Towards a Physical Internet: An investigation of barriers to IT adoption

(1)

Towards a Physical Internet:

An investigation of barriers to IT adoption

Andrea van Luyk

S2545160

Master’s thesis

MSc Supply Chain Management

Faculty of Economics and Business

University of Groningen

Supervisors:

Dr. Ir. P. Buijs

Dr. X. Zhu

(2)

(3)

2

(4)

3

Abstract

Purpose: The aim of this study is to unravel barriers to IT adoption as a precedent to Physical Internet

implementation, and to investigate why some companies in a logistics network experience these barriers more than others.

Design/methodology/approach: Empirical data is collected by means of a single case study within a

port hinterland network, where 21 interviews were conducted with all stakeholders in the network. The focus of this study rests solely on container logistics.

Findings: Companies in port hinterland networks experience 12 different barriers to IT adoption, based

on a technology-organisation-environment theoretical framework. Within these networks, some stakeholders experience more barriers due to a high level of fragmentation of the hinterland network structure in terms of capacity and capabilities.

Originality/contributions: This paper contributes to the growing body of PI literature by shedding a

light on IT adoption barriers in port hinterland networks using an empirically and theoretically grounded approach. The study thereby also makes a contribution to IT adoption literature by taking a network-perspective on adoption rather than a single-firm network-perspective. Additionally, this study investigates how and why these barriers differ across stakeholders.

Keywords: Physical Internet, information technology (IT), adoption, barriers

(5)

4

Acknowledgements

I would like to take this opportunity to thank the people who contributed to the successful completion of this study. First of all, I would like to thank both of my supervisors for their flexibility in enabling me to finish my thesis throughout the summer. I am particularly grateful to my first supervisor, Paul Buijs. Paul, thank you for your commitment to the project and your involvement in my contacts with relevant business partners. Your accurate and supportive feedback provided direction when I was drifting, and the devoted interest you have for your research field is truly infectious. I really enjoyed working with you on this project.

Furthermore, I would like to express many thanks to my contacts at the Port of Rotterdam. In particular Michiel Nijdam, as he showed immediate interest in my thesis project and warmly welcomed me at the port. He also introduced me to a set of his colleagues who were very helpful in realizing my data collection. Donald Baan, Melinda Mossel and Martijn Thijssen, thank you all for helping me get a thorough understanding of the problem and for exploiting your networks for the sake of my research. Lastly, I would like to thank my friends and family for the unconditional support and belief in the successful ending of this thesis project. My closest friends, for always cheering me on as the first of all to graduate. My brother and sister, who have the same (and perhaps even deeper) affection for academia and research, whom I knew would always appreciate my struggles. Mom, dad and Janneke, thank you for your never-ending interest in me. I am truly grateful.

(6)

5

Key abbreviations

CC Cloud computing

DST Deep sea terminal

EDI Electronic data interchange

ETA Expected time of arrival

ETD Expected time of departure

GPS General Positioning System

ILT Inland terminal

IoT Internet of Things

PI Physical Internet

RFID Radio frequency identification

SL Shipping line

TEU Twenty feet equivalent unit

(7)

6

1. Introduction

After the revolutionary introduction of the Digital Internet, a similar concept has now announced itself in the logistics industry: the Physical Internet (PI). Early research has identified PI as the blueprint for a global system of networks founded on digital, physical and operational interconnectivity (Ballot, Montreuil, & Meller, 2014; Montreuil, 2011; Montreuil, Meller, & Ballot, 2013). Similar to how the Digital Internet aimed to optimize the transmission of data, the aim of PI is to optimize the efficiency of logistics networks by means of openly consolidating loads through a set of interconnected servers and facilities (Crainic & Montreuil, 2016; Pan, Ballot, Huang, & Montreuil, 2017). For this it requires mainly two components: standardized π-containers for transporting goods, and network-level Information Technology (IT) that will coordinate the flow of π-containers through the network.

In spite of the fact that PI is a very promising innovation, literature is not yet mature nor very much grounded in theory. However, the increase in PI research indicates that the relevance of this field is growing (Treiblmaier, Mirkovski, & Lowry, 2016). Quantitative models and simulations have dominated the current research field, as is often the case with innovative concepts in the logistics industry (Pan et al., 2017). Literature reviews are increasingly being published, concluding that practical and empirically grounded studies merit particular attention (Sternberg & Norrman, 2017). Implementation of PI as a whole could take decades, but as PI is determined to have a gradual roll-out, it is highly relevant to already academically explore the implementation process regardless of the fact that there is still uncertainty on a number of conceptual issues. The adoption process of PI, and its IT properties in particular, can therefore still be further investigated at this time (Sternberg & Norrman, 2017).

(9)

8 The aim of this paper is henceforth to explore the barriers to IT adoption in port hinterland networks as a precedent to PI implementation. The initial results of this study indicate that there are multiple barriers to IT adoption which differ significantly across stakeholders in the network. Therefore, an additional aim of this study will be to also explore what rationales constitute the differences of these barriers among different stakeholders. Therefore, this paper addresses the following research question:

RQ What are the barriers to IT adoption in port hinterland networks, and why do these barriers

differ across parties in these networks?

This paper contributes to the growing body of PI literature by shedding a light on the adoption process of IT within the PI implementation process, and is among the first to conduct a broad empirical investigation with respect to PI. Studying perceived barriers and investigating why these barriers differ across parties in a network, contributes to the sparse academic literature that attempts to connect PI to practice and incentivize its implementation. In particular, an investigation is made into the perceived barriers to IT adoption and while such barriers have been extensively studied with respect to different settings, there is not much literature available on the barriers to IT adoption on a network level. Therefore, a contribution is also made to IT adoption literature in a more general sense.

The paper is structured as follows. Section 2 provides an overview of prior research on PI, port networks and IT adoption. Consecutively, the methodology for data gathering and analysis will be discussed in Section 3, together with a detailed rendition of the focal hinterland network. Section 4 discusses the emergent barriers to IT adoption. In Section 5, these findings are compared and contrasted to current literature and some observations are proposed. The paper ends with a discussion and conclusion.

2. Theoretical background

2.1 Physical Internet

(10)

9 funding programs. The need for validation research, comprising of case studies, pilots and labs, has also been stressed by ETP-ALICE, which is a key lobby group for technological innovation in logistics and SCM in Europe, and has a particular focus on the PI implementation agenda (ALICE, 2014). However, research on implementation of PI is still scarce, as only three publications briefly touch upon this subject (Ballot et al., 2014; Cimon, 2014; Fazili, Venkatadri, Cyrus, & Tajbakhsh, 2017). While all three publications discuss relevant issues with respect to PI implementation, neither of them is empirically grounded or verified.

The PI network is envisioned to optimize three types of flows: physical, digital and financial flows. Streams of physical objects, information and money flow through a network governed by Internet of Things (IoT) principles (Gubbi, Buyya, Marusic, & Palaniswami, 2013). The IoT has been defined as ‘a dynamic global network infrastructure with self-configuring capabilities based on standard and interoperable communication protocols’ (Vermesan & Friess, 2011, p. 10). The capabilities in this network infrastructure communicate about their identity, status, location or any other type of relevant information in real-time (Uckelmann, Harrison, & Michahelles, 2011). Technologies constituting IoT networks are, for example, Radio Frequency Identification (RFID) and General Positioning System (GPS), and tools exploiting both technologies could be smart sensors or tags. These tools should get ever smarter by being connected to the network and as a result of the increasing amount of data they process. For instance, π-sensors could indicate the status of fixed assets such are cranes or warehouses, and at the same time register passing vessels or containers, and learn to report this more accurately every time. The π-tags are connected to a unique container and function as the representing agent of that container (Montreuil, 2011). These π-technologies and π-tools all enable the flow of goods, information and money through the network.

2.2 Physical Internet in port hinterland networks

PI is a particularly interesting innovation to study in light of sea ports and their networks. Especially since there are a lot of capabilities and entities that could communicate to a network infrastructure based on IoT, such as containers, cranes, vehicles and terminals (Montreuil, 2011). Regardless, PI research has not yet narrowed down to a maritime setting. This might be due to the fact that a major efficiency increase has already taken place in this sector when ocean containers were standardised in the form of twenty feet equivalent units (TEU’s), and research started focusing elsewhere (Lee & Song, 2017).

(11)

10 than 90 percent of trade (Mangan, Lalwani, & Fynes, 2008) and this amount is likely to even further increase for world-class ports, as shipping lines are building ever bigger vessels that can only be berthed at a few ports globally, resulting in a concentration of cargo in these places. Secondly, even though network-level IT is not yet applicable, ports are well-known settings where processes and performances are greatly affected by their IT infrastructures (Ferreti & Schiavone, 2016; Gordon, Lee, & Lucas, 2005; Mangan et al., 2008). Lastly, in logistics networks it is key for firms to find efficient ways to move and store flows of goods, money and information in order to retain its competitive position. This will become ever more crucial with the rise of global shipping, which will result in an influx of all three flows. Due to the resulting increase in complexity at port networks, expressed in congestion issues throughout the ports and their hinterlands, this industry requires continuous adoption of innovative IT (De Martino et al., 2013).

Ports are often mentioned in PI studies as a multi-modal hub in a global logistics web. The function of such a multi-modal hub is to receive, sort, tranship, consolidate and ship both inbound and outbound π-containers (Crainic & Montreuil, 2016). Currently, ports are complex networks in itself, captivating multiple modalities and entities in one area, many of these flowing to and from the hinterland. As this is getting increasingly complex, the notion of IT, and intelligent ports in particular, became a very popular topic in academic debates about port management (Ferreti & Schiavone, 2016; Wu, Xiong, Gang, & Nyberg, 2013). Intelligent ports are defined as ‘a service system for port transportation based on modern electronic IT, whose features are to provide multifarious information services for port participants based on the collection, processing, release, exchange, analysis and usage of relevant information’ (Dong, Gang, Li, Guo, & Lv, 2013, p. 292). Notably, this definition shows a close resemblance to the definition of ports in multi-modal hubs in PI. Intelligent ports also mainly rely on IoT tools to interconnect all entities in the network. For example, Siror, Huanye & Dong (2011) have performed an empirical study focused on RFID to analyse the benefits of an intelligent port. This is a major start to PI literature, but the authors emphasize the need to also analyse the cooperation with the various participants of the entire supply chain of both maritime logistics and land logistics (Dong et al., 2013).

(12)

11

2.3 IT adoption

IT adoption is a concept that has been studied for decades in various fields and thus has a very solid empirical basis (Oliveira & Martins, 2011). One of the major frameworks contributing to the development of this literature stream was the Technology, Organisation and Environment (TOE) framework by Tornatzky & Fleischer (1990) (Figure I). Other established theories such as the Theory of Planned Behaviour (Ajzen, 1991) and the Technology Acceptance Model (Davis, 1989) were not suitable for this particular study because they only consider an individual’s choice (Oliveira, Thomas, & Espadanal, 2014). The TOE framework, on the contrary, is regarded a very suitable framework for studying a firm’s entire context, including the environment and its influence on the adoption process.

Technology context entails all technologies relevant to the firm, including those in use as well as the

technologies or applications existing in the market. Organisation context describes a firm in terms of a number of factors: the firm size, its formal linking structures (centralized or decentralized), the communication processes (formal or informal) and its available resources (human and slack).

Environment context describes the playing field where the firm conducts its business, including all

relevant stakeholders such as competitors and governments.

Figure I – The original TOE framework by Tornatzky & Fleischer (1990)

(13)

12 not that barriers must be overcome in order to progress to adoption (Cole, 2014). Many scholars have studied the facilitators and barriers to IT adoption of a specific technology, using the TOE framework as a base structure for arranging their results. Hence, multiple factors have been added or removed from the original TOE framework. A short overview of the literature per dimension follows.

Technology Both the complexity and compatibility of an innovation have been found to be major

determinants of adoption (Harris, Wang, & Wang, 2015; Thong, 1999; Wamba et al., 2016), especially concerning data processing and standard data exchange (Simmer, Pfoser, Grabner, Schauer, & Putz, 2017). An additional barrier to adoption has been proven to be the perceived risks of adopting new IT. These could be risks such as confidentiality of information and security issues (Harris et al., 2015; Kannabiran & Dharmalingam, 2012; Miorandi, Sicari, De Pellegrini, & Chlamtac, 2012), standardisation and integration issues, and unresolved technical issues (Oliveira & Martins, 2011).

Organisation One factor of the original TOE framework has been confirmed in almost every study to

enhance adoption, and that is the size of firms. Small firms will not quickly adopt IT contrary to medium and large firms. This might be partly due to a lack of financial resources of smaller enterprises (Kannabiran & Dharmalingam, 2012) or to the fact that larger firms will have more knowledgeable employees (Abdollahzadehgan, Gohary, Hussin, & Amini, 2013; Low, Chen, & Wu, 2011; Neeley, 2006; Thong, 1999). At the same time, the lack of knowledge is an often found barrier to adopting innovations (Kim & Garrison, 2010). Sea ports in particular were found to have difficulty in exploiting its knowledge of IT for the existing corporate strategy (Ferreti & Schiavone, 2016). Certain traits of employees and decision-makers such as reluctance to change can also withhold a company from innovating (Harris et al., 2015).

Environment Complementary to the four environment factors mentioned by Tornatzky & Fleischer

(1990), who studied IT adoption in a general sense, there are some industry-specific additions to the model. For example, the lack of trading partner readiness was identified as a barrier to e-businesses in Europe, which makes sense in the particular industry as IT is a critical success factor for those firms (Zhu, Kraemer, & Xu, 2003). In multi-modal transport, policy-related barriers were studied in depth by Harris et al. (2015) and involve policies concerned with customs, safety standards and cargo procedures. Lastly, Ferreti & Schiavone (2016) identified immaturity of industry standards as an inhibitor to IT adoption in maritime settings.

(14)

13

3. Methodology

3.1 Research design

In line with the purpose of this study, we conducted qualitative research. Qualitative research is most appropriate when holistic and meaningful characteristics of real-life situations and context are highly relevant to the research outcomes (Yin, 1994). As container logistics in port hinterland networks are indeed complex and contingent upon many contextual factors, qualitative research suits this study best. The unit of analysis is port hinterland networks in this single case study.

Single case studies are valuable as they lend the opportunity to study a phenomenon in great depth (Voss, Tsikriktsis, & Frohlich, 2002) and to study a unique case (Yin, 1994). This study was performed in the hinterland network of the Port of Rotterdam, ranging from the port area to its inland customers in Europe. The Rotterdam hinterland network is a very large network, comprising hundreds of companies in its scope. Studying such complex networks in-depth is therefore particularly valuable. The Port of Rotterdam is unique in the sense that it is among the largest ports in the world and is therefore a major gateway into the European hinterland. Moreover, the infrastructure for container transportation in the hinterland network is particularly well-developed (World Bank, 2016) , which allows for a smooth flow of containers to and from hinterland parties. The Port of Rotterdam also has developed a distinctive focus on IT over the past years (Port of Rotterdam, 2011). All these factors constitute a developed network where there is a solid basis for installing innovative technology, therefore making it a very useful setting for this single case study.

3.2 The Port of Rotterdam hinterland network

The Port of Rotterdam is the largest port in Europe in terms of containers handled per year (13,7 TEU in 2017) and is therefore the main gateway of European hinterland transportation in containers (Port of Rotterdam, 2017). The port used to be the biggest port of the world until 2003, when Asian ports initiated a substantial growth. Therefore, over the past 15 years, the focus of the Rotterdam Port has shifted towards a more innovative and sustainable growth rather than just an increase in the movement of goods and containers. This vision entails a strong focus on data and IT to add value (Port of Rotterdam, 2011).

(15)

14 terminals spread throughout the Netherlands. The majority of ILT operators take on the role of barge and/or train operator as well. For the sake of clarity, in this study we consider all these parties under the nominator of carrier. A consignee is the company that owns the freight and is usually the destination when the freight is imported or the start when the freight is exported. There are approximately 30.000 consignees in the Netherlands alone. The port authority is the owner of the port area and acts as a governor of the tenants and the exploited areas. A forwarder could be involved as an intermediary between consignees and carriers. IT suppliers might be involved for providing the technical infrastructure within or between companies. Lastly, overarching parties such as the government and particular associations help the parties in this network in pursuing their interests. Port authorities, governments, IT suppliers and associations are, however, not involved in the operational activities of container transportation.

Figure II - Hinterland transportation network

Besides the large number of companies involved, one factor that makes these networks even more complex is the (lack of) contractual relationships between these parties. For example, a DST only has a contractual relationship with, and thus obligations to, SL’s. As a result, SL’s will always be prioritized over carriers in handling of vessels. This also affects flows of information, as some parties are bound to communicate, but do not have a contractual relationship. We kindly refer to Appendix A for a visualisation of the network structure.

(16)

15 systems include a Barge Management System (BMS), Terminal Operating System (TOS) and Materials Resource Planning (MRP). Network and telecommunication technologies are for example web portals, EDI or API connections, e-mail or phone calls. Lastly, data that flows through the network contains for instance container identification numbers, the Estimated Time of Arrival (ETA), or Estimated Time of Departure (ETD). Additionally, the Port of Rotterdam is, as part of its core strategy, focused on developing applications and tools that will integrate information flows in the network. An integrated barge planning tool and an application aimed at optimizing the planning of deep sea vessels are two examples of efforts attempting to smoothen the flows of vessels and information within the port.

3.3 Data collection

This case study relied on mainly primary sources of data, including (a) semi-structured interviews with informants, (b) electronic interviews filled out by informants who we were unable to schedule an interview with, and (c) investigatory discussions with employees in strategic business units at the Port of Rotterdam. Additionally, some secondary sources of data were used, such as company yearly reports, executive summaries and strategic roadmaps for the sake of triangulation (Voss et al., 2002). The data collection period spanned approximately 4 months from March 2018 until June 2018. Overall, 18 interviews were conducted, 3 electronic interviews were completed and 3 investigatory discussions were held for the sake of understanding market dynamics and strategic objectives.

Informants were selected based on multiple criteria (Table I). Firstly, all informants should in some way be involved in logistics, planning, operations or transportation in a hinterland network. Secondly, all parties in this network had to be involved in the data collection in order to establish a comprehensive understanding of the entire network. We did, however, not interview any SL’s, due to the fact that they are not actively involved in hinterland transportation as their core interest lies in deep sea shipping. Thirdly, we wanted to prevent biases due to collaboration in corridors or specific geographic areas. Therefore we ensured data input from informants in different regions and we classified these regions in terms of provinces. Lastly, we included some informants from government initiatives, associations and IT suppliers to enhance our understanding of the market dynamics, different interests and opportunities or challenges in the sector. These informants had a thorough understanding of the problems in the sector and could therefore strengthen our own understanding. As the purpose of a single case study is to provide the researcher with in-depth knowledge, we aimed to interview as many informants as possible in the time span available, thereby enhancing the richness of our data.

(17)

16 companies have in use and the information that was, or was not, available to them. We were also interested in the way of communicating this information internally and externally throughout the network. Follow-up questions would go more into detail on the issues that companies encountered with respect to IT, information, communication and collaboration in the network. The interview duration varied greatly, dependent on input from the informants and the number of follow-up questions, but lasted on average 50 minutes. All interviews were fully recorded and transcribed.

Additionally, the results were validated by an expert in the carrier industry. As we acknowledge the drawbacks of a single case study, these being the reliability and generalizability of results (Voss et al., 2002), validation of results contributed to the construct validity of this research. The results of this validation effort can be found in Appendix C.

Table I – Primary data collection specifics

Company Geographic

region (province)

Position informant Length of

interview (min)

1 Initiative Noord-Holland Project manager 57

2 Initiative Utrecht Advisor shipping 24

3 DST Zuid-Holland Manager PR & communication 53 4 Barge/truck operator, SL Groningen Deputy director 56 5 Barge/truck operator, ILT Drenthe ICT Manager 59 6 Barge/truck operator, ILT Gelderland Deputy director 34 7 Barge operator, ILT Zuid-Holland Operational manager 48 8 Barge/rail operator Zuid-Holland Managing director 108

9 ILT Flevoland Operational manager E-mail

10 Consignee Zuid-Holland Manager projects customer service & logistics

42

11 Consignee Zeeland Operational manager 23

12 Consignee Groningen Logistical engineer 73 13 Consignee Utrecht Head logistical planning 53 14 Forwarder Groningen Founder & owner E-mail

15 Forwarder Brabant Forwarder deep sea 61

16 Forwarder Gelderland Director E-mail

17 Planning tool Zuid-Holland Managing director 39 18 Port community system Zuid-Holland Manager strategy & innovation 46 19 IT supplier Zuid-Holland Product manager 39 20 IT supplier Zuid-Holland Founder & co-owner 44 21 Association Zuid-Holland Project manager SCM 53 A Port authority Zuid-Holland Corporate strategist 50 B Port authority Zuid-Holland Senior business manager

logistics

(18)

17

3.4 Data analysis

The interview data was analysed according to qualitative data analysis procedures and in line with Strauss & Corbin (1990) by performing a within-case analysis (Eisenhardt, 1989). The initial stage of data analysis was aimed at generating an enhanced understanding of what barriers there are to IT adoption in this particular network. The analysis was performed using an open-coding approach for all interview transcripts, thereby adhering closely to the statements of the informants whenever possible (Gioia, Corley, & Hamilton, 2013; Strauss & Corbin, 1990). In the end, we analysed 1107 quotes and coded them as 530 first-order codes. The second-order analyses involved axial coding, seeking similarities and differences among the many first-order codes. In this process, we tried to retain the informant terms in order to stay close to the data. We consolidated the 530 first-order codes into 52 second-order themes, displaying the dynamics between the codes. Using the constant comparative method, we continuously cycled between data, themes and relevant literature in order to unveil the major themes of interest and the relationships between them. Finally, we aggregated 12 of these 52 second-order themes into the 3 dimensions of the TOE framework. Of the remaining second-second-order themes, 27 were descriptive themes for enhancing our understanding, but not displaying any barriers. The other 13 were used for additional analyses, as described below. Because of the amount of information, we particularly focused on the concepts that had limited referents in the existing literature or that “leaped out” because of their relevance (Gioia et al., 2013). The process of data analysis was not a linear process but rather a dynamic, iterative process that continued until we had a clear grasp of the depth of this case and its within-case relationships. The initial findings from this analysis are presented in Section 4.

(19)

18

4. Emergent barriers to IT adoption

In order to investigate the emergent barriers to adoption of new and innovative IT, it is worthwhile to first investigate the current status of IT infrastructure in the logistics sector. Therefore, data was gathered on parties’ current IT resources, comprising of the operating systems and the network & telecommunication technology (Table II).

Table II – Current IT infrastructure

Operating systems Network & telecommunication technology

Terminal Operating System (TOS) Content Management System (CMS) Port Community System (PCS) Barge Management System (BMS) Transport Management System (TMS) Enterprise Resource Planning (ERP) Materials Resource Planning (MRP) Custom-built in-house systems

API EDI Web portal

E-mail (including Word/Excel/PDF files) Mail

Phone Intranet Applications

(20)

19

(21)

20 Figure III on the previous page shows the data structure for our case study results. It depicts the three aggregate dimensions of the TOE framework that emerged from the initial analysis, and the preceding second-order themes and first-order codes that constitute these aggregate dimensions. These results indicate that there are 12 barriers to IT adoption in port hinterland networks, categorized in 3 dimensions. Based on this data structure, we propose an overview of all adoption barriers according to the TOE-framework (Figure IV). Whereas the figure suggests that the division between barriers is static rather than dynamic, the analysis was not as straightforward as putting themes into the most obvious dimension. There is a certain interplay between these barriers, drawing the dimensions close together. For example, characteristics of makers are inherent to market characteristics as decision-makers influence the dynamics of the market. For the sake of clarity, we discuss each of the three dimensions separately, while at the same time acknowledging their interrelatedness.

Figure IV – Emergent barriers to IT adoption

4.1 Technology barriers

(22)

21

Perceived risks of IT It seems that there is a reluctance in reliance on IT and software. First and

foremost, informants are worried about privacy and security issues. As one informant put it: ‘within the shipping industry, there is some fear about the subject of privacy, and at the same time how one could break into your security’ (advisor shipping, initiative). Another perceived risk is that of IT failure and the impact it would have on operations. A disruption due to IT failure already happens occasionally. Parties are especially holding back because they value the human factor in container logistics, and are afraid that this will disappear as a result of automation. As one informant put it, ‘the know-how will disappear, the automation will take over but if automation fails, there is no back-up. That is a major risk for us’ (deputy director, carrier).

Complexity of IT Complexity of IT is another barrier for some parties. A carrier, for example, is involved with many other parties in their daily operations. Even when improving the internal IT system, many informants indicated to have issues with the complexity of aligning all necessary information flows in their in-house system. Very often, companies have multiple software systems tied to each other in back-office operations. An additional complexity of IT projects lies in the fact that IT is not a priority for everyone: ‘those are definitely barriers for carriers. They just want it simple and not too complex, and well if you have to work in another application and it does not work again… Well they are carriers, they just want to transport from A to B’ (product manager, IT supplier). The complexity of IT is closely linked to the technological competences within a firm, underscoring the interconnectedness of themes and dimensions.

Compatibility of IT Another technological barrier seems to be the current organisation of information. There is no standardised data format for communicating container numbers, ETA’s, ETD’s, and other information. Instead, the level of digitisation ranges from communication using e-mail, phone calls and Excel files to dedicated API-connections. Some informants even reported on the use of handwritten notes or receiving container numbers in the mailbox. Incompatibility also has to do with the level of specialisation at partnering firms in a specific corridor. If there are large discrepancies in strategic focus, this could lead to development of different systems: ‘every terminal (ILT, red.) has its own specialty, for example company A is focused on handling short sea vessels. Not barges. So they develop software focused on short sea’ (ICT manager, carrier). Information flows and the level of digitisation in the industry are simply incompatible, which calls for increasing innovation but is, up to that point, unfortunately still a barrier.

Transparency Transparency of information is a technological barrier as it is related to the processing

(23)

22 information’ is involved with information ambiguity. Additionally, information ambiguity is a perfect disguise of errors in the process – whether they are deliberate or not: ‘at the moment one party makes a mistake, it is not in his interest to show that information as transparent as possible’ (project manager SCM, association). The more the network is integrating its information flows and making things transparent, the more the future of some companies is pressured.

4.2 Organisation barriers

Next to issues related to the technical details of an IT system, many barriers were found to be of an organisational nature, resulting in the following four barriers: firm size, financial resources, characteristics of the decision-maker and technological competence. Overall, the case study findings indicate that organisational barriers pose a prominent problem in IT adoption, having to do mostly with firm size and the resulting financial capacity.

Firm size The size of firms, mainly in terms of TEU’s carried per year, can act as a barrier to

especially the ability to innovate or collaborate. However, there is no consensus among informants about the impact of firm size. One informant identified 3 types of parties: ‘there are different sorts of parties. The large companies might be so big that they do not need to be involved, or they are medium-sized and they have enough volume to be open with a number of companies. And then there are the really small ones which are willing to, but do not understand or they are afraid to lose their loads’ (product manager, IT supplier). On the other hand, ‘small parties can be willing to innovate because they are visionary and can make very swift decisions’, another informant said (project manager, initiative). One effect of firm size where informants agree is the fact that large companies are the right companies to initiate use of a new technology.

(24)

23 see if there is a way to spend some money on for example sustainable transportation. So we thought, we will optimize a certain pathway in our network by carrying two containers per truck instead of one. The money we saved off that, we don’t just sit on it, but we put it in an investment on renewable fuels’ (manager projects customer service & logistics, consignee). This example illustrates that some companies are able to find ways to innovate regardless of the fact that they have little to no financial reserves for innovation.

Characteristics of the decision-maker Characteristics of the decision-maker can be a barrier to IT

adoption, as there are many older, more conservative decision-makers in this sector and the mindset is still rather traditional. Hence, particularly age and mindset can pose as an inhibitor to IT adoption: ‘many decision-makers in this sector are, with all due respect, above 50. […] They barely know how to handle an iPad. That may sound disrespectful, but how then will you ever enunciate the need to innovate?’ (ICT manager, carrier). One interviewee stated that, “especially in barging, many older decision-makers are tempted to make a decision in the business case, whereas younger decision-makers have a more entrepreneurial mindset and consider their business position in relation to their competitors. Moreover, they are also less afraid of the privacy aspect, because they think: ‘I have my mobile phone right here, everyone can track me already whether I join this initiative or not. And the older generation is just holding back in that perspective’ (advisor shipping, initiative). As a result, decision-makers significantly slow down change processes in this sector.

Technological competences There are large differences in technological competences. Similar to the age and mindset of the employees in this sector, technological know-how is also not yet very well-developed. Mainly interesting is that there is a clear role for software suppliers: ‘we are dependent on our software suppliers for innovations. And that is crucial’ (deputy director, carrier). Quite a lot of informants indicated to have a ‘very disappointing’ level of expertise in-house, or even that they ‘simply do not know how to share’. A few companies contrast this finding because they do have in-house specialists with significant IT knowledge.

4.3 Environment barriers

Lastly, many industry-specific barriers were found to be of importance when considering IT adoption. Four barriers were found to be related to the environment of an organisation: the market characteristics and the related competitive pressure, power imbalance and policies and regulations. All in all, the hinterland network is a very tumultuous, difficult environment and above all an environment that is coping with many other problems than the slow IT development.

Market characteristics Market characteristics are definitely a barrier to IT adoption as the logistics

(25)

24 quicker moving customer sector, and more importantly, the international shipping industry. Interviewees indicate that there is a long established culture that is not very easily changed: ‘many just do the same and are used to doing that in the same pattern, in the same way with the same culture’ (founder & co-owner, IT supplier), and the market is said to be very traditional with a likewise mindset. ‘If you have done the same thing for 30 years, why would you suddenly go about it differently?’ is an often heard statement. Lastly, many companies indicated to only join IT initiatives or projects when it is clear to them that more parties in the network will join this, or that they will only adopt new IT when it has turned out to be successful, i.e. the results indicate effects also known as network externalities.

Competitive pressure Additional to features that are inherent to firms and people in this particular

sector, there are certain dynamics in the competitive environment that make adoption of IT very difficult. Firstly, competitive forces hinder the willingness and the easiness to share data, because competitors could profit from this: ‘you could share some information, but it depends on who you share it with. You will naturally not share sensitive data with your competitors’ (deputy director, carrier). Secondly, because the competition among particularly carriers is very intense. This has an effect on their profit margins, as illustrated earlier, but also on the response of competitors to errors or sensitive information. ‘The problem is that competitors are always lurking, trying to deduct where a container came from and for which consignee it was. And then one can hunt for your loads’ (deputy director, carrier). Lastly, a striking finding with respect to competitive contexts was that informants experience a proliferation of ILT’s and regard that as a factor that adds fuel to fire in the carrier industry. ‘Over the past year, a proliferation of terminals came into existence, of which each has its own service line to the Port of Rotterdam’ (manager PR & communication, DST). Naturally, this contributes negatively to the existing competitive environment and the congestion issues in the port.

Power imbalance Adding to the already heavy competition, there are also a lot of power discrepancies in this sector, where powerful companies force others to conform to their rules. This can lead to problems: ‘a shipping line makes up its own rules and you will just have to deal with that. Except some rules they make up are simply not executable’ (deputy director, carrier). In particular, there is a zero tolerance policy with respect to data and information provision to DST’s, where container information must be 100% correct in order for those containers to be shipped. Such things displays a genuine fragmentation of power in this particular market and seem to cause large biases in the competitive environment.

Policy and regulation Policy and regulation are a last barrier to IT adoption as they require from

(26)

25 weights (the Verified Gross Mass) as part of the full documentation that is required for a load booking at a shipping line. Also, the recently introduced privacy regulations in Europe cause many parties to be reserved in taking IT to the next level. Many companies indicated that such regulations, as they are compulsory and have a direct impact on operations, get priority over things such as IT adoption.

5. Interpretation of results

In port hinterland networks, all parties have different interests, capabilities, geographical locations and resources and they are very much interconnected with one another. This section will elaborate upon the most apparent and relevant differences among five different parties who are operationally involved in the movement of goods, information and money concerning container logistics: SL’s, DST’s, carriers, forwarders and consignees. In the analysis, we filtered per party the barriers identified in Section 4. We distinguish between two user groups: SL’s and DST’s are in the seaside user group and carriers, forwarders and consignees are in the hinterland user group. Table III lists for each of the five parties how they experience each barrier, insofar data was available. Accordingly, we will reflect on the outcomes, juxtapose these findings to existing literature and develop some general observations.

Table III – Relation of each party to TOE factors

Technology Organisation Environment

SL Advanced IT

Transparency is a driver

Largest in size

Sufficient resources for investing

In-house competences

Intense competition among SL’s Local hinterland not important High in power & authority Focus on international regulations

DST Advanced IT

Moderate in size

Able to invest if necessary In-house competences Young, innovative people

Little competition among DST‘s Get agent power from SL’s Have to deal with both seaside and

landside policies

CAR Large differences in IT

Transparency is a barrier when competitors are involved

Large differences in size Low margins, little resources No in-house competences Large differences in people

Intense competition among carriers Differences in power; overall, very

little and linked to load sizes Have to deal with port policies

FOR Legacy IT

Transparency is a major threat to business model

Large differences in size Low margins, no resources No in-house competences Older, conservative people

Intense competition among forwarders and with platforms Differences in power; overall, very

little and linked to load sizes Have to deal with port policies

CON Large differences in IT

Large differences in size Large differences in resources Large differences in

competences

Large differences in people

Competition limited to direct competitors

Moderate power; customer

(27)

26

5.1 Seaside user group

SL’s and DST’s are in the seaside user group because their business is partly concerned with deep sea shipping, in contrast to the other parties in the analysis. On a technology level, this user group already has advanced IT systems and thrives on transparency of information because to them, information is knowledge and allows them to optimize their performance. Particularly DST’s are served by transparency, because if they are transparent in their information, the carriers coming to the port can adapt to this information. Therefore, DST’s have begun to pay more attention to optimizing the entire supply chain next to optimizing their internal operations, particularly with respect to communication (van der Horst & van der Lugt, 2011). ‘That has everything to do with the fact that most DST’s are fully automated and they are served by a solid data exchange’ (manager strategy & innovation, Port Community System). This solid data exchange is essential not only to enhance DST’s internal operations, but also to ensure compatibility with all stakeholders of the DST. Therefore, they are not in an intensely competitive environment but are instead stimulated by the Port Authority and their own strategic drivers to continuously innovate.

Technology competence constitutes both physical infrastructure and intangible knowledge (Zhu et al., 2003). SL’s are enormous multinational corporations with equally large IT systems and for that, they need sufficient in-house specialists for managing the complex and sizable IT resources. DST’s too, require in-house competences in order to respond quickly to any disruptions on the terminal. Indeed, an informant explained: ‘we are basically an IT company that also overhauls containers’ (manager PR & communication). Moreover, innovation is a driver rather than a barrier for both parties as part of their competitive position (De Martino et al., 2013). DST’s business models are centred on the swift handling of sea vessels and consecutively ensuring also the outflow of containers from the terminal. As DST’s are to a large extent automated, it is particularly interesting to continuously invest in optimisation of these IT-facilitated processes.

(28)

27 All in all, the companies in this user group already possess sophisticated IT systems that set the bar for the rest of the network to respond to. Next to that, their position in the competitive environment of ports stimulates innovation rather than limiting it. Therefore, we observe:

Observation 1: Seaside parties are less likely to experience any barriers to IT adoption due to an already advanced level of IT infrastructure, their firm size and capabilities and their position in the port hinterland competitive environment.

5.2 Hinterland user group

Carriers, forwarders and consignees are in the hinterland user group as their core business reaches from the port to inland customers. With respect to technology, there are large differences in the organisation and the maturity of systems among carriers, forwarders and consignees: ‘it runs from top of the bill to the Stone Age’ (ICT manager, carrier). In multimodal transport, low compatibility among IT applications serves as a barrier to interconnectivity (Harris et al., 2015). Whereas some carriers are a front-runner with respect to compatibility with a dedicated connection to the database of DST’s, other companies still employ multiple people to browse through DST websites all day to track their containers. However, as contradictory as it may seem, these parties are not always helped by information transparency. Because the competition in the carrier and forwarding industry is intense, every piece of information is knowledge. More specifically, a forwarders’ business model entails the optimal coordination of various parties using information about prices, capacity and container location. If such information is made increasingly transparent, forwarders might be put out of business. Consignees are different in this sense, because they are very interested in transparency of information and therefore also perceive less risks in IT adoption. However, consignee priorities are often focused on more pressing matters that are vital to operational processes such as adapting their systems to privacy or documentation regulations, which compromises the their technological advancement. Therefore:

Observation 2: Hinterland parties are more likely to experience barriers to IT adoption due to a fragmentation of IT capabilities in the hinterland network.

(29)

28 be risk-averse and maintain a short-term focus when it concerns finance (Harris et al., 2015; van der Horst & de Langen, 2008), causing a reluctance to invest in IT that will not generate (almost) immediate profits. Almost all hinterland companies rely on a third party software supplier to arrange their IT systems. They have no IT expertise in-house, ‘except for a friend of a friend who fixes a bug every once in a while’ (forwarder deep sea, forwarder). Additionally, firm size is often a proxy for human resources and the organisation of IT (Kannabiran & Dharmalingam, 2012). Indeed, decision-makers in particularly smaller firms in the hinterland network are mainly above 50 years old and rather conservative as they have been in the business for a long time. They have a limited knowledge and understanding of IT and innovations, and are not quickly urged to investigate this. Particularly carriers are experiencing difficulty in employing qualified personnel and have a shortage of IT skills that are needed to innovate (Harris et al., 2015). Consignees in general have more resources and competences available, dependent on the firm size and innovativeness. Consignees are not concerned with the actual transportation of containers, but their main interest is that their freight boards the right vessel on time. For that, they need to provide an increasing amount of information to either carriers or forwards, creating a dyadic link where goods, information and money are exchanged. This calls for enhanced interconnectedness in this part of the network. However, current logistics networks remain fragmented due to the focus on one single consignee or corridor (Montreuil, Ballot, & Fontane, 2012). All together, we propose the following observation:

Observation 3: Hinterland parties are more likely to experience barriers to IT adoption due to a fragmentation of organisational capacity and capability in the hinterland network.

(30)

29 appoint fixed slots. As a result, smaller carriers do not get fixed slots, but instead have to run the risk of having to wait on the terminal before they can load or unload their containers. The consignees have a different role in this market because they are the end customer of this logistics process. Therefore, they do not experience any barriers due to competition or market forces in the container transportation sector. In their direct competitive environment, this might be different. However, consignees do experience many struggles with changing policies and regulations, for example with changed documentation rules with respect to container shipment or new privacy regulations. Regulation in EU port networks is in fact increasing with respect to safety, security and environmental issues (Psaraftis, 2005), making it difficult for all consignees to keep up with specific regulations. After all, the container logistics sector is usually not their primary playing field. In sum:

Observation 4: Market characteristics, competitive pressure and power imbalance cause carriers and forwarders to compete rather than collaborate, thereby reinforcing the fragmentation of hinterland networks.

6. Discussion and conclusion

PI is envisioned to be implemented by 2050 (ALICE, 2014). For this to be possible, networks and IT need to be fully interconnected in order to ensure a smooth flow of goods, information and money. This paper has addressed this need by investigating the barriers to IT adoption in a network perspective. We take stock of the findings that emerged in Section 4 and interpretations discussed in Section 5, and use these to revisit our research question, comprising of two parts. The initial part, which aimed to answer what barriers there are to IT adoption in port hinterland networks, resulted in 12 barriers to IT adoption which are prevalent with respect to technology characteristics, organisational features and the competitive environment of a network. Many barriers that were established have also been highlighted in earlier studies, especially the organisational barriers as many studies have focused on single-firm adoption (Harris et al., 2015; Kannabiran & Dharmalingam, 2012; Kim & Garrison, 2010). There are two new factors which prove to be very relevant in these specific networks, namely transparency and power imbalance. These two barriers mainly become apparent as a result of studying adoption in a network perspective, displaying the competitive context in which these hinterland parties operate.

(31)

30 willing to cooperate in order to enhance their mutual performance (Wilhelm, 2011). Particularly port networks due to their heterogeneous and dynamic nature ‘in which public and private undertakings interact for satisfying different and often conflicting interests’ (De Martino et al., 2013, p. 131). In line with this statement, our study demonstrates a fragmentation of hinterland capacity and capability, which appeared to be the main cause for companies to perceive barriers to IT adoption in port hinterland networks.

PI is functionally dependent on standardisation and interconnectedness (Montreuil, 2011). While one major advantage of container logistics is the standardized TEU container, a major effort is required in harmonizing the hinterland network so that data and information can be effortlessly shared using the envisioned standardized π-protocols. A tool for harmonisation efforts has been articulated in the form of IT (Dong et al., 2013; Harris et al., 2015; Siror et al., 2011). However, the current organisation of the network in terms of both capacity and capability inhibits current innovation efforts. In light of the above remarks, stakeholders of the port hinterland networks ought to reconsider their position in, and contribution to, the network. Load-bundling efforts are a logical solution to this end but should be carefully considered as they turn out to be very difficult to scale up and manage financially (van der Horst & de Langen, 2008). Moreover, an opportunity is laid out for government and port authorities to address the fragmentation of their hinterland networks. This is a crucial opportunity as doing so successfully will make them more attractive for SL’s and enhance the competitive position of both the port and its hinterland stakeholders.

6.1 Contributions to theory

A number of contributions were made to the existing literature on PI and IT adoption. Firstly, this paper addresses the call of several academics and practitioners for future PI research to be increasingly grounded in empirical data and solid theory, and thereby attempt to connect PI to practice (Sternberg & Norrman, 2017). This study has a solid empirical basis due to the relatively comprehensive data collection, and uses the well-established TOE framework as a theoretical groundwork. As a result, a framework of perceived barriers was proposed and takes a first step towards harmonizing and unifying a complex network where PI could be initiated. The paradigm change in network management that is needed for PI (Ballot et al., 2014) is thus more articulate as a result of this study.

(32)

31 multiple parties, using a network scope rather than a single company scope. Indeed, this different perspective provides new insights into both the applicability of the TOE framework and the interpretation of barriers to IT adoption. It appears that the TOE framework, and IT adoption models in a more general sense, are equally useful in studying networks. As a result, the framework is turned into a more dynamic framework where a barrier is not simply a barrier, but rather a scale on which different parties in the network are placed. For instance, firm size poses a well-known determinant of IT adoption (Kannabiran & Dharmalingam, 2012; Zhu et al., 2003). In networks, firm size naturally differs. Party A might be small in size as opposed to his competitor which is much larger. A certain interplay between these parties would be overlooked when studying single firms, but now this interplay becomes apparent and establishes a dynamic dimension in the framework which involves factors such as competition, cooperation, dependency or power. Hence, due to studying a network perspective, barriers such as competitive pressure and power imbalance became apparent in port hinterland networks.

6.2 Limitations and opportunities for further research

This research is not without limitations. Firstly, while we have taken measures to overcome the drawbacks of a single case study, results might not be fully generalizable or appropriate into other port networks or industries. By means of methods such as triangulation of data sources and validation of results we have tried to ensure research validity and reliability. Future studies could overcome these limitations in two directions. Either by studying IT adoption in a more general sense by collecting data from different networks in different industries. Another direction for future studies could be to investigate in a more quantifiable way the fragmentation of hinterland capacity and capability, as this seems to be the major part of port hinterland networks experiencing barriers to IT adoption. Consecutively, scholars could articulate appropriate IT infrastructures for network harmonization efforts.

(33)

32 by studying the hinterland sector in a more generalizable study, for example employing a multiple case or survey study among carriers or forwarders.

There is an opportunity for scholars in the PI research field to extend the findings of this study much further. Researchers could look into specific barriers in order to investigate solutions to overcome these barriers, for instance with respect to data governance or business model transformation. Furthermore, as PI is determined to revolutionize not only the container logistics sector, but also commercial logistics industries and possibly many more industries, future research could still explore the antecedents for IT adoption in such settings. These might differ from the antecedents in this study, as the nature and complexity might be largely different. Additionally, an opportunity lies in extending the theoretical basis of network-perspective adoption models. In a general sense, technology adoption across organisational boundaries is much more relevant nowadays as technologies increasingly span entire networks and involve numerous users. Extending the IT adoption research field in this direction would possibly also allow for new actionable insights with respect to decision-making in competitive environments and business model innovation. Since innovative companies to a much larger extent use technologies that are dependent on the number of users for its success, for instance platform business models, an adoption framework facilitating these network traits would certainly be an opportunity for research.

Notes

(34)

33

References

Abdollahzadehgan, A., Gohary, M. M., Hussin, A. R. C., & Amini, M. (2013). The organizational critical success factors for adopting cloud computing in SME’s. Journal of Information Systems

Research and Innovation, 4(1), 67–74. Retrieved from https://ssrn.com/abstract=2333028

Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human Decision

Processes, 50(2), 179–211. https://doi.org/10.1016/0749-5978(91)90020-T

ALICE. (2014). Information systems for interconnected logistics roadmap. ALICE. Retrieved from http://www.etp-logistics.eu/alice/en/publications/

Baker, J. (2012). The Technology-Organization-Environment framework. In Information systems

theory: explaining and predicting our digital society (Vol. 1, pp. 231–245). New York, NY:

Springer. https://doi.org/10.1007/978-1-4419-9707-4

Ballot, E., Montreuil, B., & Meller, R. D. (2014). The Physical internet: the network of logistics

networks. Paris, France: La Documentation Française.

Byrd, T. A., & Turner, D. E. (2001). An exploratory examination of the relationship between flexible IT infrastructure and competitive advantage. Information and Management, 39(1), 41–52. https://doi.org/10.1016/S0378-7206(01)00078-7

Cimon, Y. (2014). Implementing the physical internet real-world interface: beyond business models, the devil is in the details. In Proceedings of 1st International Physical Internet Conference (pp. 1–9). Québec City, Canada.

Cole, R. (2014). Tracing the origin: the use of blockchain in supply chain provenance.

Crainic, T. G., & Montreuil, B. (2016). Physical internet enabled hyperconnected city logistics. In

Transportation Research Procedia (Vol. 12, pp. 383–398).

https://doi.org/10.1016/j.trpro.2016.02.074

Davies, I., Mason, R., & Lalwani, C. (2007). Assessing the impact of ICT on UK general haulage companies. International Journal of Production Economics, 106(1), 12–27.

https://doi.org/10.1016/j.ijpe.2006.04.007

Davis, F. (1989). Perceived usefulness, perceived ease of use, and user acceptance of information technology. MIS Quarterly, 13(3), 319–340. https://doi.org/10.2307/249008

de Búrca, S., Fynes, B., & Marshall, D. (2005). Strategic technology adoption: extending ERP across the supply chain. Journal of Enterprise Information Management, 18(4), 427–440.

https://doi.org/10.1108/17410390510609581

De Martino, M., Errichiello, L., Marasco, A., & Morvillo, A. (2013). Logistics innovation in sea ports: an inter-organizational perspective. Research in Transportation Business and Management, 8, 123–133. https://doi.org/10.1016/j.rtbm.2013.05.001

Dong, X., Gang, X., Li, Y., Guo, X., & Lv, Y. (2013). Intelligent ports based on internet of things. In

(35)

34

Informatics (pp. 292–296). Dongguan, China. https://doi.org/10.1109/SOLI.2013.6611428

Eisenhardt, K. M. (1989). Building theories from case study research. The Academy of Management

Review, 14(4), 532–550. https://doi.org/https://doi.org/10.5465/amr.1989.4308385

Fazili, M., Venkatadri, U., Cyrus, P., & Tajbakhsh, M. (2017). Physical internet, conventional and hybrid logistic systems: a routing optimisation-based comparison using the Eastern Canada road network case study. International Journal of Production Research, 55(9), 2703–2730.

https://doi.org/10.1080/00207543.2017.1285075

Ferreti, M., & Schiavone, F. (2016). Internet of things and business processes redesign in seaports: the case of Hamburg. Business Process Management Journal, 22(2), 271–284.

https://doi.org/http://dx.doi.org/10.1108/JEIM-07-2014-0077

Gioia, D. A., Corley, K. G., & Hamilton, A. L. (2013). Seeking qualitative rigor in inductive research: notes on the Gioia methodology. Organizational Research Methods, 16(1), 15–31.

https://doi.org/10.1177/1094428112452151

Gordon, J. R. M., Lee, P. M., & Lucas, H. C. (2005). A resource-based view of competitive advantage at the Port of Singapore. Journal of Strategic Information Systems, 14(1), 69–86.

https://doi.org/10.1016/j.jsis.2004.10.001

Gubbi, J., Buyya, R., Marusic, S., & Palaniswami, M. (2013). Internet of Things (IoT): a vision, architectural elements, and future directions. Future Generation Computer Systems, 29(7), 1645– 1660. https://doi.org/10.1016/j.future.2013.01.010

Harris, I., Wang, Y., & Wang, H. (2015). ICT in multimodal transport and technological trends: unleashing potential for the future. International Journal of Production Economics, 159, 88–103. https://doi.org/10.1016/j.ijpe.2014.09.005

Hearnshaw, E. J. S., & Wilson, M. M. J. (2013). A complex network approach to supply chain network theory. International Journal of Operations & Production Management, 33(4), 442– 469. https://doi.org/10.1108/01443571311307343

Kannabiran, G., & Dharmalingam, P. (2012). Enablers and inhibitors of advanced information technologies adoption by SME’s: an empirical study of auto ancillaries in India. Journal of

Enterprise Information Management, 25(2), 186–209.

https://doi.org/http://dx.doi.org/10.1108/JEIM-07-2014-0077

Kim, S., & Garrison, G. (2010). Understanding users’ behaviors regarding supply chain technology: determinants impacting the adoption and implementation of RFID technology in South Korea.

International Journal of Information Management, 30(5), 388–398.

https://doi.org/10.1016/j.ijinfomgt.2010.02.008

Kuan, K. K. Y., & Chau, P. Y. K. (2001). A perception-based model for EDI adoption in small businesses using a technology-organization-environment framework. Information and

Management, 38(8), 507–521. https://doi.org/10.1016/S0378-7206(01)00073-8

(36)

35 research opportunities. Transportation Research Part B: Methodological, 95, 442–474.

https://doi.org/10.1016/j.trb.2016.05.001

Low, C., Chen, Y., & Wu, M. (2011). Understanding the determinants of cloud computing adoption.

Industrial Management & Data Systems2, 111(7), 1006–1023.

https://doi.org/10.1108/02635571111161262

Lucassen, I. M. P. J., & Dogger, T. (2012). Synchromodality pilot study: identification of bottlenecks and possibilities for a network between Rotterdam, Moerdijk and Tilburg. Retrieved July 11, 2018, from

https://www.maasvlakte2.com/en/index/show/id/594/Master+Plan+for+hinterland+transport Mangan, J., Lalwani, C., & Fynes, B. (2008). Port-centric logistics. The International Journal of

Logistics Management, 19(1), 29–41. https://doi.org/10.1108/09574090810872587

Miorandi, D., Sicari, S., De Pellegrini, F., & Chlamtac, I. (2012). Internet of things: vision, applications and research challenges. Ad Hoc Networks, 10(7), 1497–1516.

https://doi.org/10.1016/j.adhoc.2012.02.016

Montreuil, B. (2011). Toward a physical internet: meeting the global logistics sustainability grand challenge. Logistics Research, 3(2–3), 71–87. https://doi.org/10.1007/s12159-011-0045-x Montreuil, B., Ballot, E., & Fontane, F. (2012). An open logistics interconnection model for the

physical internet. In IFAC Proceedings Volumes (pp. 327–332). IFAC. https://doi.org/10.3182/20120523-3-RO-2023.00385

Montreuil, B., Meller, R. D., & Ballot, E. (2013). Physical internet foundations. In Proceedings of the

14th IFAC Symposium on Information Control Problems in Manufacturing (pp. 151–166).

Bucharest, Romania. https://doi.org/10.1007/978-3-642-35852-4-10

Nag, R., Corley, K. G., & Gioia, D. A. (2007). The intersection of organizational identity, knowledge, and practice: attempting strategic change via knowledge grafting. Academy of Management

Journal, 50(4), 821–847. https://doi.org/10.5465/AMJ.2007.26279173

Neeley, C. K. R. (2006). Connective technology adoption in the supply chain: the role of

organizational, interorganizational and technology-related factors. ProQuest Information and Learning company. Unpublished PhD dissertation, University of North Texas.

Oliveira, T., & Martins, M. (2011). Literature review of information technology adoption models at firm level. The Electronic Journal Information Systems Evaluation, 14(1), 110 – 121.

https://doi.org/1566 - 6379

Oliveira, T., Thomas, M., & Espadanal, M. (2014). Assessing the determinants of cloud computing adoption: an analysis of the manufacturing and services sectors. Information & Management,

51(5), 497–510. https://doi.org/10.1016/j.im.2014.03.006

Pan, S., Ballot, E., Huang, G. Q., & Montreuil, B. (2017). Physical internet and interconnected logistics services: research and applications. International Journal of Production Research,

Towards a Physical Internet: An investigation of barriers to IT adoption