process. We would also like to thank Ingrid Källström, Johan Svärd, Victor Pettersson,

Andreas Enstr¨om, Maria ˚Aslin, Rene Klostermann, Karsten Gl¨ucker, Claes Holm, Ajit Krishnan

Kayrat and Je↵ery Hou at H&M for sharing your knowledge and experience with us.

Further, we would like to raise our sincere gratitude to our supvervisor at KTH, Bo Karlson, for giving us guidance, encouragement and valuable feedback. Lastly, thanks to our seminar leader, Maria Hammar´en, and our seminar group at KTH for contributing with thoughtful aspects that allowed us to further improve our thesis.

(14)

Abbreviations

3PL Third-Party Logistics ABC Activity Based Costing CFS Central Freight Station CMZD Customized

DC Distribution Center

MHE Material Handling Equipment SC Standard Carton

SCM Supply Chain Management SKU Stock Keeping Unit

(15)

List of figures

1 Illustration of the research process. . . 12

2 The supply chain process, revised figure, Richards (2011). . . 21

3 Warehouse activities as a percentage of total cost, revised figure, Richards (2011). . 25

4 Commonly used flowchart symbols, revised figure (Oskarsson, 2013). . . 30

5 Main principles of ABC, revised figure (Gerdin, 1995). . . 32

6 The supply chain structure at H&M. . . 35

7 The transport carton flow from manufacturing site to DC. . . 36

8 The transport carton flow in a Store DC . . . 38

9 The transport carton flow in an Online DC. . . 39

10 A generic transport carton. . . 41

11 String transport carton. . . 43

12 Unfolded shirt. . . 44

13 Illustration of half-folding and shirt-folding, including how each folding method is packed into cartons. . . 44

14 Content of packed transport cartons. . . 45

15 Key stakeholder units. . . 46

16 Summary of first round interviews. . . 52

17 Total carton distribution for the store and online channel combined. . . 53

18 Carton distribution by channel. . . 54

19 Carton distribution jackets and blazers. . . 55

20 Carton distribution jersey and knitwear. . . 55

21 Customized cartons distributed by product division. . . 56

22 Ladieswear packed in customized cartons. . . 57

23 Total carton flow cost (USD) As-Is compared with two cases. . . 59

24 Main principles of ABC applied to the transport carton flow. . . 61

25 Input-sheet in model for Company X. . . 64

26 Costs-sheet in model for Company X. . . 65

27 Assumptions-sheet in model for Company X. . . 66

28 Results-sheet in model for Company X. . . 66

(16)

List of tables

1 Interviews and observations conducted during the case study. . . 16

2 Existing unique standard cartons, where the length range between 1-10 and the width ranges between 1-5, where 10 and 5 are highest. . . 42

3 Key stakeholder responsibility and how the stakeholders are a↵ected by the transport carton flow. . . 46

4 Percentage impact of total carton flow cost per business unit. . . 60

5 Description of resources undertaken in the model. . . 62

6 Description of activitites undertaken in the model. . . 63

(17)

1. INTRODUCTION

1 Introduction

This section presents the setting of the thesis. First an overview of the background to the thesis subject is presented, followed by the purpose and the research questions. Thereafter the delimita-tions of the thesis, and the thesis expected contribution, to academia and to the case company, are described. The section ends with a description of the thesis disposition.

1.1 Background

The fashion industry is a global multi-billion-dollar industry (FashionUnited, 2019) characterized by strong historical growth and ever changing market conditions (Statista, 2019). Over the last 20 years, the fashion apparel industry has undergone profound transformation (Bhardwaj and Fairhurst, 2010). Until the mid 1980s, the fashion industry was characterized by low cost mass pro-duction of standardized styles. In the beginning of the 1990s, retailers shifted towards expanded product ranges and became more responsive to upcoming fashion trends. Thus, the fashion indus-try has gone from only having spring/summer and autumn/winter collections, to release several collections of products during a year.

Today, customer demands have forced retailers to prioritize low cost garments, flexibility in design, quality, and speed to market to stay competitive (Bhardwaj and Fairhurst, 2010). Big fashion retailers such as H&M, Gap, Benetton and Inditex group do not commonly manufacture their garments, instead the manufacturing is outsourced and thus their supply chains consist of large networks of external suppliers (Escalona Orcao and P´erez, 2014). A global supply chain leads to cost advantages, whereas a local supply chain facilitates fast response to market, which is a dilemma retailers are facing today when deciding on a sourcing strategy (Guercini and Runfola, 2004). As a consequence of the increased globalization, the complexity of retailers’ supply chains as well as the complexity of logistical activities have increased. Thus, geographical integration within the supply chain is crucial to stay successful as a retailer and maintain an efficient supply chain.

Packaging has a significant impact on logistics and thus supply chain efficiency (Hellstr¨om (2007),

Saghir (2004a), Johnsson (1998), Twede (1992)). However, how products should be distributed from producer to consumer is a subject that have been devoted little attention in previous research, whereas emphasis rather have been on whether the product can be produced and sold (Gustafsson et al., 2005). To stay successful, retailers must give attention to how products are distributed and what packaging solutions that should be used. Products are often enclosed and distributed in packaging with the objectives to contain, protect, handle, deliver or present goods (Paine, 2002). Improved packaging logistics would reduce handling costs and provide more secure loads, which in turn would lead to less damaged products (Gustafsson et al., 2005). Thus, by considering packaging logistics, business and supply chain performance could be enhanced and lead to higher profit margins, lower costs and reduced environmental impact.

(18)

1. INTRODUCTION

1.2 Purpose & research questions

The aim of this research is to understand necessary considerations of how a company’s transport cartons a↵ect supply chain efficiency and supply chain costs, from a holistic perspective. In order to operationalize the aim of this research, the following research question and sub research questions have been formulated;

RQ: How does a set of transport cartons impact supply chain efficiency and supply chain costs? SRQ1: What stakeholder units within the supply chain are a↵ected by the transport cartons? SRQ2: What aspects are important to consider when deciding on a set of transport cartons? SRQ3: What is the transport carton distribution and how are the transport cartons used at H&M today?

SRQ4: How can the total cost of the transport carton flow vary depending on the set of transport cartons?

1.3 Delimitations

As the phenomenon investigated is highly complex, delimitations were made. The research was delimited to examine the transport carton flow from manufacturing sites to distribution centers (DCs), and not further shipment to end customers or stores.

As H&M is one of the biggest retailers in the world, a vast amount of data exists, and therefore further delimitations to only analyze data from the German and Central European market were made. These markets were thought to be representative for this study mainly due to two reasons. Firstly, they are the largest planning markets in Europe and have several distribution centers, tak-ing the complexity with mixed cartons and sales accuracy into consideration. Furthermore, the distribution centers in these markets are partly automated, one of the important aspects when analyzing solutions suitable for the future. To successfully be able to process the vast volume of data provided by H&M, it was delimited to only include data for the year of 2018. This delimi-tation of only analyzing one year of data allowed for analysis across all seasons, which a↵ects the collections that H&M releases and in turn what transport cartons that are used to distribute the products.

1.4 Expected contribution

This thesis aims to empirically contribute to the research area of packaging logistics with a unique case study, of one of the world’s largest fast fashion retailers, aiming to identify important aspects to consider related to transport cartons to enable for an efficient supply chain. More specifically, the thesis aims to illustrate potential opportunities and challenges related to transport carton usage and how it a↵ects supply chain efficiency.

(19)

1. INTRODUCTION

Furthermore, the thesis aims to contribute to H&M by investigating what impact the transport cartons have on its supply chain efficiency and by conducting a business case, quantifying potential cost savings. To be able to conduct the business case, a cost estimation model will be constructed, which also will be delivered to H&M.

1.5 Disposition

The outline of the thesis is as follows:

Introduction: This section presents the setting of the thesis. First an overview of the background to the thesis subject is presented, followed by the purpose and the research questions. Thereafter the delimitations of the thesis, and the thesis expected contribution, to academia and to the case company, are described. The section ends with a description of the thesis disposition.

Method: This section presents the methods used while conducting the thesis. It starts with a description of the research process which was carried out in three stages, namely a pre-study, a main study and an analysis of the results. The methods used in each stage of the research process are then thoroughly presented. The chapter ends with a discussion of the quality of the research, by discussing the thesis validity, reliability, generalizability and research ethics.

Literature and theory: This section presents previous literature and theories related to the subject under study in order to provide an understanding of the context of the research subject. First, an introduction to supply chain and supply chain management theory is presented, followed by a literature review of existing research related to warehousing, transport container loading and packaging. Last, the fundamentals of process mapping and activity-based costing are introduced, which are used as theoretical frameworks in this research.

Case company: This section gives an introduction to the case company and its operations related to the transport cartons in order to provide an understanding of the organization and subject under study. The section is based on findings from the pre-study and observations, describing the current transport carton situation. Moreover, Section 4.5 serves to answer the first sub research question.

Results & analysis: In this section, the findings from the case study are presented. First, impor-tant aspects that need to be considered related to transport cartons are identified. Second, results from data analysis of the current transport carton situation at H&M are presented. Last, results from the cost estimation model given di↵erent scenarios are presented. Analysis and explanations of the findings are presented along with the results.

Description of cost estimation model: This section presents a description of the general cost estimation model, which was constructed based on findings from the case study to calculate the total transport carton flow costs for the scenarios presented in Section 5.3. Further, the model’s main features and calculations are described. The section ends by presenting delimitations and limitations of the model.

(20)

1. INTRODUCTION

Discussion: This section contains a discussion about the findings in this study. First, findings related to the transport cartons are discussed. Second, criticism against the construction of the model and how it might have a↵ected the results are presented. Third, opportunities for further business improvements, detected during the case study, are discussed as well as the generalizability of the findings and sustainability aspects related to the research subject.

Conclusions: This section presents conclusive remarks and the answer to the research question evaluated in this thesis. Moreover, opportunities for further research are discussed.

(21)

2. METHOD

2 Method

This section presents the methods used while conducting the thesis. It starts with a description of the research process which was carried out in three stages, namely a pre-study, a main study and an analysis of the results. The methods used in each stage of the research process are then thoroughly presented. The chapter ends with a discussion of the quality of the research, by discussing the thesis validity, reliability, generalizability and research ethics.

2.1 Research process

The research topic of this thesis originated from a subject in need of investigation identified by H&M. The methodology used followed an inductive approach, designed as a case study. Case studies are suitable when the research purpose is to investigate and analyze a phenomenon in its natural context using various methods to understand the phenomenon (Collis and Hussey, 2013). In addition, case studies are useful when looking at holistic characteristics of phenomena which makes it suitable for this thesis that aims to investigate a company’s set of transport cartons from a holistic perspective (Yin, 2009). The research was initiated with a pre-study, which aimed to acquaint the authors with the organization and create a deeper understanding of the empirical context. Based on findings from the pre-study, the transport carton flow was mapped and the key stakeholder units were identified.

The pre-study was followed by a main study, which can be divided into three parts. First, semi-structured interviews with the identified key stakeholders at H&M were held. Second, quantitative data analyzes were conducted, which aimed to examine the current distribution of the transport cartons at H&M. Last, a business case was conducted, with the aim of comparing the costs for the current set of transport cartons at H&M with other potential sets of transport cartons. Throughout the entire case study, a literature review was conducted.

(22)

2. METHOD

2.1.1 Literature review

A literature review was initiated, starting with wide information gathering related to the area of research which then was narrowed down to specific topics as the scope of the master thesis clarified. According to Blomkvist and Hallin (2015) this is rather a typical approach when conducting a literature review due to complexity of the area of research, which makes it crucial to obtain a thorough understanding of the research area as well as understand how to position the study relative to previous research within the field. The literature review was carried out through the entire research process and was based on carefully selected sources.

Practically, the literature review was conducted by thoroughly searching for scientific literature sources using the databases KTH Primo, Web of Science and Google Scholar. Information about theories and concepts were also retrieved from dictionaries and books covering specific themes. To ensure qualified sources, the searches were filtered to only include journals and publications. Thus, the literature review was based on relevant sources from recent years and older publications where the content were considered applicable despite the publication year. Moreover, the results from previous research which were closely related to the area of research, lead to that the subject of investigation was clarified and that the investigated research questions could be formulated. Keywords used for finding relevant literature were; Supply Chain, Supply Chain Management, supply chain costs, supply chain networks, retail supply chain, omni-channel supply chain, supply chain constraints, warehouse management, warehouse challenges, transport cartons, transport op-timization, container loading, packaging, packing problems, packaging logistics, process mapping, and Activity-Based Costing.

2.1.2 Case study

According to Collis and Hussey (2013) a case study usually follows five primary steps; 1) selecting the case, 2) preliminary investigations, 3) data collection, 4) data analysis and 5) writing the report. The actions taken in step one to four are further described below.

Selecting the case

As the thesis topic originated from a subject in need of investigation, identified by the case company, there was no case selection process. The complexity with transport cartons is however a common issue for a large number of retailers (BCG, 2013), highlighting the need for investigation of the subject. Several supply chain nodes are impacted by the transport cartons, picturing the complexity of a carton strategy well.

As H&M is one of the largest fashion retailers in the world (FashionUnited, 2019), with a heavy-logistic supply chain, it is a suitable case company for studying a transport carton flow. Further, a large fashion retailer is thought to be a more relevant choice for the purpose of this thesis opposed to a small fashion retailer with less complex supply chain. Moreover, the current transport cartons at H&M has not been designed to consider all aspects from the di↵erent nodes in the supply chain.

(23)

2. METHOD

Neither for its current multi-channel supply chain, nor for tomorrow’s potential omni supply chain. H&M is therefore, for our aim, a highly suitable case of a fashion retailer with a transport carton strategy lacking a holistic perspective.

Preliminary investigation

The preliminary investigation is a process of becoming familiar with the context in which the case study will be performed. Collis and Hussey (2013), commonly refer to the preliminary investi-gations as the pre-study. Therefore, before setting the scope and formulating research questions to be investigated during this thesis, a pre-study was made in order to gain an understanding of the overall industry in which the case company operates. This preliminary stage was also the stage where both academic and the case company’s expectations were defined. As the thesis study originated from the case company supervisors’ hypothesis, being objective during this stage of the research process was of extra importance as the perceived ’issue’ could be a subjective imagination. Therefore, before setting the scope of the thesis, several interviews were held with relevant internal key stakeholders and observations at distribution centers were made. Objectivity was achieved by interviewing employees from di↵erent units in the supply chain to obtain as many perspectives as possible of the investigated problem. Parallel to the interviews and observations at the DCs, a liter-ature study was conducted to get a general point of view based on previous research within the field.

Data collection

Interviews

Interviews were held to gain a deeper understanding of how the case company and its supply chain processes work. According to Blomkvist and Hallin (2015), interviews are suitable in an initial stage of the research to develop in-depth understanding, find new dimensions of the studied area as well as find ambiguity. Thus, interviews with stakeholders from di↵erent departments were con-ducted to obtain a holistic point of view and input from di↵erent perspectives. More specifically, interviews were held with stakeholders from seven departments; Buying Office, Production Office, International Freight and Transport, Warehouse Logistics, Garment Care, Sales and Sustainability. These stakeholders were selected after discussions with the supervisors at H&M. They were chosen as they all have interests related to the transport cartons, which are needed to gain an overall un-derstanding of the transport carton flow, from order placement to end customer delivery. Moreover, an interview with Warehouse Logistics Business Intelligence department was conducted to discuss secondary data related to the transport cartons.

First, unstructured pre-study interviews were held to develop in-depth understanding about the area of research and clarify the investigated research problem. Second, semi-structured case study interviews were conducted, which also allowed for follow-up questions. The case study interviews could further be divided into three interview rounds. In the first interview round, all seven depart-ments were asked about challenges related to transport cartons, and about strict requiredepart-ments and preferences on a potential new set of transport cartons. In the second interview round, the two

(24)

2. METHOD

divisions Warehouse Logistics and Garment Care were asked follow-up questions on empirical find-ings from the quantitative data analysis. The third interview round consisted of questions related to cost drivers for the transport cartons and di↵erent carton sizes. All departments except Buying Office and the Sustainability department were included in the third interview round. The reason to why these two departments were excluded, was because they do not have any costs directly attributable to the transport cartons.

All interviews were documented in writing and the duration varied between 30-90 minutes. Most of the interviews were face-to-face, but five of the interviews during the first and third interview rounds, were conducted using Skype which was due to geographical circumstances. Further, follow-up questions were asked through e-mail. All predefined questions during the first interview round, where the interviewees were asked similar questions related to the transport cartons, can be found in Appendix A.

Observations

To understand the distribution center processes, two distribution center visits to Eskilstuna and

Bor˚as were made to observe the daily distribution center routines. Observations are of exploratory

character and appropriate to investigate daily work routines (Blomkvist and Hallin, 2015). Since observation as a methodology can be perceived subjective, the observations were systematically documented, which Blomkvist and Hallin (2015) argues is important to stay credible as an obser-vant. Throughout the DC visits, questions were asked to clarify DC processes.

Data from secondary sources

In the process of compiling a status analysis, secondary data collected from H&M’s databases were analyzed. The secondary data was extracted with help from the business unit Business Intelligence. As H&M’s databases contain a significant volume of data, it had to be reduced to a subset of the H&M assortment to create prerequisites for analysis. As such, the subset was restricted to only include data containing information related to transport carton shipments to the German store market and the Central Europe online market, during 2018. These two markets were selected be-cause they represent two large destination markets for store and online respectively, resulting in a large volume of data. By analyzing a large volume of data, the result become more trustworthy as it includes the complexity that large destination markets entail, such as for example several distribution centers. Data from both the store and online destination markets were analyzed as the automation level in the warehouses di↵ers. Further, these two markets were selected as they are automated to a great extent and thus are representative for future distribution centers.

(25)

2. METHOD

Summary of data collection

Table 1 below summarizes all interviews and observations that were conducted during this the-sis.

Table 1: Interviews and observations conducted during the case study.

Stage Date Department Interview type Interview round

Pre-study 2018-12-07 Warehouse Logistics Unstructured Preliminary

Pre-study 2019-01-23 Garment Care Unstructured Preliminary

Pre-study 2019-01-25 Warehouse Logistics Unstructured Preliminary

Business Intelligence

Pre-study 2019-02-01 Warehouse Visit Observation Preliminary

Eskilstuna

Pre-study 2019-02-05 Warehouse Visit Observation Preliminary

Bor˚as

Case study 2019-02-01 Garment Care Unstructured First round

Case study 2019-02-06 Warehouse Controlling Semi-structured First round

Case study 2019-02-07 International Freight Semi-structured First round

& Transport

Case study 2019-02-08 Sales Semi-structured First round

Case study 2019-02-11 Sustainability Semi-structured First round

Case study 2019-02-18 Buying Office Semi-structured First round

Case study 2019-02-20 Production Office Semi-structured First round

Case study 2019-02-21 Warehouse Logistics Semi-structured First round

Automation

Case study 2019-03-11 Warehouse Logistics Semi-structured Second round

Case study 2019-03-12 Garment Care Semi-structured Second round

Case study 2019-03-01 Production Office E-mail Third round

Case study 2019-03-01 Garment Care Semi-structured Third round

Case study 2019-03-12 Warehouse Controlling Semi-structured Third round

Case study 2019-03-13 Sales E-mail Third round

Case study 2019-03-18 Warehouse Controlling Semi-structured Third round

Case study 2019-03-22 International Freight Semi-structured Third round

& Transport

(26)

2. METHOD

Data analysis

Collis and Hussey (2013) highlights the importance of becoming familiar with the data when con-ducting a with-in case study to build up separate descriptions of events, opinions and phenomena that can be used to identify patterns.

In this thesis study, data triangulation was achieved through collection of data from di↵erent sources as interviews, observations and secondary data. Through the use of multiple data sources, bias was minimized and validity established (Yin, 2009). Several methods for structuring data analyzes exist (Blomkvist and Hallin, 2015). In this research, a thematic approach was used, which is commonly used when analyzing qualitative empiric. Thus, the data were sorted in di↵erent categories (requirements, process, costs) with the objective to later answer the research questions based on the di↵erent categories. The findings from interviews and observations were compiled and presented based on the mentioned themes and to describe the empirical setting as well as the transport carton flow.

The analysis of the secondary data was mainly conducted using Microsoft Excel, but as the compu-tational power of Excel is limited, the software Tableau was also used as a complement to overview and derive relevant data for the analyses. More specifically, four quantitative data analyzes were done analyzing the transport carton distribution, carton distribution per product group, distribu-tion of customized transport cartons and the product quantity packed in the main standard cartons.

Transport carton distribution

The actual usage of di↵erent carton sizes was analyzed, for both channels combined and separately. All standard carton sizes were included, but grouped together depending on its size, resulting in a total of 17 carton groups. Each unique standard carton size was grouped together with the standard cartons with height variations of these. Further, the customized and string cartons were grouped together to create a more plausible view of what fraction of the total number of cartons they constituted. Data from all seven departments, Cosmetic, Divided, Ladieswear, Menswear, Home department, Other and Sizes 50-170, and all product categories were analyzed.

Carton distribution per product group

The carton usage per product category was analyzed by grouping products with similar character-istics together. In this analysis, only data from the ladieswear, menswear and divided departments were included, which accounts for the largest amount of H&M’s revenue. A total of 15 product groups were analyzed; Jerseys, Jackets & Blazers, Blouses, Denim & Trousers, Shoes, Swimwear & Lingerie, Dresses, Knitwear, Sportswear, Socks & Tights, Shorts & Skirts, Suits, Woven tops, Accessories and Bags. Product categories with unclear descriptions or occasional deliveries such as for example New development, Mixed collection, Press collection, Promotion, Giveaways and Other items were excluded from this analysis.

(27)

2. METHOD

Customized transport cartons

Data of carton usage of customized cartons was analyzed for all departments, including all product categories. While conducting this analysis, the product categories were grouped together as in the analysis described above. Except for analyzing the usage of the customized cartons, the customized carton measures were compared with the standard transport carton measures.

Main standard carton and product quantity

The product unit quantity packed into the two main standard cartons was analyzed. In this analysis, the data was filtered to only include data from the departments, ladieswear, menswear and divided. Further, only full cartons were included, to avoid cartons packed with irrepresentative number of products. Both the average unit quantity and the mean value were analyzed to ensure that the data was representative and that the measures did not deviate too much.

2.1.3 Model creation

A cost estimation model was created to evaluate the business case of changing the set of transport cartons. A business case is an investigation of whether a future potential project will be profitable, and is conducted before a project is approved (Lester, 2017). The business case can be referred to as a document that establishes what the problem is, what the solution will be and what it will cost in terms of resources and time. The underlying purpose of the document is to inform a final decision maker that will decide whether to take on a project or not (Blais, 2012).

The model was constructed to evaluate costs related to the existing set of transport cartons, com-pared with other sets of transport cartons. The model was built using Excel and was based on data given from the stakeholder units in the third round interviews. Thus, the model was based on costs related to the transport cartons and given assumptions. In absence of reliable data re-garding product quantity packed into the di↵erent carton sizes, the model was built to take two potential scenarios regarding product quantity into account. Furthermore, the model was evaluated along with stakeholders to ensure that the main cost drivers were taken into account. A thorough description of the model can be found in Section 6.

2.2 Quality of the research

When conducting a case study, it is fundamental to consider validity and reliability to ensure good research quality (Blomkvist and Hallin, 2015). These aspects, in addition to genaralizability and research ethics, have been discussed as follows to strengthen the credibility of this research.

2.2.1 Validity & reliability

To ensure the quality of the case study the dimensions validity and reliability were considered. Validity is of importance to ensure that the research is giving a fair representation of reality.

(28)

Con-2. METHOD

clusions should be supported by a chain of evidence, preferably triangulated using multiple data sources (Gibbert, Ruigrok, and Wicki, 2008). This thesis is built upon both quantitative and quali-tative data sources which, thus, strengthen the thesis validity. The quantiquali-tative data was extracted from H&M’s databases. The qualitative data was gathered from unstructured and semi-structured interviews and by observations during distribution center visits. Shah and Corley (2006) emphasize the importance of unstructured interviews when trying to understand all underlying perspectives which increase the construct validity of the study, depicting a more accurate representation of re-ality. Unstructured interviews were therefore held during the first stage of the case study in order to allow each stakeholder to emphasize what was most important from their point of view. The validity of the thesis is further strengthened by the fact that it covers multiple perspectives as interviews were held with stakeholders from several di↵erent business units.

The reliability dimension refers to whether a study has been conducted in the right way (Blomkvist and Hallin, 2015). A study’s reliability is furthermore increased by ensuring that the results of the study would be found again if the study was repeated. Therefore, to facilitate the replication of the case study, a thorough description of the methodology used when conducting the case study was presented. In addition, the cost estimation model used to evaluate the business case and its features are described in detail. Moreover, semi-structured interviews were held, aiding in strengthening the case study’s reliability.

The conclusion drawn from qualitative data can, however, be hard to repeat due to individual interviewee and interviewer interpretations which may impact the results (Collis and Hussey, 2013). The risk of subjective interpretation during the interviews, is however somewhat reduced as two researchers held the interviews, which possibly increases the accuracy of the thesis findings (Kaulio and Karlsson, 1998). As the case-company name is provided the reliability is further increased (Gibbert, Ruigrok, and Wicki, 2008). There are however several improvement areas related to the case study’s reliability. Firstly, recording and transcribing the interviews had facilitated replication of the case study and limited the risk of subjective interpretations. Secondly, the quantitative data that the case study was based on is company specific, and thus the reliability would have been favoured if the case study was based on data from several case companies. Lastly, some of the analyzes performed in this case study could be difficult to replicate as information, such as carton measurements and costs, was confidential and thus could not be included in the report.

2.2.2 Generalizability

The case study methodology is criticized for being unscientific in the way that it is subjective and primitive, which makes it difficult to draw conclusions to a phenomena, or explanandum (Blomkvist and Hallin, 2015). Since this thesis is based on a single case study the findings do not allow for statistical generalization. Instead analytical generalization is used, where generalization from empirical observations to the theory are taken into consideration (Yin, 2009). The analytical generalizability is achieved through a discussion of how the results’ in this case study can be applicable to other, similar cases.

(29)

2. METHOD

The findings in this master thesis study can be generalized to logistic heavy retail companies, where products are delivered in carton boxes and the design process not is restricted to fit specific carton measures. Moreover, research findings from di↵erent nodes in the supply chain network can be applicable to other industries with heavy logistics. A detailed discussion about how the findings in this case study are analytically generalizable will be presented in Section 7.4.

2.2.3 Research ethics

Research ethics is of great importance, especially when managing expectations from dual stake-holders (Blomkvist and Hallin, 2015). In this case, cooperating with a large retail company, caution was taken not to let the case company gain too much influence on the thesis project. Moreover, the research fulfilled the four general rules, secrecy, professional secrecy, anonymity and confidentiality for good research practice formulated by the Swedish Research Council (2017). All participants contributing to the master thesis study was informed about its purpose and their rights about consent. In the outset of the master thesis process, nondisclosure agreements were signed to ensure that confidential information was only to be used with the purpose of writing this report and not going to be jeopardized, neither during nor after the master thesis process. Since the data provided by H&M was highly confidential, consisting of detailed information about carton box shipments, it was manipulated in this report. However, this does not a↵ect the final results and conclusions in the report. Further, research misconduct defined by All European Academies (2017) as fabrication, falsification or plagiarism have been avoided by aligning results and findings with stakeholders before presenting them, justifying exclusions and manipulating data throughout the report and always referring to other researchers’ work when necessary.

(30)

3. LITERATURE AND THEORY

3 Literature and theory

This section presents previous literature and theories related to the subject under study in order to provide an understanding of the context of the research subject. First, an introduction to supply chain and supply chain management theory is presented, followed by a literature review of existing research related to warehousing, transport container loading and packaging. Last, the fundamentals of process mapping and activity-based costing are introduced, which are used as theoretical frame-works in this research.

3.1 Supply chain theory

3.1.1 Supply chain and supply chain management

Today, it is rarely the case that a single-firm is responsible alone for all the processes related to product creation from raw material to distribution to end customer. Instead, products reach the end customers through supply chains. Supply chain is a multi-defined concept that has been revised several times throughout history. Christopher (2011) defines supply chain as ’the network of orga-nizations that are involved, through upstream and downstream linkages, in the various processes and activities that produce value in the form of products and services delivered to the ultimate consumer’. The supply chain often constitutes of independent firms, handling various units in the supply chain, collaborating to maximize the overall supply chain performance as well as their own objectives. The common denominator of the di↵erent definitions is the consistency of di↵erent enti-ties and their interplay to deliver products or services to end customers. Below, Figure 2 illustrates the supply chain process.

Figure 2: The supply chain process, revised figure, Richards (2011).

To optimize overall business performance, Lummus and Vokurka (1999) argue that it is crucial for companies to manage the supply chain network since it can enhance a firm’s competitiveness through reduction of lead times across the supply chain. The term used for this business activity is Supply Chain Management, SCM. The Council of Supply Chain Management Professionals (2019)

(31)

defines supply chain management as: ’Supply chain management encompasses the planning and management of all activities involved in sourcing and procurement, conversion, and all logistics management activities’. Thus, supply chain management includes coordination and collaboration within and across companies constituting the supply chain network, which among other entities includes suppliers, intermediaries, third party service providers, and customers. Practitioners agree that supply chain management is crucial for a firm’s competitiveness, and benefits such as reduction in supply chain costs and increased service level have been highlighted in several empirical studies (Pibernik et al., 2011).

Naturally, supply chain management has a big impact on the supply chain cost. Nilsson (2000) even argues that flow economy is the term that best describe supply chain management. Supply chain cost can be defined as ’all relevant costs in the supply chain of the company or organization in question’ and is often confused with logistics cost which is a narrower concept that refers to costs related to distribution, transportation or warehouse costs (Pettersson and Segerstedt, 2012). To ensure cost minimization and at the same time maximize customer value throughout the supply chain is challenging (Lee and Billington, 1993). Several studies discusses problems related to supply chain inefficiency resulting from isolated improvements in local or partial sections of the supply chain (Lee, 2000), further highlighting the need for supply chain management. Especially for large businesses, it has been common to breakdown the business into operational components such as purchasing, production and distribution and optimizing each operational component of the supply chain separately (Ibrahimov et al., 2012). Though, companies have in recent years, realized that optimization of isolated silos is not efficient, and instead e↵orts are concentrated to optimize the whole system and overall supply chain.

3.1.2 Fast fashion supply chain

Uncertainties in customer demand and fast changing buying pattern have challenged the industry and forced retailers to adopt to new supply chain strategies, such as fast fashion supply chains which are characterized by a larger number of product o↵erings simultaneously, in e↵ort to diminish the use of product discounting (Huang, Hsu, and Ho, 2014). In fast fashion supply chains, retailers strive to align supply quantities in response to observed product demand.

The two most common supply chain strategies to meet product demand are push and pull strategies (Yang, Cai, and Chen, 2018). A push supply chain is characterized by that retailers orders products from the supplier before the selling season, which comes with a certain inventory risk. In the pull supply chain strategy, retailers’ place at-one orders in the selling season, instead leaving suppliers to face a bigger risk. Historically, the push strategy has been most popular, but as the customers shopping pattern has changed, the pull strategy has shown to be the most successful when it comes to ordering the optimal order quantity.

A strategic question, when dealing with meeting the right product demand, is how the manufactur-ing should be carried out (S¸en, 2008). Trade-o↵s between in-house production or outsourcmanufactur-ing are made, which has di↵erent advantages and disadvantages. In-house production ensures control over

(32)

quality and time and minimizes communication problems within the supply chain network. On the other side, outsourcing the production to contractors comes with the advantage of being more cost efficient and more flexible (Brown and Rice, 1998). An additional issue is the strategic choice of where to locate production sites, which results in trade-o↵s between supply chain responsiveness and cost efficiency, leading to that many retailers select cost efficiency and o↵-shore production in Asia and Latin America (S¸en, 2008). To shorten lead times, Stratton and Warburton (2003) found out that many retailers used local manufacturers to produce fast-response products. In situations when local manufacturing is not possible, retailers order large product quantities to minimize the risk of not meeting the customers demand at a certain selling season.

3.1.3 Supply chain structures for retailers

There are three commonly used supply chain structures for fast fashion retailers, single channel, multi channel and omni-channel. The single channel approach is used by retailers operating in one channel, i.e. pure online retailers and BM retailers (retailers that have at least one physical location, i.e. stores residing in local shopping malls). The supply chains for retailers operating both online and in-store are referred to as multi channel and have historically structured their supply chains in segregated channels, with separated systems. This is a consequence of the fact that many actors started o↵ as single-channel retailers and later on expanded their businesses to include other channels. Since it usually has been no common interface between the di↵erent channels’ supply chains the retailer’s customers have either shopped and returned products in-store or online. Thus, not allowing customers to order online and pick up in store or vice versa. As retailers recently have started facing challenges due to new customer demands on seamless shopping experiences, many retailers are now in the process of merging its supply chains. Going from separated multi channel solutions to unified solutions, independent of channel, commonly referred to as omni-channel supply

chains (H¨ubner, Wollenburg, and Holzapfel, 2016). The transition towards an omni-channel supply

chain is, according to H¨ubner, Wollenburg, and Holzapfel (2016), challenging but seen from a

holistic perspective the transition generates considerable competitive advantages.

3.2 Warehousing

Warehouses have historically been referred to as cost centres (Richards, 2011), but is today seen as vital parts in the supply chain, enabling delivery of the right product, in the right quantity, to the right customer, at the right place, at the right time, in the right condition, at the right price. Several types of warehouses exist, operated by di↵erent suppliers, including raw material suppliers, manufacturers, retailers and companies handling reverse logistics. Warehouses are a way for suppliers to hedge against uncertain demand patterns and high seasonality. Besides, warehouses cover for production shutdowns and allow for bulk buying.

A type of warehouse typically used in retail supply chains is referred to as a Distribution Center (DC). A DC di↵ers from a warehouse as it, apart from storing products, provides value-added

(33)

ser-3. LITERATURE AND THEORY

vices (VAS) such as steaming, cross docking, packaging etc. Thus, DC operations is more complex than general warehouse operations, resulting in more advanced warehouse equipment. Moreover, a distribution center acts as the bridge between the supplier and its consumers, and typically both retail and warehouse orders are shipped from distribution centers instead of warehouses (Mani, 2019).

An individual item that passes through the item flow at a warehouse is referred to as a stock keeping unit, SKU (Rouwenhorst et al., 2000). Typically, the item flow through a warehouse can be divided into four distinct phases; the receiving process, the storage process, the order picking process and the shipping process, which are described as follows.

Receiving

SKU’s arrive by truck or internal transport. Items are checked or repacked to di↵erent storage modules before being put away to await transport to the next process.

Storing

The SKU’s are placed in storage locations. The storage area can be of two di↵erent types, reserve area or the forward area. Smaller quantities of products are often stored in the forward area, where items are easier to access for an order picker. The process when SKU’s are transferred from the reserve to the forward storage is referred to as replenishment.

Order picking

The process when items are retrieved from their storage locations and transported to storing and/or consolidation process, where consolidation is grouping of items destined for the same customer. The order picking can be done either manually or automated to a certain extent.

Shipping

Orders are checked, packed and loaded into carriers that will deliver the products to end cus-tomers.

Gu, Goetschalckx, and Mcginnis (2010) argue that the layout design of the warehouse is a key component of warehouse optimization. This is further strengthened by Caron, Marchet, and Perego (2000), that argue that the layout design impacts more than 60 percent of the total travel distance within the warehouse. However, warehouse design is complex and depends on trade-o↵s between di↵erent costs. One of the biggest challenges when it comes to warehouse design is the technical capabilities. Storage units and systems have to function for di↵erent types of products with di↵erent characteristics and sizes, thus leading to difficulties for warehouse optimization (Rouwenhorst et al., 2000).

To further highlight the role of the warehouse in the supply chain, warehouse costs make up be-tween one to five percent of total sales depending on the company and its type of product. In addition to this, warehousing constitutes about 22 percent of a company’s total logistics cost, with inventory carrying costs at a further 23 percent. The warehouse impact on costs highlights the importance of warehouse optimization and warehouse management. Below, Figure 3 illustrates warehouse activities as a percentage of the total cost (Richards, 2011).

(34)

Figure 3: Warehouse activities as a percentage of total cost, revised figure, Richards (2011).

As illustrated in Figure 3, order picking constitutes the costliest activity at the warehouse. Its labor intensive, challenging to automate and can be complicated to plan. The result is that companies face the challenging trade-o↵ between speed, accuracy and costs (Richards, 2011). To increase productivity in the picking process, preparation and understanding of the products sales pattern are essential. Where products are located a↵ects travel distance and strain on employees, which in turn increase productivity and reduces cost.

3.3 Transport container loading

Transport containers are designed to facilitate and speed up handling of freight and to protect the transported goods against weather and pilferage (Steenken, Voß, and Stahlbock, 2004). Several types of standard containers exist, among them are the short twenty feet containers, forty feet containers and forty-five feet containers. Some goods, requiring special handling conditions, are transported in containers with additional properties such as reefer containers or oversized contain-ers.

Constraints in container loading can be either hard or soft (Bortfeldt and W¨ascher, 2013). Hard

constraints must be met, whereas violations are tolerated for soft constraints to a certain limit. One hard constraint that cannot be overseen is the space constraints set by the dimensions of the containers. In previous research weight limits have also been addressed as hard constraints (Liu et al. (2011), Gehring and Bortfeldt (1997), Terno et al. (2000), Chan et al. (2006), Egeblad et al. (2010), Liu et al. (2011)). Another challenge when it comes to weight, is the weight distribution that needs to be spread evenly inside the container (Bortfeldt and Gehring, 2001). Otherwise, there is a risk that the cargo shifts during container movement, which will be reduced if the weight is balanced. Though, this constraint is considered to be a soft constraint. Other risks that may

(35)

arise followed by unbalanced weight are an uneven distribution of axle weight when the container is transported on a truck or that handling operations become complicated, and in worst case, impossible (Ratcli↵ and Bischo↵, 1998).

Moreover, the vertical orientation of a box is limited to one or two dimensions and should not be the smallest surface. The vertical orientation of the boxes also limits the load-bearing strength of a box. Thus, vertical orientation constraints minimize the risk that packaging get damaged and ensure stability of the load. An additional hard constraint is the stacking constraint, which restricts how many boxes that can be stacked on top of each other. The load-bearing strength of the boxes set this constraint and determines the pressure it can take, which is defined by the strength of the box’s side walls before it crashes (Bischo↵ and Ratcli↵, 1995). A box’s load-bearing strength depends on several factors, the orientation of the box when loading the container, the content of the box, and at last the surrounding conditions such as humidity, duration of the transport and the way of stacking. To avoid damage of loading, restrictions of how to stack boxes exist, prohibiting larger boxes from being put on smaller or fragile carton boxes, and limiting the number of boxes that can be stacked on each other (Junqueira, Morabito, and Yamashita, 2012). Researchers (Lin et al. (1993), Egeblad et al. (2010)) also argue that heavy items should be placed below lighter ones. Zhao et al. (2016) explains that how to load a container depend of whether the boxes are homogeneous or not. If the boxes placed in the container are rather homogeneous, loading trough wall building and layer building are the two most common strategies, whereas placing the boxes one at a time is most common when boxes are strongly heterogeneous.

There are also other soft allocation constraints such as connectivity constraints (Liu et al., 2011), which requires that the cargo is shipped together, or separation constraints, which demand that some items should not be shipped together in the same container (Eley, 2003). For example, according to Eley (2003), goods such as food and perfumery should not be shipped together. To facilitate for operations as loading and unloading the containers, pallets should be loaded with items of the same type (Terno et al., 2000). Positioning constraints have been assigned as both soft or hard constraints. The positioning of goods depends on the actual product, its size and its weight. Haessler and Brian Talbot (1990) brings up the complexity with loading close to the door, where heavy items should not be placed. Relative positioning constraints have further been discussed by several researchers (Bischo↵ and Ratcli↵ (1995), Haessler and Brian Talbot (1990)). They bring up that di↵erent goods can have a negative e↵ect on each other’s quality, and that some items require a certain distance to each other whereas other items do not. Multi-drop situations are another problem when optimizing container loading, requiring goods delivered to the same place to be strategically placed to avoid unnecessary unloading and reloading operations. Moreover, specific load-related constraints exist, which have been addressed as one of the most important issues according to several researchers (Bortfeldt and W¨ascher, 2013). Load-related constraints can be stability constraints or complexity constraints. If the load is not stable, the cartons can get damaged and get difficult to handle for employees during loading operations. Load stability can be achieved through two di↵erent ways of placing the carton boxes, vertically or horizontally, which has di↵erent advantages and disadvantages. De Castro Silva, Soma, and Maculan (2003) argue that vertical stability prevents carton boxes from falling down when being placed on top of other carton

(36)

boxes. Bischo↵ and Ratcli↵ (1995) further argue that horizontal stability prevent carton boxes to shift inside the container when its moved. Last, complexity constraints relate to limitations of technology and human resources, when complex loading patterns occur.

3.4 Packaging

To define packaging, three definitions suggested by Paine (1981) are commonly used, which refers to packaging as:

1. “A coordinated system of preparing goods for transport, distribution, storage, retailing and end-use”

2. “A means of ensuring safe delivery to the ultimate consumer in sound condition at minimum cost”

3. “A techno-economic function aimed at minimizing costs of delivery while maximizing sales (and hence profits)”

Packaging are further commonly divided into three di↵erent levels, primary packaging, secondary packaging and tertiary packaging, depending on where in the supply chain the packaging is found. Primary packaging, also referred to as sales packaging, is the packaging that end-customer’s see in stores. Secondary packaging, also referred to as grouped packaging, is the packaging used to enclose several primary packages, foremost used to make the handling in stores more efficient. Tertiary packaging, also referred to as transport packaging, is the packaging used to transport products from production to warehouses or stores and are usually not seen by end customers (). Examples of tertiary packages are wood pallets, shrink wraps and cardboard boxes.

3.4.1 The interface between packaging and supply chain efficiency

Packaging is strategically important as it has significant impact on logistics and supply chain per-formance (Hellstr¨om, 2007; Saghir, 2004b). The packaging a↵ects the efficiency of several logistic activities along the supply chain, from the point of packing to the point of emptying and recycling the packaging. To exemplify, a manufacturer wants packages easy to fill and store, a transport provider wants packages that enables for efficient loading whereas warehouses and stores want packages that enables for efficient picking, stacking, storing and replenishing. In resemblance, Hell-str¨om (2007) stresses that packaging a↵ect the cost of every logistic activity. Yet, logistical aspects

have frequently been disregarded when designing packaging containers (Saghir, 2004b). Hellstr¨om

(2007) argues that the reason to the disregarding of packaging’s impact on logistic activities could be due to lack of a holistic perspective. He further stresses the importance of that the packag-ing system should perform well with all present nodes in the supply chain in order to minimize sub-optimization. Further, Saghir (2004b) argues that the disregarding could be due to that pack-aging often is considered as a minor sub system of logistics, with limited influence on supply chain performance. Today businesses have started to realize that packaging is an important element

(37)

that can have large impact when trying to establish an efficient supply chain and the interaction and integration of logistics activities and packaging activities is now considered as an important

competitive advantage leading to increased business performance (Hellstr¨om, 2007). Besides taking

packaging logistics into consideration, Saghir (2004b) further highlights the importance of packing instructions and argues that packing instructions directly influence the time it takes to complete packaging of a product which in turn impacts lead times.

3.4.2 Packaging problems

A logistical challenge that retailers are facing is how to distribute a large variety of products to stores all over the world (Sung and Jang, 2018). Fisher (2004) mean that the complexity of this problem increases as seasonal fashion products need to be distributed to retailer’s own brand stores in thousands. In addition, fast response to the market is critical due to volatility in demand and short selling season for fashion goods (S¸en (2008), Mehrjoo and Pasek (2015)). Therefore, retailers commonly use assort-packing. Assort-packing is when the products are mixed, and packaging is done with consideration to the stores to which they will be distributed, to shorten the product distribution lead time from factories to retail stores (Chettri and Sharma, 2008). However, the traditional packing strategy for retailers is solid-packing, where factories pack the boxes with a single type of product or stock keeping unit. When the solid-packed cartons arrive at the distribution center they are unpacked, repacked, and then the products are distributed to the stores to which they are assigned. Solid-packing facilitates for the factories and is thus efficient for inbound delivery (DC deliveries). However, the distribution lead time are extended due to the need of extra labor activities at the destination market. Another constraint when it comes to assort-packaging and accurate product distribution to stores is the carton box size, which determines how many items that fit the carton box. Sung and Jang (2018) argues that the pack configuration is the biggest challenge for assort-packing. If products are distributed in the right carton sizes, the cartons might not need to be unpacked and repacked and will thus save the DC labor force a lot of time. However, pack configuration is complex, involving several parties which are impacted of the carton box size. Among them are third-party manufacturers, producing and packaging di↵erent products for several brands into carton boxes. That several parties are involved make it difficult to provide a set of carton boxes that meet every customers demand.

According to (Lee et al., 2014), using a one-sized packaging box for all products would minimize the inventory cost but would not be practical since it will lead to under and overstocking to the stores. On the other hand, if there is a packing type for all products, inventory costs will a↵ect the warehouse cost negatively. Moreover, Wilson (1965) argues that packaging problems make it difficult for cost optimization. Cardboard costs and warehouse space costs are minimized if utilization is maximized and the box fits each product exactly. Though, the cost savings from using di↵erent carton box sizes, are often o↵set because of higher price of ordering boxes in small quantities and due to increased complexity of handling the boxes in the warehouses. Thus, using carton boxes of only one size, minimize the costs for box inventory, handling and purchase. These contradictions lead to the problem of selecting an optimal number and sizes of carton boxes that