The impact of the Physical Internet on storage location and configuration of different seaports

(1)

The impact of the Physical Internet on storage

location and configuration of different seaports

Master thesis, Supply Chain Management

University of Groningen, Faculty of Economics and Business 15 July 2015

LEONIE VAN HELVERT Student number: 2030519 Folkingestraat 7a 9711 JS Groningen tel.: +31 (0)6-38742697 E-mail: l.m.van.helvert@student.rug.nl First Supervisor Prof. Dr. I.F.A. Vis

Second Supervisor Dr. Ir. P. Buijs Supervisor Field of Study

(2)

2

ABSTRACT

(3)

3

1.3.2 Literature study 06 1.3.3 Scenario study 07 1.3.3.1 Final scenarios 07 1.3.4 Case study 07 1.3.4.1 Case selection 08 1.3.4.2 Interviews 08 1.3.4.3 Data Analysis 09 1.4 Outline 09 2. THEORETICAL BACKGROUND 10 2.1 Role of seaports 10 2.1.1 storage at seaports 10 2.1.2 Seaport classification 10 2.2 Physical Internet 11 2.2.1 Modularity 12 2.2.2 Interconnectivity 13 2.2.3 Redistribution 13

2.3 The Physical Internet and Seaports 13

3. SCENARIO STUDY 14

3.1 Creating factors 14

3.2 Scenario trees 18

3.2.1 Seaside scenario tree 18

3.2.2 Landside scenario tree 20

3.3 Choosing of the scenarios 22

4. RESULTS

24

4.1 Scenarios 24

4.1.1 Interview S. Pan 24

4.1.2 New conceptual scenarios 25

4.1.3 Interview L. Tavasszy, second time 27

4.1.4 Formation of final scenarios 28

4.2 Interviews at practitioners 31

4.2.1 Port of Rotterdam 31

4.2.2 Husa Transportation 32

4.2.3 Groningen Seaports 32

(4)

4

5. DISCUSSION

34 5.1 Current storage 34 5.2 Dominant port 34 5.2.1 Scenario 1 34 5.2.2 Scenario 2 35 5.2.3 Scenario 3 35 5.3 Ordinary port 36 5.3.1 Scenario 1 36 5.3.2 Scenario 2 36 5.3.3 Scenario 3 37

5.4 Comparison based on seaport typology 38

5.4.1 Storage 38

5.4.2 Role of seaports 38

6. CONCLUSION

40

6.1 Summary and recommendations 40

6.2 Limitations and further research 41

(5)

5

1. RESEARCH OBJECTIVE

1.1 Introduction

Current logistics performance remains limited by a system of independent supply chain networks. Here, firms use dedicated logistics services within their own network, where short-term transport notices cause less than truckload shipping (Sarraj, Ballot, Pan, Hakimi, & Montreuil, 2014). This, among others, has led to a global transportation efficiency estimated at lower than 10%, which illustrates the unsustainability of how physical objects are transported, handled and stored (Montreuil, 2011). The Global Logistics Sustainability grand challenge addresses this issue and requires a new logistics model to do so (Montreuil, 2011). The Physical Internet Initiative (PI) is such a new framework for innovations that enables the creation of more efficient supply chain networks and operations. By creating an open logistics network, higher levels of efficiency can be obtained than with individual networks. This is done by employing interconnectivity on physical, digital and operational levels (Montreuil, Meller, Thivierge, & Montreuil, 2012). Identifying opportunities for the PI is the highest research priority according to the European Technology Platform for Logistics (ALICE, 2014a). Previous research on the PI has focused mainly on setting the basis and explaining its components as well as creating open mobility webs (e.g., Montreuil, Meller, & Ballot, 2010; Montreuil, 2011; Hakimi, Montreuil, Sarraj, Ballot, & Pan, 2012). However, no research on the implementation of the PI has been performed for seaports, even though they are named as hubs that are influenced by the PI (Hakimi et al., 2012) and their strategies run as far ahead as 2030. These strategies now do not take new logistics models such as the PI into account (Port of Rotterdam: Port Vision 2030; Groningen Seaports: Havenvisie 2030).

Therefore, this research focuses on the effects of the PI on seaports. Though there are multiple areas that need to be researched before a full image of the PI is possible, each area deserves its own study. This research will focus on storage, since a changing environment influences the optimal storage strategies of seaports (Wan, Liu, & Tsai, 2009). Storage locations could change to providing different proportions or even moving to different locations altogether if that is what shipping firms prefer. This study intends to comprehend the effects of the PI on the role of seaports and specifically on the demand for storage. For seaports, this is important to take into account when making decisions regarding e.g. technology or routes to invest in. If the different trajectories are not considered, the wrong policy might be adapted.

(6)

6

1.2 Research Questions

The main research question of this thesis is “What is the possible impact of the Physical

Internet Initiative on the demand for storage in different seaport types?”

Sub questions:

1. What is known in literature and professional publications with regards to the characteristics of container storage?

2. What is known in literature and professional publications with regards to different types of seaports?

3. What is known in literature and other forms of (academic or professional) publications with regards to the Physical Internet Initiative and storage aspects?

4. What are different scenario methodologies?

5. What are different scenarios that the Physical Internet Initiative could lead to?

6. How do the created Physical Internet Initiative scenarios influence the demand for storage and storage locations for containers in different types of seaports?

7. What are opportunities and threats for different types of seaports in the created Physical Internet Initiative scenarios?

8. How do the created Physical Internet Initiative scenarios influence the role of seaports?

1.3 Methodology

1.3.1 Main research question

The aim of this research is to gain insight into the possible effects that the implementation of the PI could have on the demand for storage location and configuration at different types of seaports and/or their hinterland. In order to answer the main research question, a combination of two approaches was utilised. First, a scenario study was undertaken to create possible scenarios that depict different worlds where the PI is realised. A multiple case study was then used to determine the influence of the scenarios on seaports, both in terms of storage and strategy. The approaches are explained below.

The focus of this study is import, as this covers a majority of the shipments (Groningen Seaports, 2015). The unit of analysis is seaports, which implies that each seaport in the research accounts for one case. Though ‘seaport’ might seem broad, different institutional models exist (Zauner, 2008). The broad term was chosen to ensure inclusion of all models.

1.3.2 Literature study

(7)

7

1.3.3 Scenario study

For sub questions 4 and 5 a scenario study was undertaken. Scenarios seek to identify factors that shape conditions under which a certain future might unfold (McDowall & Eames, 2006). Scenario studies can help firms remain competitive by analysing possible future requirements needed. This relates to the major contribution of scenario studies: thinking in alternative worlds, which can aid in dealing with uncertainty (von der Gracht & Darkow, 2010).

To select the appropriate scenario method, a literature search was done to outline possibilities after which a semi-structured interview took place. The interview was held with L. Tavasszy, who is a professor in the field of freight transport and logistics as well as knowledgeable on scenario studies. The interview protocol can be found in Appendix 1, part A. The questions focused both on scenario methodologies on their own and in combination with the PI in order to determine the proper methodology for this research. An exploratory analysis based on scenario trees was chosen because it is still unknown how the critical factors will act and how the PI will be realised. Back casting is utilised: starting from the assumption that a certain form of the logistics market is desirable, here the realisation of the PI, and researching how this market is possible (McDowall & Eames, 2006).

1.3.3.1 Final scenarios

To form the final scenarios, interviews were held with university scholars that are experts in the field of the PI. Semi-structured interviews were undertaken since they provide structure while allowing adaptations during the interview. They also allow pursuit of complex issues surrounding novel subjects (Zauner, 2008). The first interview was with S. Pan, a researcher in the field of supply chain management with a focus on open logistics and the PI. He has been involved in several publications on the PI, of which one as main author. The interview protocol can be found in Appendix 1, part B. Revisions to the scenarios were made and taken to the second interview to evaluate new possibilities while gaining input on progress made. This interview was with L. Tavasszy, the same expert as for the scenario methodologies, so he was aware of the study. Also, he is part of the European Technology Platform where the PI has a very high research priority (ALICE, 2014a). The interview protocol can be found in Appendix 1, part C. Based on these results the final scenarios were determined.

1.3.4 Case study

(8)

8

1.3.4.1 Case selection

In order to collect all necessary data, a purposeful selection of cases was made to include replication logic. For this research, the case selection was made based on theoretical replication, where the results for the cases were expected to be different but the reasoning behind it logical and predictable (Karlsson, 2009). The most important selection criteria relates to the type of seaport. As will be explained in the literature review, this can be further divided into a shipping network and inland network. The location of seaports also played a role in the case selection, since this could determine whether other storage locations were readily available. Finally, the presence of container terminals could determine the intensity of the influence of the PI and therefore it is also taken into account. Two cases are ultimately selected: Port of Rotterdam and Groningen Seaports, as can be seen in table 1.

Table 1. Case selection criteria

Selection criteria Port of Rotterdam Groningen Seaports

Shipping network Continental Regional

Inland network Multifunctional Intermodal

Location of port South Holland Groningen

Presence of container terminals

Yes Yes, but no focus point

1.3.4.2 Interviews

For the Port of Rotterdam, an interview was held with M. Nijdam, who works at the corporate strategy department and is familiar with the PI. Moreover, he has performed research on cross-border cooperation between seaports, which relates to the open network of the PI. A second interview was held at Husa transportation with S. Ramovic, the business development manager. He works at rail terminal Veendam, which is located in the hinterland and thus provides a different perspective on storage and strategic aspects. He has contact with both the Port of Rotterdam and Groningen Seaports, from whom Husa rents their inland terminal. The last interview was held as a roundtable session at Groningen Seaports. Present here were T. Smit from Groningen Seaports, S. Timmer from Sealane and F. Potze from ColliCare Logistics. This group represented the port authority and some of its partners that work in import and logistics. The diversity ensured different perspectives, while the group was kept small to prevent dominance of one member (Karlsson, 2009). Also, a phone interview was held with E. Bertholet, the business manager logistics at Groningen Seaports, to ensure inclusion of all storage aspects while not taking too long for the already extensive roundtable session. The interview protocols can be found in appendix 1, part D and E. Questions related to current storage and the influence of the PI scenarios on storage and strategy.

(9)

9 seaports are globally oriented and thereby are affected by international affairs and global economic conditions, more so than other firms (Ircha, 2001). The SWOT analysis was used as background for the questions asked at seaports that related to strategy. The results are mentioned in chapter 5 for completeness, though they are not used in the discussion as the scenarios were not included in the workshop.

1.3.4.3 Data analysis

According to Eisenhardt (1989), data analysis consists of two main steps: analysis within one case followed by a pattern search amongst different cases. For the analysis of one case, a distinction was made between relevant and irrelevant data. This data reduction was done by removing different words and phrases that did not directly relate to storage or strategy. This made different aspects visible that came up in interviews. The relevant data was organised based on scenario and subject (location, configuration or strategy). The research setup then looks as follows:

Figure 1. Research setup

1.4 Outline

The rest of this paper is structured as follows: previous literature will be discussed in chapter two. Chapter three will hold the factors and formation of initial scenarios, after which chapter four will hold the final scenarios based on interview results as well as results from interviews at seaports. Chapter five is a discussion of the effects of the scenarios on storage and strategy. Chapter six is the conclusion.

The Physical Internet - in general - - storage aspects -

Seaport typology

Interviews & workshops with seaports

Creation of conceptual scenarios

Verification interviews with experts on the Physical Internet Traditional focus on

(10)

10

2. THEORETICAL BACKGROUND

This section provides an overview of the role of seaports, both regarding storage aspects and seaport typologies. Furthermore, the role of the PI is discussed. For the PI, it is important to realise its exact components and ideologies, but also the aspects that will have the most influence on seaports to be able to determine its influence on storage.

2.1 Role of seaports

In the last decades, the role of seaports in general has been changing to working in an environment that is concentrated on supply chain networks and logistics, rather than focusing solely on the seaport’s own attributes (Panayides & Song, 2013). This is due to changes in the logistics system, which seaport competitiveness is dependent on but which is external to the seaport (e.g., Verhoeven, 2010). A seaport that works within a supply chain network can show a decrease in wastage and cost reduction by the use of Just-In-Time (JIT) systems (Panayides & Song, 2009). Furthermore, due to an increasing size of container vessels and thus their capacity to carry containers, a hub-and-spoke network has become popular because shipping companies and seaports need to cater to this (Imai, Shintani, & Papadimitriou, 2009).

2.1.1 Storage at seaports

All seaports face decisions regarding the optimal storage locations for their freight before transportation occurs. For this temporary storage, a container terminal serves as a short-term location for containers after which they are loaded onto a different vessel (Wan et al., 2009). These terminals make use of intermodal shipping, where the same transport unit is used even though the usage of different transport modes is possible (Sörensen & Vanovermeire, 2013). With the widespread implementation of containers, seaports adapted their terminals to include containers and perform transshipments (Steenken, Voß, & Stahlbock, 2004). This changed the layout of seaports and increased the storage of containers at terminals based in or near seaports. This was done either by building new terminals or making increased use of container stacking (Jiang, Chew, Lee, & Tan, 2013). An example is the Maasvlakte 2 at the Port of Rotterdam. Here, 2,000 hectares of sea are drained to boost the accessibility of the seaport while at the same time increasing the depths of the sea1. The container handling capacity will be doubled, which is deemed necessary due to its expected growth from 25 per cent to 42 per cent of total handling (Port of Rotterdam: Port Vision 2030).

2.1.2 Seaport classification

There are many different types of measures for seaports. Previous research has focussed, among others, on the performance measurements and evaluation factors of seaports (Murphy, Dalenberg, & Daley, 1988; Bichou & Gray, 2004). Another aspect of seaport typology is ecological. Here, seaports are seen in terms of their spatial and environmental development: sustainability, waterfront regeneration, and governance issues (Bichou & Gray, 2005;

1

(11)

11 Notteboom & Rodrigue, 2005). Supply chain elements have also been covered in combination with complexity and clusters (de Langen, 2002; Robinson, 2002). Other focuses have been on size, influence, (none) sea-borne trade and sectors (Bichou & Gray, 2005).

For the purpose of this research, these classifications are all either too broad; supply chain wide – or too narrow; governance issues. In their paper, Park and Medda (2010) introduce a seaport classification system that focuses on a combination of modality and hinterland connections, both of which are needed for an understanding of demand for storage location. First, a shipping network is needed for a seaport to work as a gateway: it connects the seaport with other seaports. These other seaports can be in other regions, but also other countries. The mode of transportation can be per ship, but other modes can be used as well. Depending on the location of the seaport, it plays a role in a continental, regional or feeder shipping network. A continental shipping network supplies direct global services to other countries or even continents through scheduled shipping routes. A regional shipping network, on the other hand, focuses mainly on direct routes to the same region. Other countries are usually managed indirectly. A feeder shipping network mainly has indirect routes also for the regional services. Aside from shipping networks, inland networks are needed for a seaport to connect with its hinterland and backward areas. This regards the mode of transportation used. First, a multifunctional inland network used multiple modes of transportation and supplies diverse services. An intermodal inland network has a small hinterland and backward area, which themselves connect to other inland modes of transport. Finally, a simple inland network connects the seaport to just its hinterland with mainly truck and rail, hardly by waterway. The following classification matrix can then be made:

Table 2. Seaport classification matrix (Park & Medda, 2010)

Shipping network

Continental Regional Feeder

Inland network

Multifunctional Dominant Port Versatile Port Specialized Port

Intermodal Superior Port Ordinary Port Industrial Port

Simple Intermediary Port Developing Port Peripheral Port

2.2 Physical Internet

(12)

12 The PI knows 13 founding and organisation principles, which can be found in table 3 (Montreuil, Ballot, Glardon, & McGinnis, 2011). In this research, three elements are used that relate to these principles as they are expected to be of greatest influence on seaport storage: modularity, interconnectivity and redistribution. Though the PI entails entanglement of all 13 principles, several come forward most in the area of this study. The other principles are less relevant and therefore excluded. The elements are clarified below in relation to the principles.

Table 3. Founding and organisation principles of the PI (Montreuil et al., 2011)

Founding principles Organisation principles

Instrumentality Interconnectivity Responsibility Uniformity Metasystemisation Accessibility Openness Unicity Universality Encapsulation Agentification Contracting Certification 2.2.1 Modularity

Modularity relates to several of the 13 principles, most noteworthy encapsulation and unicity, through the use of π-containers. It is employed in the PI by generalising and extending the current used standardised packaging systems to a grand scale (Montreuil, 2011). Whereas at the moment, the usage of containers of 20 or 40 twenty-foot equivalent unit (TEU) is possible alongside a smaller level of modularity used for parcel logistics, the PI puts objects within those containers in a standardised set of measurements. From here onwards, these are called TEU containers (20 or 40 TEU) and containers (smaller sizes). The composition of the π-containers can be found in figure 2, which shows how all of the different sizes fit together. The different lengths in meters are 0.12; 0.24; 0.36; 0.48; 0.6; 1.2; 2.4; 3.6; 4.8; 6; and 12. π-containers are transported to their destination by using the most economical and efficient route through the use of shared transport, handling and storage (Ballot, Montreuil, & Fontane, 2010). Moreover, the π-containers are able to interlock automatically and are equipped with sensors that make identification possible, which make the π-containers flow through transport and handling easily (Montreuil, 2011).

(13)

13

2.2.2 Interconnectivity

Interconnectivity relates to universal connections of all principles of the PI to develop a global interlaced network (Montreuil et al., 2011; Sarraj et al., 2014). Aside from using modularity to increase interconnectivity, which is discussed as a separate factor, it relates to digital and operational connections (Montreuil et al., 2012). Digital connectivity ensures exchange of information across the PI. This exchange includes e.g. tracking objects, messaging amongst actors and visibility of the flow. Operational connectivity ensures that business processes are interlaced through a standard interface to increase the usability of the PI. Furthermore, due to these technological and operational connections, use of the standardised interface ensures that transfer of the π-container is possible through any means of transport, handling or storage (Montreuil, 2011). The π-container acts as an autonomous agent through the use of connective technologies such as radio-frequency identification (RFID), which allows tracking anywhere in the world at all times.

2.2.3 Redistribution

Redistribution deals with the physical interconnectivity of the PI (Montreuil et al., 2012). With the use of redistribution, physical entities such as the π-containers can seamlessly flow through the logistics system of the PI. To realise this, an open distribution web is needed that stores objects in the most efficient location. Packages can use any mode of transportation and be stored in any warehouse. The use of dedicated warehousing does not happen anymore, but organisations can use the PI to determine which location is best at that point in time.

2.3 The Physical Internet and Seaports

Empirical research on the implications of the PI is lacking due to its so far limited application in practice, especially where seaports are concerned. Consequently, the influence that the PI could have on seaports remains unclear despite the fact that the role of seaports is already transitioning and preparation is needed for when the PI is realised. For storage, this now takes place at or near the seaport, but it is unknown whether this will be the same in a world where the PI is realised. It can be said that modularity is implemented in the use of TEU containers, which indicates the effects of standardised packaging. However, their transportation network is not joined: numerous different supply chains overlap each other (Montreuil et al., 2012). If the π-containers are implemented in an open global system, seaports need to know how they can prepare for this and what their role could be. Moreover, the influence of the PI on seaports could be different for alternate types of seaports with varying importance of elements.

(14)

14

3. SCENARIO STUDY

This chapter determines the conceptual scenarios based on scenario trees of different factors. The factors are chosen based on a stepwise reduction approach.

3.1 Creating factors

In this study, factors will first be chosen that are most likely to have the highest impact on the realisation of the Physical Internet. During the search for publications, the emphasis was on scenarios dealing with some form of logistics or supply chains. Table 4 shows an overview of the 11 scenario studies used and their number of factors. In total, 77 factors appeared, though several were related to the same focus point, e.g. environment, climate change and environmental trends. The full list of factors can be found in Appendix 2. Not included are smaller studies that focus on one factor.

Table 4. Scenario studies

Research # Factors

1. Four futures of Europe 5

2. Future ports scenarios for 21ST century ports 5 3. Future scenarios: Implications for port planning 9 4. Intelligent infrastructure futures. The scenarios – towards 2055 5

5. Logistics 2050, A Scenario Study 14

6. Port Vision 2030 9

7. Prospects and opportunities of ICT and media 9

8. Report on transport scenarios with a 20 and 40 year horizon 8 9. Scenarios for the logistics services industry: A Delphi-based analysis for 2025 5

10. Shell energy scenarios to 2050 4

11. Some effects of hinterland infrastructure pricing on port competitiveness:

Case of Antwerp 4

(15)

15 1. # mentions

2. PIRelation 3. Scopeofstudy

4. Overlap

Table 5. Created factors

Factors from scenario studies Created factor

Environmental stewardship – Environment – Level of climate change – Climate change – Environment – Environmental trends –

Environment

Climate change

Resilience and reliability – Security and safety – Safety – Robustness – Political stability and economic security

Control Demand patterns – Labour supply – Capital mobility – Society –

Dominant consumer needs – Society – Social acceptance – Demographic trends – Social change – Socio-cultural – Demand

Demand patterns

Finance and economics – Land side accessibility – Container demand – Economics – Level and distribution of income – Labour market and knowledge economy development – Overall economy – Economic trends – Economic, import and export assumptions

Economic growth

Energy prices – Energy price and energy mix – Availability and price of raw materials – Availability of raw materials – Energy mix – Energy trends – Resources

Energy

International trade – Container demand – Economics – Distribution of production and world trade – World economy – Economical –

Economic trends – Economic, import and export assumptions

Globalisation

Trade regulation – European production – Policy issues – Political-legal

Governance – protectionism Regulatory and spending policies – Logistics and transport regulation

– Investments by the state or private enterprises – Infrastructural development and congestion – Policy assumptions

Governance – regulations Port competition – Ship size – Transport scale – Port assumptions Hub & spoke

Media sector – Media use /behaviour Media

Capacity – City-port relationship – Modal shift – Quality of urban development – Infrastructure for international exchange – Supply chain integration – Industrial structure – Specific assumptions for inland navigation

Other

Technical change – Safety and security – Productivity – ICT systems and robotics – ICT – ICT sector – Technological trends –

Technological – Technology

Technology – automation & RFID Material technology – Technical advances – Technological Technology –

3D printing

(16)

16

 # Mentions: all possibilities under the heading ‘other’ were excluded since they only came up once.

 PI relation: this eliminated the factor media, as this related to the means used and technology behind media, not to the power of media or any logistics aspect (Müncher Kreis, 2009).

 Scope of study: climate change relates to the level of global warming and has consequences on supply chain resilience (Steen Petersen et al., 2009). However, the PI was created as an answer to impending climate change (Montreuil, 2011) and the realisation of the PI is thus part of a solution to climate change.

 Scope of study: control in a PI-world needs to be researched before it can be used. Also, the PI enhances control (Ballot et al., 2010): it influences control instead of the other way around.

(17)

17

Table 6. Final set of factors

Seaside factors

Globalisation This relates to the internationalisation of a firm’s supply chain, mostly to lower production costs, and requires global freight transportation (Rodrigue, 2012). If globalisation decreases, transport distances will decrease and thus the status of ocean cargo (Deutsche Post AG, 2012).

Outcome: intercontinental or regional

Governance – protectionism

The contrast is in the influence of the government on the economy and its eagerness to cooperate with international organisations (de Mooij & Tang, 2003). Countries safeguard themselves if protectionism blooms. If not, trade is liberalised with minimal governmental influence.

Outcome: protectionism or liberalisation of trade

Hub & spoke A hub seaport is a node that connects smaller spoke seaports (Asgari, Farahani, & Goh, 2013). This can occur globally, where mega-ships occasionally visit sizeable hubs (Imai et al., 2009) or locally, where transport occurs between smaller regional hubs at a higher frequency.

Outcome: global or local

Technology – automation & RFID

Automation enables intelligent tools that can automatically operate and communicate (ALICE, 2014d). RFID enables real-time tracking and identification, which facilitate monitoring and automation (e.g. DeutschePostAG,2012). They enableopennessand interconnections.

Outcome: implemented in mass form, yes or no

Landside factors

Demand patterns – E-commerce

Demand patterns are a complex web that drives the supply chain structure (Taneja et al., 2009; Rodrigue, 2012). E-commerce has been growing at double digit speed (ALICE, 2014a), which might benefit modularity and thus the implementation of the π-containers.

Outcome: implemented in mass form, yes or no

Economic growth The need for ocean freight depends on the socio-economic situation as economic- and transport developments are linked (Rodrigue, Comtois, & Slack, 2006). Economic growth determines the size of demand of seaports, where high growth produces more demand and thus chances.

Outcome: high or low

Governance – regulations

Seaport governance can be public, private or mixed. For the PI, it should serve as interface between freight systems (ALICE, 2014a). A public authority can use policy instruments to stimulate developments.

Outcome: majority public, majority private or mixed

Technology – 3D printing

Objects can be directly manufactured layer by layer through the use of a 3D printer (Deutsche Post AG, 2012). If it is widely implemented, for seaports the flow of raw materials increases while that of (semi-) final goods decreases (e.g. ALICE, 2014d).

(18)

18

3.2 Scenario trees

Two scenario trees are constructed. In order to build the scenario trees, the four final factors from both seaside and landside were put in causal order if this seemed appropriate based on previous literature. This was not necessary for all factors, but it was useful for the three elements of the PI that could be either crossed off or would definitely be realised. In the next section, the relation of each element to the factors will be discussed, which leads to the expected elements realised from the different scenario tree outcomes.

3.2.1 Seaside scenario tree

Figure 3. Scenario tree seaside Governance

Protectionism Globalisation Hub & Spoke

Technology Automation & RFID

Protectionism Liberalisation of trade Intercontinental Regional Global Local Yes No Yes No Yes No Yes No Yes No Yes No Yes No Yes No Global Local Global Local Global Local Intercontinental Regional Start

(19)

19 Regarding interconnectivity, the most important factor is technology, since the proper technology is required in order for modules to be connected (Montreuil et al., 2012). If this is not the case, interconnectivity is not possible regardless of the outcome of other factors. If there is liberalisation of trade, interconnectivity has a high possibility unless trade takes place regionally and within local networks, since global collaboration in some form is necessary for innovative technology to be created (Deutsche Post AG, 2012). Countries might separately come quite far, but the level required will not be reached. If protectionism occurs, interconnectivity will not take place, since nationalistic measures do not allow for a high level of cooperation on grand scale (de Mooij & Tang, 2003).

The last factor, redistribution, has a high chance of taking place in a liberalised world, since this allows for openness and collaboration (ALICE, 2014a). If countries are protectionist over their locations, warehouses and the economy in general, the chances that redistribution will occur therefore drop. The only way redistribution might still take place in a protectionist society is if trade takes place regionally and with a localised hub and spoke network. This way, redistribution occurs on a smaller, country- or region- based scale. The following table with outcomes can then be formed:

Table 7. Seaside factors outcomes

# Governance Globalisation Hub/ spoke

Tech Modularity / interconnectivity / redistribution

(20)

20

3.2.2 Landside scenario tree

Figure 4. Scenario tree landside

Governance

Regulations Economic Growth

Technology 3D Printing Demand Patterns E-commerce High Low Yes No Yes No Yes No Yes No Yes No Yes No Yes No Yes No Yes No Yes No Yes No Yes No High Low Start Yes No Yes No Yes No Yes No Yes No Yes No High Low Majority Public Majority Private Mixed

(21)

21 Interconnectivity will be realised in public settings if economic growth is high and there is 3D printing with e-commerce or no 3D printing. Concrete plans are needed for governments to spend their scarce resources. Therefore, if growth is high and there is 3D printing but no e-commerce, there is no interconnectivity. If economic growth is low in public settings, there will be no investments in interconnectivity in general. For the two other regulations, private and public, interconnectivity will be realised if there is high growth or if there is low growth but no 3D printing. In the former situation, there are funds to invest while the latter situation requires efficiency. If there is low growth and 3D printing, no interconnectivity is realised since money will be spend on more pressing matters (Stocker et al., 2014).

With regards to redistribution, this occurs if regulations are public. This way, it will be easier to gain a large network for storage by governmental support actions in developing logistics systems (OECD/ITF, 2015). If the market is privatised, redistribution is unlikely because firms are not driven to start themselves. The differences for the other factors therefore lie in a mixed sector due to influences of both the private and public sector. If economic growth is high, new developments such as redistribution are undertaken (Curry, Hodgson, Kelnar, & Wilson, 2006). If economic growth is low, redistribution will only occur in the presence of e-commerce, since efficient storage is required. The following table can then be formed:

Table 8. Landside factors outcomes

# Governance Economic growth

3D E-commerce Modularity / interconnectivity / redistribution

2.1 Majority public High Yes Yes + / + / +

2.2 Majority public High Yes No - / - / +

2.3 Majority public High No Yes + / + / +

2.4 Majority public High No No + / + / +

2.5 Majority public Low Yes Yes - / - / +

2.6 Majority public Low Yes No - / - / +

2.7 Majority public Low No Yes - / - / +

2.8 Majority public Low No No - / - / +

2.9 Majority private High Yes Yes + / + / -

2.10 Majority private High Yes No - / + / -

2.11 Majority private High No Yes + / + / -

2.12 Majority private High No No + / + / -

2.13 Majority private Low Yes Yes - / - / -

2.14 Majority private Low Yes No - / - / -

2.15 Majority private Low No Yes + / + / -

2.16 Majority private Low No No + / + / -

2.17 Mixed High Yes Yes + / + / +

2.18 Mixed High Yes No - / + / +

2.19 Mixed High No Yes + / + / +

2.20 Mixed High No No + / + / +

2.21 Mixed Low Yes Yes - / - / +

2.22 Mixed Low Yes No - / - / -

2.23 Mixed Low No Yes ? / + / +

(22)

22 1. Black swan 2. Impossible combinations 3. Likelihood of occurring 4. Different worlds

3.3 Choosing of the scenarios

As tables 7 and 8 show, all scenarios have been given a number. For the development of the conceptual scenarios, at least three scenarios are needed to describe the possible alternative futures (Valeriu, 2014), while more than four scenarios are undesired since this is outside one’s cognitive possibilities (Steen Petersen et al., 2009). The choice was therefore made to construct four conceptual scenarios. To determine which scenarios would be used, first those that contained either a question mark or did not contain any of the PI elements were removed from the selection. This eliminated 1.3 through 1.6, 1.15, 2.13, 2.14, and 2.22 through 2.24. Two scenarios could be formed from combining factors from seaside and landside, while the other scenarios were formed from one of the two sides. A stepwise deletion approach was undertaken to decide on the combinations to be utilised. For combinations to be possible, the same outcome in terms of elements had to be realised.

First, decisions were made for the scenarios from both sides. Here, no black swans were incorporated. During the initial scenario interview with L. Tavasszy he mentioned that these are conditions with deep uncertainty that could have a vast impact on the future. For the scenarios, that would be 3D printing as this has the ability to disrupt whole supply chains to their core, relating to scenario 2.1, 2.2, 2.5, 2.6, 2.17 and 2.21. The second step entailed removal of impossible combinations. Here, scenarios that had regional/global, intercontinental/local and e-commerce/local were removed: numbers 1.11, 1.13 and 2.7. The third step related to the likelihood of occurrence. This was based on combinations of factors and eliminated scenario 1.7. Finally, the fourth step deleted combinations that would already definitely occur in other scenarios as to increase the differences between worlds (Deutsche Post AG, 2012). This eliminated 2.3, 2.4 and 2.20 and leads to the following two scenarios:

Scenario 1: modularity, interconnectivity and redistribution

Governance Globalisation Hub / spoke Technology

Liberalisation of trade Intercontinental Global Yes

Regulations Economic growth 3D printing E-commerce

Mixed High No Yes

Scenario 2: redistribution

Protectionism Regional Local No

(23)

23 1. ≥ ++ 2. Black swan 3. Impossible combinations 4. Likelihood of occurring 5. Different worlds

Scenarios 3 and 4 were made from the remaining scenarios and are based on only one aspect, either seaside or landside. The

decision-making process here is mostly the same, with a few changes. In the first step, it was important to choose scenarios with two elements of the physical internet in order to create the biggest change in future worlds from current society. This eliminated 1.1, 1.2 and 2.10. In step two, instead of eliminating black swans, one black swan was implemented with 3D printing to comprehend its influence and how seaports could prepare. All landside scenarios without 3D printing were therefore not taken into account, removing 2.11, 2.12, 2.15 and 2.16. The third step was impossible combinations and eliminates 1.10, 1.12 and 1.14. For the likelihood of occurring, nothing had to change while the last step, related to different worlds, excludes 2.18. This leads to the following scenarios:

Scenario 3: modularity and redistribution

Liberalisation of trade Regional Local No

Scenario 4: modularity and interconnectivity

Majority private High Yes Yes

Appendix 5, part A and B show a colour scheme of the decision process for clarification purposes. To determine the influence of the different scenarios on the role of seaports and their storage, it is important to understand which future has an effect on what type of seaport and how. To determine this, the following conceptual model will be researched:

Table 9. Conceptual model with potential scenarios

PI scenarios Determined by factors 1 2 3 4 PI elements Determined by micro-factors

Modularity Yes No Yes Yes

Redistribution Yes No No Yes

Interconnectivity Yes Yes Yes No

Seaport typology

Port type 1 ?

(24)

24

4. RESULTS

This section provides an overview of the findings, divided into two segments. First, the results from the scenario study are presented and the final scenarios created based on verification interviews with experts on the PI. This answers sub question 5. The scenarios are then used in interviews at seaports, which is the second part: the results from those interviews, leading up to answering sub questions 6, 7 and 8 in chapter 5.

4.1 Scenarios

To determine correctness of the factors as well as their influence and the final scenarios, two interviews were held with university scholars. After the first interview, the scenarios were adapted and taken to the second interview for confirmation. The final scenarios were created after the second interview. Methodological details are explained in chapter 1. Unless mentioned otherwise, the results provided were given by the interviewee(s).

4.1.1 Interview S. Pan

First, a discussion of the factors took place. The interviewee was uncertain about the influence of 3D printing, especially since not everything can be printed, such as shampoo. Moreover, he thought that π-containers might also be used to transport raw materials, thereby minimising the effect of 3D printing on the logistics system. This factor needs to be reconsidered as the interviewee is uncertain and does not know if it will have an impact. The factor e-commerce came forward as being important, since it is a trend and “more and more people are doing

online shopping”. The same can be said for regulations, where especially the outcome mixed

is significant. Finally, redistribution does not only blossom in public or mixed governance, but is also thought to be possible from private governance. This is due to the possibility of flow management where “[private firms] rent a space in a platform which is provided [at

seaports] and they use this space to create some service for […] the in and outbound flows.”

(25)

25

4.1.2 New conceptual scenarios

After the first interview, all new insights mentioned above were taken into consideration and the following new factor outcome tables could be created:

Table 10. Seaside factor outcomes, version 2

# Protectionism Globalisation Hub/ spoke

Tech Modularity / interconnectivity / redistribution

(26)

26

Table 11. Landside factor outcomes, version 2

# Regulations Economic growth

3D E-commerce Modularity / interconnectivity / redistribution

2.1 Majority public High Yes Yes + / + / +

2.2 Majority public High Yes No - / + / +

2.3 Majority public High No Yes + / + / +

2.4 Majority public High No No + / + / +

2.5 Majority public Low Yes Yes - / - / -

2.6 Majority public Low Yes No - / - / -

2.7 Majority public Low No Yes - / - / -

2.8 Majority public Low No No - / - / -

2.9 Majority private High Yes Yes + / + / +

2.10 Majority private High Yes No + / + / +

2.11 Majority private High No Yes + / + / +

2.12 Majority private High No No + / + / +

2.13 Majority private Low Yes Yes + / - / -

2.14 Majority private Low Yes No + / - / -

2.15 Majority private Low No Yes + / + / -

2.16 Majority private Low No No + / + / -

2.17 Mixed High Yes Yes + / + / +

2.18 Mixed High Yes No + / + / +

2.19 Mixed High No Yes + / + / +

2.20 Mixed High No No + / + / +

2.21 Mixed Low Yes Yes - / - / -

2.22 Mixed Low Yes No - / - / -

2.23 Mixed Low No Yes + / + / +

2.24 Mixed Low No No - / + / +

From the interview, scenario 1 was found plausible and consistent and the outcome of the separate PI elements has not changed based on the new insights (# 1.9 and 2.19). Therefore, it will remain as before. For the other scenarios, combinations could now be made between modularity/interconnectivity and interconnectivity/redistribution. These combinations will therefore form scenario 2 and 3. A possible fourth scenario was opted for, but since this would again have all three elements it is expected that the change for seaports would be trivial. The choice was thus made to use three scenarios, which is an appropriate number according to Valeriu (2014).

(27)

27

Scenario 1: modularity, interconnectivity and redistribution

Liberalisation of trade Intercontinental Global Yes

Mixed High No Yes

Scenario 2: modularity and interconnectivity

Liberalisation of trade Regional Local Yes

Majority private Low No No

Scenario 3 (two options for landside): interconnectivity and redistribution

Protectionism Regional Local Yes

Majority public High Yes No

Mixed Low No No

4.1.3 Interview L. Tavasszy, second time

Again a discussion of the factors took place. Most important here was the influence of 3D printing, specifically if transportation would take place in π-containers or bulk, since this was unclear from the interview with S. Pan. Though there is no image of the distribution structure, L. Tavasszy stated that “if supply chains remain as they are it would be favourable for the

π-container” and therefore it was discussed that bulk would be more likely if 3D printing carries

through. For e-commerce, it is quite clear that it will push through since “predictions are

currently that it will grow with 18% per year”. For the scenarios however, probability of an

affair does not matter but that it occurs or not, as came from the first interview with L. Tavasszy. Finally, the factor governance – protectionism was scrutinised. Here, it seems suggested that policies from port authorities can influence the speed of realisation of the PI, while it came from the interview that it should be more of an indirect effect that influences e.g. technology development.

(28)

28 the necessity for efficiency in low economic growth situations in a private setting, which drives the development of technology. The realised elements were also agreed upon. Finally, scenario 3 with 3D printing is both plausible and consistent. The realisation of the elements interconnectivity and redistribution was agreed upon as well as the removal of modularity, since the necessity of the π-container is omitted due to 3D printing.

4.1.4 Formation of final scenarios

All insights were used to form the final scenarios. The second versions of the conceptual scenarios were taken as base to see if adaptations were needed. Scenarios 1 and 2 were deemed plausible and consistent and the elements realised were agreed upon. Therefore, these scenarios did not change. For scenario 3, 3D printing can be seen as a black swan that determines the omission of modularity and therefore the option with 3D printing was chosen. The final scenarios are described in detail below. Each scenario starts with an image depicting the elements realised and the configuration of the eight factors.

Scenario 1: Untamed Globalisation

In scenario 1, trade is liberalised and takes place intercontinentally in a global hub & spoke system. This implies that mega ships arrive at international hub seaports at low frequency but with a high volume of TEU containers. The PI blossoms and the TEU containers are filled with smaller π-containers that move all over the world. Automation and RFID, as well as the technology behind the π-containers, are fully developed and available in the market: real-time tracking and an automated transportation flow are possible and interconnectivity is realised. Port authority has mixed regulations and economic growth is high, creating possibilities for innovations, particularly redistribution: the logistics system is open and storage is not dedicated. There is limited 3D printing in this world, but e-commerce thrives and online shopping is done from all over the world.

Untamed Globalisation Modularity Interconnectivity Redistribution Liberalisation of trade Intercontinental trade Global Hub & Spoke Technology developed

Mixed regulations High economic growth

(29)

29

Scenario 2: Local Preferences

In scenario 2, trade is again liberalised. However, it takes place regionally in a local hub & spoke system, which implies that smaller ships arrive at hub seaports at higher frequency but with a lower volume of TEU containers. Since modularity is realised, these TEU containers are filled with smaller π-containers. Automation and RFID, as well as the technology behind the π-containers, are fully developed and available in the market, making interconnectivity possible. This way, packages can be transported automatically and can be tracked at all times and places.

Port authorities are privatised and economic growth is low. Both modularity and interconnectivity can be realised despite low economic growth, because privatisation ensures the availability of funds through the need for efficiency. However, it limits possibilities for redistribution, since firms are not willing to open their storage systems. Moreover, though liberalisation stimulates redistribution, the combination of worldwide possibilities with local trade causes vague borders of what is possible and available and what not. Therefore, storage facilities remain dedicated to private firms and networks. There is limited 3D printing in this world and limited e-commerce.

Local Preferences Modularity Interconnectivity Redistribution Liberalisation of trade Regional trade Local Hub & Spoke Technology developed

Private regulations Low economic growth

(30)

30

Scenario 3: Protectionistic Localisation

In scenario 3, protectionistic measures have been taken by countries or small trade regions. Trade thus takes place regionally in a local hub & spoke system, which implies that smaller ships arrive at hub seaports at higher frequency but with a lower volume of trade. Interconnectivity and redistribution are enabled due to the combination of these three factors: there is protectionism, but trade takes place regionally with a local hub/spoke system, therefore enabling interconnectivity and redistribution on a smaller scale within countries or small trade regions. Economic growth is high, but there is limited e-commerce. In this world, 3D printing has taken over and thus caused a disruption in the container supply chain: there is no modularity as ocean freight takes place in bulk. The technology behind the PI is developed and regulations are public.

Resulting from the final scenarios, the following conceptual model is created, adapted from table 9:

Table 12. Conceptual model with final scenarios

Protectionistic Localisation Modularity Interconnectivity Redistribution PI scenarios Determined by factors 1 2 3 PI elements Determined by micro-factors

Modularity Yes Yes No

Redistribution Yes No Yes

Interconnectivity Yes Yes Yes

Seaport typology Port type 1 ? Port type 2 ? Protectionism Regional trade Local Hub & Spoke

No technology Public regulations High economic growth

(31)

31

4.2 Interviews at practitioners

After the final scenarios were determined, their influences on seaports needed to be identified. Three interviews were held at different seaports and some partners as well as a workshop with young professionals, as explained in chapter 1. The outcomes will be described below and then scrutinised in chapter 5. Unless mentioned otherwise, the results provided were given by the interviewee(s).

4.2.1 Port of Rotterdam

Current storage shows a trend where “different goods are stored at different locations”. Household goods generally move from the seaport to warehouses and distribution centres in the hinterland as quickly as possible. As soon as the TEU containers arrive at the warehouse, they are opened and emptied. For the small section that is stored in warehouses at the seaport, this is also in smaller units than a TEU container. There sometimes occurs temporary storage at the seaport terminal where TEU containers await further transport: this storage remains in TEU container and should be as short as possible. It is sometimes used by the client as storage as well, though this is discouraged by the seaport. For this temporary storage at the seaport, it is possible in π-containers though that would be difficult in the current system. Moreover, it would “increase the number of operations which means the space required at the terminal

grows” due to increased handling of small objects and therefore reduce efficiency. A

breakdown location is required, but at the seaport the directional flow of goods is still quite thick. A possibility could be at Alblasserdam, where a new TEU container transferium was just opened to get TEU containers off the Maasvlakte swiftly. TEU containers would “quickly

go there through inland waterways and further container breakdown could take place there”.

Figure 5 shows the location of the TEU container transferium Alblasserdam.

Figure 5. TEU container Transferium Alblasserdam2

For scenario 1, π-containers should be transferred in batches as large as possible, so “the

appearance of the container will not really change”. This is the case at the deep sea terminal,

because breakdown of the TEU containers to π-containers will occur later: in warehouses at the seaport and further away. For strategy, advantages of the PI lie in the interconnectivity aspect which increases efficiency. Moreover, the distinctive character of logistics service providers diminishes. The competitive position of the port would then improve, since “it is

only about scaling and efficiency and Rotterdam is better in that than competition”.

(32)

32 Regarding scenario 2, “smaller ships increase the possibility of working with π-containers at

the terminal”. Though most storage would still take place outside the seaport, TEU containers

could be emptied at the terminal and moved in π-containers to the hinterland. This scenario nevertheless brings mostly threats to the Port of Rotterdam, since global trade drops and with it economies of scale: this threatens the competitiveness of the port, which is based on scale. Scenario 3 decreases the workload of the seaport and brings more work to the hinterland, since that is where production takes place. Large-scale storage of the bulk could take place at the seaport itself. This scenario has a “commercial challenge to get the materials through the

Port of Rotterdam” and perhaps even to get large 3D printers near the seaport. A downside is

the possible needlessness of the Maasvlakte 2, since less space is required.

4.2.2 Husa Transportation

Currently, TEU container storage takes place on the terminal in Veendam because Groningen Seaports is purely for handling and Husa is closer to the customer. TEU containers can be stacked at the terminal fully loaded, emptied at a warehouse on site or transported to the customer immediately. Storage at the terminal is possible in π-containers, but “for the firm it

has to pay off” financially as they need new handling materials to work with π-containers.

Scenario 1 would not change the location of storage. In terms of storage in π-containers, this would not be the case for the seaports but is definitely possible at Husa, from where storage can then be distributed in π-containers. However, this happens if “the customer wants it and

we wait until there is demand for it”. Strategy wise, large ships would still choose Rotterdam

over Groningen, though opportunities could arise if ships make use of all big and small hubs on their route. The question is then whether it is “efficient to dock at every seaport or not”. For scenario 2, π-containers determine the preferred storage locations, but this remains in the hinterland. The switch to π-containers should occur in the hinterland, but once there as early as possible. There are positive points to this scenario, the position of Groningen Seaports will become more important and will “be taken into the network of the Physical Internet”.

Storage in scenario 3 needs to take place close to the customer: the producer. Bulk transport increases the importance of seaports not focused on TEU containers and thus benefits Groningen Seaports. On the other hand, the commercial challenge is mentioned again where “they have to build a good network with distribution in the surrounding areas”.

4.2.3 Groningen Seaports

(33)

33 Storage in scenario 1 takes place in the hinterland, because it is important to get away from the seaport as quickly as possible, since only loading and unloading takes place there. The location in the hinterland is then determined by “what brings together the most modalities”. In this scenario e-commerce thrives. Therefore, though it increases handling storage should be in the small units of the π-containers. In terms of strategy, it is mostly an opportunity. The impact might be smaller than at large seaports; it might enable surprising new flows rooted in the PI standards.

Scenario 2 does not change anything location wise, though the form of storage could change to π-containers since this could provide a vantage ground in technology. However, this is mostly determined by whether there is demand for π-containers by the customers. A threat here is that no planners are needed anymore and there is hardly any distinctiveness between different logistics service providers due to standardisation of processes and transparency. The role of the seaport itself, however, might not change due to that but due to e.g. the local business climate: total competitiveness.

Finally, for scenario 3, storage would move to cheap locations since the value of bulk is relatively low. This would be at Groningen Seaports rather than the Port of Rotterdam. Moreover, Groningen Seaports is already an industrial area and for this type of port there is “room for handling that is also affordable”.

4.2.4 Workshop with young professionals

The workshop did not take into account a specific scenario but pictured a world where all three elements were realised. For the strengths and opportunities, this related especially to the connection with the hinterland of seaports. The idea of constructing a pipeline was opted. With this, the standardised π-containers can easily be transported to cheaper storage locations. Furthermore, the extended gate concept was brought up. This improves the flow and moves customs to a later stage.

However, there were also plenty of possible pitfalls. First of all, the borders of terminals will fade as there is no clear ownership. Relating this to ownership of containers, the costs for empty container trafficking need to be paid, but by whom? If nobody owns it, nobody wants to pay for it. It is still quite uncertain what will happen. The example of the Maasvlakte 2 was used, where an increase in handling costs combined with decrease in required space might make it superfluous. On the other hand, it is also possible that the use of proper technology could give it a boost and increase its position.

(34)

34

5. DISCUSSION

This section delivers a link between the previous sections. The relation with previous literature on seaport storage is discussed based on the results from the case studies at different types of seaports. Furthermore, the expected effect of the PI on storage location and configuration is determined as well as the influence the PI could have on different types of seaports in different scenarios. As shown in chapter 1, the Port of Rotterdam represents a dominant port with a continental shipping network and multifunctional inland network. Groningen Seaports signifies an ordinary port with regional shipping network and intermodal inland network. This terminology is based on the classification by Park and Medda (2010).

5.1 Current storage

Before comparisons can be made based on storage at and between seaports, knowledge of the current situation is required. From previous literature, a TEU container terminal can be used for short-term TEU container storage (Wan et al., 2009). It relates to results from the interviews, where this is mentioned for those TEU containers that await further transport. This temporary storage would always be in TEU containers. However, it came from interviews that TEU containers are transported to the customer and thus away from the seaport as quickly as possible. This is opposite to Jiang et al. (2013), who state that seaports changed their layout to increase TEU container storage possibilities at the terminal. Moreover, what came from multiple interviews is that TEU containers are emptied as soon as they reach a warehouse. Storage in warehouses is thus always in a smaller form than a TEU container.

5.2 Dominant port

The first seaport type to be discussed is the dominant port. A comparison of both storage and strategy aspects is made based on the scenarios, which will therefore be discussed one by one.

5.2.1 Scenario 1

For scenario 1, M. Nijdam stated that TEU container storage location would most likely remain in the hinterland. Regarding its configuration, the appearance of the TEU container would not change either: its breakdown would not be done at the seaport terminal but at the next step in the chain. It seems that scenario 1 would not provide differences with current storage systems, which might be because it is the scenario that the world is seemingly moving towards as came from the scenario interviews with S. Pan: “somehow it is already happening

today” and L. Tavasszy: “current scenario where we are headed towards”. Moreover,

(35)

35 Regarding strategy, the hub & spoke network enables the use of large vessels and economies of scale, which aids in bringing in liner shipping companies (Low & Tang, 2012). This coincides with the interview at a dominant port, where opportunities were found due to the diminishing distinctiveness of logistics service providers. It is then more likely that customers will move towards the seaport with the best economies of scale, which is the dominant port. This is further increased by the large ship sizes, for which the dominant port is most capable of accommodating (Asteris, Collins, & Jones, 2012). It would thus benefit greatly from the implementation of the PI in scenario 1 as economies of scale provide the dominant port with a reinforcement of their leading competitive position (Zauner, 2008; Low & Tang, 2012).

As for scenario 2, variance occurs from current storage. Though the hinterland is still highly used as storage location, there is a chance that π-containers will be used at the seaport itself. This is due to the localisation of trade and the hub & spoke system, which is thought to enable handling of π-containers at the seaport itself. However, this would relate to a small percentage as the rest would still move to the hinterland in TEU containers, which links to theory where terminal activities are capital intensive and handling should be the main service provided (Tovar, Jara-Díaz, & Trujillo, 2007; Ramos-Real & Tovar, 2010). Though trade is regional, a smaller amount of scale economies can still occur due to cargo handling (Tovar et al., 2007).

Strategy wise, this is almost the opposite of scenario 1. The dominant port is threatened by a drop in global trade and therefore in their competitiveness on economies of scale. This occurs since globalisation drives maritime transport where 80% of all global transport is done by sea vessels (OECD/ITF, 2015) and the dominant port’s advantages in this system were based on the global hub & spoke network (Low & Tang, 2012).

(36)

36 For the dominant port, table 13 can then be formed with storage location and configuration in the different scenarios. Strategy aspects will be taken into account at a later stage as this is dependent on the combination of different seaports.

Table 13. Storage locations and configurations of dominant ports

Dominant port

Location of storage Configuration of storage

Current In hinterland TEU container

Scenario 1 In hinterland TEU container

Scenario 2 Both at seaport and in hinterland Both TEU container and π-container

Scenario 3 At seaport Bulk

5.3 Ordinary port

The second seaport type to be discussed is the ordinary port. Again, a comparison of both storage and strategic aspects is made based on the different scenarios. The scenarios will be discussed one by one.

All interviewees stated that the location of TEU container storage would remain in the hinterland, because it should move away from the seaport as quickly as possible: the focus of the terminal is on handling (Tovar et al., 2007). Breakdown to π-containers could occur at Husa in Veendam, thus in the hinterland. However, only if the customer wants this, as customer power is high (Aronietis et al., 2010). As also mentioned for the dominant port, this is related to the current storage system as the current situation relates most to the world depicted in scenario 1.

For ordinary ports, scenario 1 is thought to be an opportunity to create new flows rooted in the PI standards, as movement between hubs is enabled in an interconnected market (Pan, Nigrelli, Ballot, Sarraj, & Yang, 2015). However, as the diminishing distinctiveness of logistics service providers takes place, it is more likely that the customers will move towards the seaport that has the lowest costs due to economies of scale (Low & Tang, 2012; Pan et al., 2015). For ordinary ports, the importance of economies of scale might defer routes away from them to one intercontinental mega hub: the dominant port. It is important to try to realise new flows, which can be done through intermodality and high quality transport networks (Veenstra & Zuidwijk, 2010), but this might be difficult without economies of scale.

The impact of the Physical Internet on storage location and configuration of different seaports