A security-driven design for high value cargo transport within the Physical Internet

(1)

A security-driven design for high value cargo

transport within the Physical Internet

Master Thesis

27. January 2020

(2)

List of Tables

5.1 Interaction Description - Perform Secure Transport . . . 32

5.2 Interaction Description - Perform PI Functions . . . 35

5.3 Interaction Description - Secure PI Transport . . . 38

5.4 Interaction Description - Secure PI Hub . . . 41

5.5 Interaction Description - Secure Transport Mode . . . 44

5.6 Interaction Description - Perform PI container Security functions . . . 47

5.7 Interaction Description - Use Information . . . 48

5.8 Interaction Description - Fortify Container . . . 50

List of Figures

2.1 Perpetrator Ability and Position (Ekwall, 2009) . . . 7

3.1 DSR Framework . . . 17

4.1 BPMN of the current high value secure transport en route . . . 21

5.1 Functional Architecture . . . 28

5.2 Functional Architecture Extension Perform PI container security functions . . . 29

5.3 Perform Secure Transport . . . 30

5.4 IDEF0 Perform Secure Transport . . . 31

5.5 Perform PI Functions . . . 33

5.6 IDEF0 Perform PI Functions . . . 34

5.7 Secure PI transport . . . 36

5.8 IDEF0 Secure PI transport . . . 37

(3)

5.10 IDEF0 Secure PI Hub . . . 40

5.11 Secure Transport Mode . . . 42

5.12 IDEF0 Secure Transport Mode . . . 43

5.13 Perform PI container security functions . . . 45

5.14 IDEF0 Perform PI container security functions . . . 46

5.15 Use Information . . . 48

5.16 IDEF0 Use Information . . . 49

5.17 Fortify Container . . . 50

(4)

List of Abbreviations

CBP Customs and Border Protection

DSR Design Science Research

EM EA Europe, the Middle East and Africa

F T L Full truck load

GDP Gross Domestic Product

H Handling

IIS Incident Information Service

LT L Less than truck load

M O Modus Operandi

P Packaging

P I Physical Internet

SC Supply Chain

T Transport

T AP A Transport Asset Protection Association

(5)

Abstract

The current logistics processes for high value logistics such as gold and diamonds are closed networks op-erated by specialized companies that are cost intensive. Additionally, cargo theft is an issue in the current logistics sector that is routinely ignored by both industry as well as researchers. Even though the present literature on the Physical Internet (PI), which has been proposed as a solution to combat inefficiencies and issues in the current logistics sector through encapsulation and an open and interconnected network, includes statements that indicate that the PI is able to assist with the network security, no concrete solutions have been presented to integrate high value cargo into the PI without compromising security. A functional model as well as specific requirements have been developed in this research to allow such integration into the PI. Especially the deter function, that aims to reduce criminal opportunity of perpetrators, is a novelty in the literature and industry. Specific communication protocols between relevant stakeholders in this model com-plement the security given through deterring perpetrators and are integral to the success of this model.

(6)

Acknowledgements

This research was conducted as a Master Thesis at the Unviversity of Groningen as part of a larger research of the Physical Internet. I would like to thank all of the people that assisted during the writing of this thesis. First, I would like to thank my wife, Disha Mangsuli, for being there all throughout my masters and sup-porting and believing in me even when I did not believe in myself. Without you, none of this would have been possible.

Furthermore, I would like to sincerely thank my supervisors, Dr. N.B. Szirbik and Dr. Ir. T. Bortolotti, for assisting me with discussions and feedback during my research.

I would like to thank all of my fellow PI researchers in the research lab for the excellent discussions and feedback throughout the entirety of this research.

Next to that, I would like to thank all of the experts I talked to for taking time out of their busy schedules to validate this design and improve it with their expertise.

Last but not least, I would like to thank my friends and family for always being there and being supportive throughout this process.

With kind regards,

Constantin Gehling

(7)

1 Introduction

Over the years, high value transports have become increasingly specialized in a way that only specialized transport companies can offer customers a full security guarantee. This is a departure from previous trans-portation methods used to guarantee security of the high value cargo. The most prominent example of a high value goods shipment being transported through an open system is the case of the Cullinan diamond being transported through the regular mail while deceiving perpetrators by sending visible security escorts with a dummy shipment (Bariand et al., 1992). This trend directly coincides with the overall trend of the logistics systems becoming more specialized to the point where they fail to address major challenges regarding 21st century issues such as environmental, economic and social sustainability (Montreuil, 2011). A push has been made by scholars to develop a new logistics concept called the Physical Internet, hereafter called PI, to address such issues in the current logistics systems. The main concepts of the PI are that it encapsulates contents into PI containers and then transports them through an open and interconnected worldwide infras-tructure. In order to achieve the vision of the PI, common standards of handling containers, data, routing operations and all other factors influencing the environmental, economic and social sustainability have to be developed and validated.

One of the issues identified by Montreuil (2011) is that the current logistics networks are not secure. The current logistics system is powerless against well-coordinated and widespread attack on goods within the supply chain (SC) and therefore passes the price of these stolen goods to their customers. It is estimated that theft in the SC is a problem to the degree of 1% of the global GDP being lost to it (Boone et al., 2016). The PI gives us an opportunity to define standards and protocols with these issues in mind in order to improve the security of the logistics sector drastically. The properties, functionalities and protocols of the PI containers will be of vital importance to this process as they are the physical barrier between the products and the criminals attempting to steal it.

(11)

the specific actions that can be taken. The absence of high value transport in the PI therefore constitutes a relevant research gap that this thesis answers through the research question:

”Which requirements and needed functionalities must the PI-system cover in order to manage the security of high value goods in an effective and efficient way?”

Through the combination of existing PI literature with cargo theft literature and the addition of current industry standard, these requirements and functionalities will be investigated. Based on these insights, the functional model of high value cargo transportation within the PI will be reverse engineered and redesigned using the Icam Definition for Function Modelling (IDEF0).

(12)

2 Background

”While the stone travelled anonymously by mail, spectacular security precautions were taken to transfer an empty box from Pretoria to London. A similar strategy was used to send the stone from London to the

diamond-cutting factory of the Asscher brothers in Amsterdam” Bariand et al. (1992)

The above quote describes the security measures taken to ship the Cullinan Diamond, the worlds largest diamond, in 1907. It shows an example of how a high value item can be transported using the open and accessible system of the British postal service. This raised the question if there is a possibility to use the PI, as an open and accessible system, to securely transport high value goods.

2.1 Necessity of PI

The term PI was first mentioned in an economist article (Markillie, 2006) and suggested the similarities between the logistics sector and the digital internet in the way the products are moved through the nodes and distribution centres. Scholars and academics have taken up this topic of PI and identified massive issues within the current logistics sector. Montreuil (2011) identified 13 of these main issues:

• Shipping air and packaging • Empty travel as norm • Truckers are modern cowboys • Products mostly idle

• Many products never sold or used

• Production and storage facilities poorly used • Products don’t reach where needed most

• Unnecessary movement of products around the world • Fast and reliable inter-modal transport still a dream • Products in; through and out of cities difficult • Smart automation and technology hard to justify • Innovation is strangled

(13)

Both shipping air and packaging as well as empty travel are a massive problem in today’s logistics sector (Montreuil et al., 2012). A report from the USA shows that even when trucks are travelling loaded, they are only 60% full due to the rest being packaging and air. Additionally, on average 20% of all kilometres driven by truck in the USA in 2009 where with a completely empty container with more being nearly empty (Montreuil, 2011). Both these issues significantly increase the cost of transportation both environmentally due to increased emissions as well as economically through increased prices and incurred costs of empty travel.

Workers in the transportation industry, and specifically long-haul truckers, are modern cowboys due to the large amount of time they spent on the road. As most goods are currently transported on trucks within countries, these truckers usually spent long stretches of time on the road without seeing their family. These working conditions place immense physical, emotional and social stress on these truckers and can lead to serious mental health problems, safety concerns on the road as well as a decrease of productivity (Apostolopoulos et al., 2016). These long-haul trips are proven to result in increased sleep deprivation, which accounts for 58% of accidents in the US (Hege et al., 2016; Montreuil, 2011).

Idle products cause huge economic stress on companies as they signify costs tied up in inventory. A number of companies in the fashion retails industry for example struggle with a rise in their inventories due to unsold items (Green, 2018; Hanbury, 2018), while a number of companies had to turn to alternative storing methods such as mobile storage in containers (Phillips, 2018). Even with many companies adopting Lean principles and Just-in-Time inventory management, the amount of inventory compared to sales has been stable for a number of years (U.S. Census Bureau, 2019). On the other hand, not having inventories available for online sales is costing the online consumer goods and retail industry in excess of $22 billion (Corsten and Gruen, 2018).

As the current way of working in the logistics sector normally means production and storage requirements vary seasonally (Gu et al., 2010). This automatically leads to issues in planning the warehouse and production facilities. They will either be underutilized for most of the year or they will exceed their capacity during peak times. This shows that these facilities are poorly used throughout the logistics sector (Montreuil, 2011).

(14)

Urban freight comprises between 6-18% of all traffic within an urban region and accounts for almost 20% of energy use and CO2 emissions (Schliwa et al., 2015).

Even though there has been development in the logistics sector since Montreuil (2011) identified the difficulty to implement smart automation and strangled innovation as issues, the cost of implementing such technologies still prevents more companies from using them (Arunachalam et al., 2017). Furthermore, differences in the technologies and processes between different actors hinder the technological development. Most storage and handling systems that are currently used are based on the materials and goods they handle, which leads to marginal developments in these technologies due to a lack of standardisation.

The security and robustness of the logistics networks is the last issue Montreuil (2011) has identified. The robustness of the networks refers to the ability of the network to handle disruptions due to natural disasters as well as supply shortages, while the security of the network is the ability to deal with criminal activities as well as terrorism.

All of these issues lead to the current logistics system and usage of such logistics system not being envi-ronmentally, socially and economically sustainable and therefore not suitable for the future of logistics.

2.2 Cargo Theft

The security of SC networks has many different aspects. Cargo theft is one of these aspects that impacts SC’s constantly however it is largely ignored within the academic literature, in practical considerations for organizations as well as on a legislative level at the government. In today’s SC’s cargo theft is seen as a cost of doing business (Ekwall and Lantz, 2018). The Federal Bureau of Investigation (2013) defines cargo theft as:

”the criminal taking of any cargo . . . that constitutes . . . a commercial shipment of freight moving in commerce . . . [C]argo shall be deemed as moving in commerce at all points between the point of origin and the final destination, regardless of any temporary stop while awaiting transshipment or otherwise.”

(15)

annually (Boone et al., 2016). Obtaining accurate numbers however is difficult as many companies do not want to report some incidents. Most of the numbers only reflect the actual value of the goods that are being stolen without taking into account the full cost to replace the goods and send them again (Ekwall and Lantz, 2013). The full impact of cargo theft is thought to be around 1% of global GDP (Hoffer, 2010). Based on current data from The World Bank Group (2019) cargo theft would cost $200 Billion and $850 Billion in the USA and worldwide respectively.

The routine activity theory in crime defines crime based on three factors that have to come together (Ekwall and Lantz, 2015a, 2013):

• Motivated perpetrator

• Target (goods and equipment)

• Location and the lack of capable guardian

2.2.1 Perpetrators

(16)

Figure 2.1: Perpetrator Ability and Position (Ekwall, 2009)

The first distinction mentioned in the matrix is the level of the perpetrators ability. The fist type is the opportunistic perpetrator, who commits the crime without any prior planning. His actions are based on impulse decisions as the target that becomes available is desirable or usable. The opportunistic perpetrator is often dissuaded by the presence of visible security measures at the locations as there is no prior planning and therefore no prior knowledge of the security measures in place (Ekwall and Lantz, 2013).

The professional perpetrator on the other hand steals because he knows that he can sell the stolen goods for money afterwards. The actions by the professional perpetrators are planned in advance by identifying the weakest link in the security measures. Unpredictable elements such as human security or dogs are often targeted. additionally this type of perpetrator has the capacity to learn the technical expertise required to overcome predictable countermeasures (Ekwall and Lantz, 2015b).

(17)

The internal perpetrator on the other hand either has direct employment, previous employment or a similar relationship with the targeted organization and therefore has access to insider information that help plan and increasing the possibility of success of the theft of the cargo (Ekwall, 2009). Information can also be passed on from company insiders to the external perpetrators helping them plan the theft. Internal perpetrators pose the greatest threat to organizations. According to the U.S. Chamber of Commerce 75% of employees have stolen from their employer at least once (Walker, 2018). Ekwall and Lantz (2015b) mentions that the best way to prevent theft from internal perpetrators is by establishing proper work routines and protocols for the employees.

2.2.2 Criminal Opportunity

According to the routine activity theory, the combination of location and target is defined as the criminal opportunity (Ekwall and Lantz, 2015a). The criminal opportunity defines the following two characteristics of a cargo theft:

• Location

• Type of Incident

Ekwall and Lantz (2013) defines six different locations that are present in most SC’s. For the purpose of this research, SC facilities, which are the facilities of the owner of the goods either before transport or after will not be considered. Additionally, third party facilities and transport mode facilities will be grouped together as transportation facilities. Thus the following four locations are relevant for this research:

• Non-Secure parking • Secured Parking • Transportation Facility • En Route

(18)

for only 9% of incidents however it represented 24% of the value lost, meaning that the individual thefts were of considerably higher value. Lastly, on route incidents accounted for 14% of the incidents and 24% of the value (Ekwall and Lantz, 2013).

Various research has been done on the type of cargo theft incidents as well as modi operandi (Ekwall, 2009; Ekwall and Lantz, 2015a). For the purpose of this research the type of cargo theft incidents will be divided into the following categories

• Hijacking • Robbery • Theft of Vehicle

• Theft from Vehicle/Location • Deception

Cambridge English Dictionary (2019) defines hijacking is taking control of a vehicle during a journey, especially using violence. In our context, hijacking is the use of force, violence or threat with the vehicle and/or the goods being stolen (Ekwall and Lantz, 2018). This includes instances when a truck is forcibly stopped on a road.

According to the California Penal Code § 211, robbery is the felonious taking of personal property in the possession of another,from his person or immediate presence, and against his will, accomplished by means of force or fear. For this research, robbery does not include the forced stop of vehicles (Ekwall and Lantz, 2018). Robbery and hijacking are the only two types of theft incidents where violence is used.

The remaining three incident types are all non violent. The theft of vehicles is the theft of either the full truck with the trailer and the load or just the trailer and the load (Ekwall and Lantz, 2018). It is the theft of unattended vehicles or trailers without using force or violence.

The theft from the vehicle or location is the theft of goods without using violence or force. It is the theft of loads from unattended vehicles or trailers or from a facility (Ekwall and Lantz, 2018). Burglaries are included in this category.

(19)

fake logistics companies are set up in order to have goods delivered to or via them are also included in this category.

When looking at the TAPA EMEA IIS statistics, it can be seen that the instances of violence being used is relatively low. Even though there are differing opinions on the frequency of such violent instances, it is estimated to be around 5-10% of all cases (Ekwall and Lantz, 2018). This statistic however diminishes the problem as a survey conducted by the International Road Transport Union (2008) shows that one in six truck drivers interviewed had been attacked at least once over the last five years. Their online questionnaire also revealed that out of 2000 responses, there were almost 500 attacks on drivers.

Additionally, the value of the stolen goods is significantly higher with the use of violent incident types such as hijacking and robbery. However, incidents that involve internal help also have a significantly higher mean value. Hijacking, robbery and internal incidents have a loss value of over 200,000 Euros per incident compared to just 36,000 Euros when using theft from vehicle/location, the most common method (Ekwall and Lantz, 2015a).

2.3 Characteristics of PI relating to Network security

In addition to pointing out the problems with the current SC, Montreuil (2011) also identifies 13 character-istics of the PI and matches problems with charactercharacter-istics. The following seven charactercharacter-istics are addressing the network security and robustness issue:

• Objects encapsulated in world standard modular containers • Universal inter-connectivity

• Smart networked containers embedding smart objects • Distributed multi-segment inter-modal transport • Open global supply web

• Open performance monitoring and capability certification • Webbed reliability and resilience of networks

(20)

2.3.1 Encapsulation

Encapsulation in certain areas of the SC already exists today. Standard 20ft and 40ft shipping containers are the most prominent example of this. Additionally to these containers, massive logistics companies such as UPS, FedEx and DHL use encapsulation through their use of standardized parcels (Montreuil et al., 2013). The PI aims to expand this encapsulation to have globally standardized PI containers that can be transported by all of the shippers and carriers alike. These PI containers should be designed to be robust and secure in order to protect the encapsulated objects inside (Montreuil et al., 2015). Yang et al. (2018) states that PI containers need to be standardized in their encapsulation of the goods in order to facilitate the connection of logistics services. He describes the encapsulation features as being easy to interlock to other containers as well as other equipment and structures. These PI containers offer protection to their encapsulated goods as they are designed to be robust as well as being easy to seal (Yang et al., 2018). In order to handle all the different possible use cases for the PI containers, Montreuil et al. (2015) propose to develop PI containers of various different structural grades. Montreuil et al. (2015) also proposes the following three tiered PI container system that replaces all current packaging and containerization methods:

• Transport (T) Container • Handling (H) Container • Packaging (P) Container

All of the above containers will be PI containers, however there are major differences between the three. According to Montreuil et al. (2015), the T-container is planned to replace the current shipping containers with a number of modular dimensions. These containers are designed to withstand harsh external conditions such as storms and tough seas. This requires them to be more structurally robust, increasing the security of the encapsulated goods or other containers.

The H-Containers are planned to replace current basic handling unit loads such as pallets, boxes or crates (Montreuil et al., 2015). T-Containers are designed to withhold less rough conditions compared to T-Containers as they will mostly be encapsulated within the larger T-containers. Nonetheless, the structure of H-Containers should still be able to withstand handling within facilities and on carriers (Sallez et al., 2016).

(21)

encapsulate individual goods, their PI capabilities should enable them to be handled and sorted with a high degree of efficiency and accuracy.

2.3.2 Universal Interconnectivity

This universal interconnectivity is based on a set of standard protocols that have been specifically developed in order to allow for a cheap, easy and fast interconnection of PI containers through the PI’s nodes and routes (Montreuil, 2011). These standards and protocols will be applicable over the whole world to ensure fast and seamless interconnection of the containers. These protocols determine a range of activities from loading and unloading to orientation and storage activities. Furthermore, the protocols should be designed to be used in a fully automated and/or human-assisted scenario (Montreuil, 2011). These protocols will improve the handling efficiency at PI hubs and make the SC faster and more secure.

2.3.3 Smart Networked Containers

The PI containers will be built with smart capabilities which lead to an improvement of their performance. Each container will have a uniquely identifiable smart tag that, when scanned, will supply all necessary information about the PI container (Sallez et al., 2016; Landsch¨utzer et al., 2014). These smart tags allow the containers to distribute the information to the actors that need it. These smart PI containers will contain data security mechanisms that will only allow authorized parties to access information about the contents of the container (Sallez et al., 2016). Additionally to the information about the current goods, the PI containers will be able to offer information about past shipments. Montreuil (2011) suggests using RFID and/or GPS capabilities with the PI containers smart tags. Ultimately, these smart tags will be able to communicate fully with the internet, allowing for on the spot decision making of the containers to enable self-control through the PI, making them more secure (Montreuil, 2011).

2.3.4 Distributed Inter-modal Transport

(22)

2.3.5 Open Global Supply Web

Standardizing all of the protocols and systems into the global PI system, allows for the exploitation of an open global supply web. All of the supply chains and logistics actors that currently operate are using a networked that is closed to the vast majority of other actors. The PI allows organizations to exploit all of the SC networks that exists today within the PI network (Ballot et al., 2012). All of the warehouses and distribution centers that are currently used by a very limited number of enterprises will be accessible to all. This enables the PI to exploit the distributed inter-modal transport. In the USA alone, PI containers will be able to exploit the vast network of over 530,000 warehouses and distribution centers, leading to a greatly reduced travel time as the PI containers will be able to route themselves in the most efficient manner (Montreuil, 2011).

2.3.6 Open Performance Monitoring

The PI containers smart capabilities allows the actors within the PI to monitor the performance of the sys-tem in real time. Key performance indices such as speed, service level, reliability and also safety and security can be accessed in real time (Montreuil, 2011; Montreuil et al., 2010). Furthermore, the smart capabilities allow the PI containers to monitor their own state such as shock, temperature, products contained, as well as live location information (Yang et al., 2018). In case of any disruptions in its systems or irregularities, the PI container has the capability to communicate these issues immediately to the required actors. E-seals on the containers can alert the authorities in case of unauthorized tampering immediately, in order to minimize or avoid any possible losses (Yang et al., 2018). These live tracking capabilities enable the PI containers to use real-time fact-based decision-making. Even though a lot of information will become available, the PI system still respects the confidentiality of certain transactions (Montreuil, 2011), thus improving the security of the system. The vision of the PI however does not conclude with giving the PI containers smart capabilities; certified PI conveyors, PI facilities, PI roads and even whole regions and cities should be established, enabling full monitoring of the PI container within the system (Montreuil, 2011).

2.3.7 Webbed Reliability of Networks

(23)

design the most effective and efficient route. Common protocols and structures at all of the PI nodes allow them to be substituted if necessary. The PI monitors the accessibility of all of the PI nodes and routes in order to assist the containers in their routing decisions (Montreuil et al., 2012).

2.4 Gap in Literature and Research Questions

Even though the literature on PI mentions how its characteristics can help improve the security of the PI supply chain, there has not been no attempt yet on how to explicitly design the security mechanisms within the PI. Furthermore, the transport of high value goods within the PI has not been mentioned in the literature. This gap is apparent as there are numerous cargo transport companies that provide solutions to high value transport in the current logistics environment which has not been examined and added to the PI literature. An early example of the transport of high value goods is the shipping of the biggest diamond ever found from its mine in South Africa, to its new owner in England and again to the cutters in the Netherlands. Massive precautions were taken to transport this diamond via ship to England with police escort, only to use deception as the diamond was anonymously sent by mail (Bariand et al., 1992). In the subsequent travel to the cutters, deception was also used to mask the delivery of this diamond.

Shippers currently offering high value transport services have numerous, more sophisticated security mechanisms to ensure the security of the goods. Specialty high value shippers such as G4SI (2019); Rutges (2019) and VVT Europa (2019) as well as bigger shippers such as DHL (2019) and Mainfreight (2019) offer high value transport with similar security mechanisms:

• Security trained employees

• GPS track and trace with monitoring • Modern Locks

• Secure Parking • On board cameras • Digital Logbooks • Panic Buttons

(24)

Most of these mechanisms however are not mentioned explicitly by either the cargo theft or the PI litera-ture. Furthermore, a lot of the mechanisms currently employed require significant manpower. Adding these mechanisms into the PI to be automated can save significant costs.

The interactions of all of the involved stakeholders along with the PI specific characteristics and measures remains to be investigated. Therefore the following main research question has been developed:

(25)

3 Methodology

The following section will describe the methodology used in this paper. DSR was the method chosen to conduct this research as according to Dresch et al. (2015a), it is ”is a science that seeks to develop and design solutions to improve existing systems, solve problems, or even create new artifacts that contribute to better human performance”.

3.1 Research Questions

Wieringa (2009) defines a four step approach to DSR studies,focusing on problem investigation, design solution, design validation and solution implementation. This research will only focus on the first three phases of this framework. The above mentioned main research question ”which requirements and needed functionalities must the PI-system cover in order to manage the security of high value goods in an effective and efficient way?” is therefore split into smaller questions for each of the three phases:

Problem Investigation

• What are the issues with cargo theft in the current logistics industry? • What is already known in terms of security within the PI literature?

Design Solution

• What does a possible solution for security of high value goods in the PI look like?

Design Validation

(26)

3.2 Research Design

Figure 3.1 shows the steps that are necessary in order to conduct this research based on the framework by Wieringa (2009).

Figure 3.1: DSR Framework

3.3 Analysis

(27)

Therefore, the first step in this research was to analyze the current literature of the PI regarding the security of high value good and to identify the current characteristics of the PI that assist in securing the goods. The integrative literature review also does not solely focus on a single topic but rather is designed to combine insights from multiple different topics and research fields (Snyder, 2019). For this research, the PI literature was supplemented by a literature review of current security issues in the logistics industry. As this literature is also relatively new, the focus was put on cargo theft issues in the trucking industry. This revealed a gap in the current literature, not only regarding supply chain security within the PI but supply chain security in general. A detailed review of the most significant issues of cargo theft was conducted. This was a problem-driven investigation according to Wieringa (2009), in order to fully understand the problem and requirements needed. These requirements and issues, that were identified during the literature review phase of the PI and cargo theft literature, were complemented by requirements found through the analysis of current industry leading high value security transport companies.

3.4 Design Solution

The second step in the framework by Wieringa (2009) is the development of a design solution based on the analysis of the problem. Since there is no paper available yet that describes the secure transport of high value goods in the PI context, a new model has to be developed. First, a list of relevant stakeholders based on PI literature with combined with cargo theft literature was established. Based on that a use case scenario of a one time order for high value cargo transport within the PI was described and needed requirements and other functionalities were developed based on that. The model developed with IDEF0 then was based on the characteristics of the PI found in the existing literature and expanded with the requirements found in the analysis phase. As this model cannot be compared to previous, existing models of high value transport in the PI, it was crucial to develop operational scenarios of the model, based on the literature review, and design the model accordingly. Alternatives for crucial functions of the model were developed as well along with interaction diagrams between the different functions within the system and detailed descriptions of these interaction were created.

3.5 Design Validation

(28)

explained to the industry experts with all of the necessary assumptions. Additional information on the basic PI system as well as the design alternatives will also be explained to the experts to strengthen the validation. Semi structured interviews with a broad range of open questions regarding the feasibility of the proposed system will be performed in order to allow for unbiased opinion of each expert participant (Adams, 2015).

4 Analysis

The supply chain of high value items, such as gold, is highly secretive and confidential in order to protect the integrity of the supply chain. This makes it difficult for researchers to locate best practices in the industry in order to advance the supply chain security in regards to cargo theft. The only information available is that of the logistics service providers with their offer and details about their capabilities for high value transport security, which will be the basis of this analysis along with guidelines developed by TAPA (2017). The current high value shipment is limited to a few companies and requires additional manpower compared to regular shipments. The goods owner has to contact the specialty shippers and negotiate prices and services based on their ability to transport the high value goods from origin to destination facility. These shippers will then arrange and perform the full shipment from origin to destination facilities.

When the high valuable goods owner currently wants to send the goods to a destination facility, they have to contact the specialty shipper and communicate the exact details of their shipment. In coordination with the shipper then, the details of the security mechanisms and extra security has to be negotiated. Based on the solutions offered by a number of shippers there are a few security mechanisms that are identical:

• GPS in truck and trailer with monitoring • Secured parking places

• Special trained drivers and employees • Modern locks

• Emergency buttons and communication

(29)

generally a single truck with either one driver or two driver, depending on the distances and decisions made by the shipper, are used. These trucks and trailers, equipped with the security mechanisms mentioned above, will be monitored from a control center by either employees of the shipper, or in central command stations, depending on the shipper used. The shipper can decide, when planning the route, if one driver or two drivers are needed in order to get the truck form origin to destination. If this cannot be achieved without having to stop for fuel or for prescribed diver break times, the route has to be adapted ahead of the shipment to include only partial routes that start and end at secured parking locations or secured break points. These have to be designed ahead of schedule, making them vulnerable to short term changes in circumstances on the road. In case of these high value transport conducted by a shipper, one truck will usually be used the entirety of the route.

Figure 4.1 shows a simplified framework for decisions in the high value secure transport en route for trucks. This framework is based on two important questions:

1. Do we need breaks between the origin and the destination facility?

2. Do the sensor/ monitoring system and/or the driver report irregularities?

(30)

(31)

5 Design Solution

5.1 Main Optimistic Scenario

In the following chapter the analysis of a possible high value secure transport within the PI will be done through developing and explaining the main optimistic scenario of the proposed security system. This will be done through explaining the stakeholders involved in such a high value transport scenario first. After that a use case scenario of a one time order within the system will be laid out. This will be followed up with an explanation of the needed functionalities and requirements of the proposed system in order to successfully secure the high value goods throughout the whole PI. In order to do that, the entire decisions of all involved stakeholders before and during the high value transport will be described.

5.1.1 Stakeholders

The following is a list of all the involved stakeholders in the PI for the secure high value transport.

• Goods Owner • Shippers • PI Hub • Law Enforcement • Insurance companies • PI Container Owner • PI container Goods Owner

The function of owner of the goods owner is to request the shipment in the PI. In this high value security model, the goods owner has to decide if the goods are worth enough to warrant a high value transport within the PI or if the regular PI shipment is sufficient for delivery. Getting the desired services with the required security, at the desired time and within a price range is the main objective of the goods owner.

Shippers

(32)

goods from origin to destination as it depends on the distances and the capabilities of the shipper. The goal of the shipper is to transfer the goods without security issues within a certain time frame for a certain price. The shippers are looking to decrease the transport effort needed while increasing the loading level of the modes.

PI Hub

The PI hubs are mainly responsible for loading and unloading of the PI container to and from the different modes of transport serviced by the hubs. Additionally to that, the PI hubs serve as secure parking lots within the PI for times when the containers cannot be moved. The main goal of these hubs is to handle the PI container loading and unloading in the most time and cost effective way possible while still providing a high level of security.

Law Enforcement

Law enforcement plays a crucial role in this system to ensure the safety of all the stakeholders involved and to guarantee a quick response in case of any security breaches in the system. Additionally, law enforcement is a primary source of information to determine the security of routes.

Insurance Companies

Insurance companies within the high value PI transport act as another risk mitigation tool for the goods owners involved. They provide the needed layer of security for the goods owner in case the system does fail and the goods are lost. The goal of the insurance companies is to provide this risk mitigation to the system, but also to earn money.

PI container Owner

The PI container owner provide the PI containers to be used in this system. For the high value transport, the PI container owner has to arrange for a robust PI container to be available at the right place and time and for the right price to the request of the goods owner. The PI container owner has to guarantee that all the security mechanisms of the PI container work as intended. They have to guarantee the structural integrity of the provided containers and the functioning of all installed sensors, tracking or other security mechanisms.

PI container

(33)

stakeholders. The PI container is responsible for choosing the best route through the system that ensures high security, low cost as well as low amount of time needed. Another goal for the container is to find suitable transport opportunities in order to keep the idle time to a minimum.

5.1.2 Use case scenario of one time order within PI

1. Goods owner orders a high value secure transport within the PI. Pick-up time, location, price range, extra security measures needed, as well as shipment details such as weight and size will be needed to request such shipment.

2. The matchmaking process assigns a suitable PI container to the request based on the shipment re-quirements outlined earlier.

3. The PI container arrives at the goods owner facility and the high value cargo is loaded unto the containers by authorized personnel.

4. Only stopping at secured parking spots and/or PI hubs, the goods are transported using a variety of shippers and transport modes. The routing decisions are made by the PI containers based on cost, speed and security of the routes and modes.

5. During the whole transportation process, the PI container will monitor its own status through a variety of sensors build into the PI container.

6. After delivery of the PI container at the location facility, the high value cargo is unloaded and the PI container released back into the PI system.

7. The high value PI container is inspected to ensure all functions and sensors work properly, before it is cleared for future matchmaking.

5.1.3 Security PI system requirements and needed functionalities

Hiring and internal security considerations

(34)

Container requirements for high value

The PI containers used for high value goods transport within the PI will need to be structurally more secure than a normal PI container, while at the same time being indistinguishable from an ordinary PI container. The PI containers also have to have extensive monitoring abilities that need to be ensured throughout the entire journey of the goods. Before going through the assignment process, the PI containers for high value goods have to be checked to ensure all monitoring equipment is working properly. Smart tags have to be present in all of the PI containers that contain the details of the shipment. These full details are only visible to authorized employees at the origin and destination facility, as well as customs and border patrol (CBP) agents, within authorized areas only. Other than these people, only the PI container will know the contents of itself. The employees of shippers as well as PI hubs will only know the destination location as communicated by the container. Additionally to the smart tag, the PI container needs to have monitoring capabilities such as live location, shock, temperature as well as electronic seals that have communication abilities (Yang et al., 2018).

As mentioned earlier in the background, there are 3 types of PI containers (P, H and T) (Montreuil et al., 2015). All of these containers should have a version of them capable of performing these high value transports, depending on the characteristics of the goods.

Routing considerations

(35)

Extra Security considerations

When a high value transport then becomes necessary for the goods owner, there are a few decisions to be made regarding the shipment.

• Type of PI containers? • Number of PI containers? • Number of drivers? • Security escort?

The first question is if high security PI containers are needed for the transport at hand or if the standard PI shipping is sufficiently secured compared to the risks of losing the goods. Additionally, the size and weight of the goods to be shipped will determine the size of the PI containers used. Preferably a secure PI H container is used as this will be loaded into a PI T container on the trucks, thus providing an extra layer of security. If the goods are too large or too heavy to fit within a secure PI H container, a secure PI T container has to be planned for.

The second question is if one container is enough to send out or if a number of container is needed, only one of which will contain the actual goods. This will lead to a reduced opportunity for the perpetrators as they cannot be sure of the contents of the PI containers upon leaving a PI hub. This decision is made by the PI container upon consideration of all relevant information for each partial route.

The third question is how many drivers are needed for the transport of the high value goods. This depends on the location of the goods owner facility as well as the location of the nearest available and useful PI hub or secured parking facility. A general rule for the trucks is that they are not allowed to stop anywhere except at secured parking facilities or at PI hubs, which the PI containers will communicate with the drivers. This means that multiple drivers may be required if the distance is too large for a single driver to reach any secure parking spot or PI hub within one allowed shift.

(36)

Emergency response protocols

The PI system will automatically monitor each of the high value secure PI containers, and it will trigger an alarm as soon as one of the following is detected:

• GPS of PI container disabled or masked

• Tampering with either of the PI containers and/or breaking of the E-seals • Unauthorized changes in routes of the truck

In case of any of the sensors installed in the containers picking up anomalies, the system will contact the driver immediately, who then needs to communicate back to the PI system that either everything is okay and it was a false alarm or confirm the security breach. The driver will only have a specified time to respond to the request. If the driver fails to respond to the request by the PI system in a timely manner, or he confirms the alarms triggered by the sensors, the PI system will call law enforcement officers to the exact location of the high value secure PI container. In case of communication breakdowns with the container or the driver, law enforcement will be sent to the last known location. In case of special security escorts accompanying the high value secure PI container, they will be asked to confirm the alarm as well. Both the driver of the truck or the security escort additionally can notify law enforcement immediately in case of security breaches or suspect behaviour, even before the sensors can pick it up. These mechanisms will be able to act in case of hijacking attempts on the transport mode.

5.2 Functional Architecture

(37)

(38)

(39)

5.2.1 Perform Secure Transport

The function ”Perform Secure Transport”, shown in Figure 5.3, is decomposed of the functions ”Perform PI functions”, ”Perform Goods Owner Functions”, ”Insure Goods” and ”Perform Law Enforcement Functions”.

Figure 5.3: Perform Secure Transport

(40)

(41)

Interactions Descriptions

Insurance Request Insurance request is initiated by the goods owner and it contains all necessary information for the Insurance companies to provide the insurance

Insured Goods This interaction confirms that the goods will be insured along the transport. It provides an extra layer of security in case of failures of the system

Shipment Request The shipment request is triggered by the goods owner and contains all necessary information such as size, weight, value, description of the goods along with any extra security details that the customer might want.

Tracking Informa-tion

Tracking information helps authorized personnel at the goods owner to track the shipment throughout the whole logistics chain.

PI alarm The PI alarm is triggered by the contingency function and allows the Law enforcement to respond to issues within the system. Tracking and location information needs to be send along with the alarm to law enforcement to allow for easier locating.

Manual Alarm The manual alarm is either triggered by the driver of the transport mode or the PI hub employee in case of emergency. Information on location also needs to be send to law enforcement along with the alarm.

Open Lock This interaction allows law enforcement to open the container to check and verify its contents. An inspection request is needed to trigger this interaction.

Inspection Request This interaction can be done by authorized law enforcement personnel that need to access or verify the contents of the PI container. Along with the authorized personnel, this interaction has to be done within authorized zones in order to further verify that validity of this request. This should be used especially for border patrol and customs agents within their designated facilities.

Law Enforcement Response

Law enforcement response is the physical movement of law enforcement per-sonnel to the location of the triggered alarm.

(42)

5.2.2 Perform PI Functions

The 0 function ”Perform PI functions” is divided into two sub-functions, as can be seen in Figure 5.5 below.

Figure 5.5: Perform PI Functions

(43)

(44)

Container Request This interaction requests the correct container from the matchmaking process based on the specifications that have been decided.

Container en Route This interaction originates from the transport mode and updates the rest of the PI functions to make planning easier.

Loading Informa-tion

Loading information is the information for the rest of the PI functions to where the container needs to go next within a PI hub. The core PI functions will then assign the container to the right transport mode.

Correct PI con-tainer

The correct PI container is delivered to the system based on the specifications within the container request.

Released Container The core PI functions tell the secure system when the container is released and put on a transport mode. From then on the secure PI transport function will handle the container until the next PI hub.

(45)

5.2.3 Secure PI transport

The 0.Sec function ”Secure PI Transport” is divided into three sub-functions, as can be seen below in Figure 5.7.

Figure 5.7: Secure PI transport

(46)

(47)

Route Information The route information are the driving instructions for the transport mode and the driver developed by the PI container through analysing all available data to ensure high security.

PI Alarm Confir-mation Request

When the sensors within the container report tampering or breach, either the PI hub or the driver of the transport mode is requested to confirm the alarm, depending on where the container is at the time of the alarm. There is a time limit in which the employees have to respond to the alarm, otherwise the alarm is confirmed.

PI Hub Status Re-quest

The PI container requests the status of the PI hub in order to make sure it is safe to go through that PI hub.

Transport Mode Status Request

The PI container requests the status of the transport mode.

Transport Mode Status

The transport mode status is used for all levels of decision making, to decide if a transport mode is safe to plan with and if the current transport mode is safe.

PI Alarm Confir-mation

Once the alarm is confirmed the contingency actions will trigger the PI alarm to law enforcement.

PI Hub Status The PI hub status is used for all level of planning: strategic, tactical and real-time in order to determine if it is safe to use this PI hub

(48)

5.2.4 Secure PI Hub

The 2 function ”Secure PI Hub” is divided into three sub-functions as can be seen in Figure 5.9 below.

Figure 5.9: Secure PI Hub

(49)

(50)

Inspection The PI hub back office periodically inspects the security mechanisms that secure the perimeter of the PI hub to make sure that no breaches are possible.

Inspection Results The inspection results allow the PI hub back office to update the container about their status. If the inspection found no issues with the perimeter security, the PI hub can be used by the high value container.

PI Hub sensor Data Sensors are installed at the PI hub perimeters to immediately detect tampering with the perimeter and communicate their findings with the back office so that the PI hub status can be updated in real time.

Background Check Background checks have to be conducted by the PI hub back office to ensure each of their workers is cleared.

Background Infor-mation

The PI hub employees have to share their background information with the back office to clear them for working in the Hub.

Alarm Check When the PI container an alarm confirmation request, this is received by the back office and the employees are asked to investigate the alarm within a cer-tain amount of time, otherwise the PI alarm is considered as confirmed and contingency actions will be conducted.

(51)

5.2.5 Secure Transport Mode

The 3 function ”Secure Transport Mode” is divided up into three sub-functions as can be seen in Figure 5.11 below.

Figure 5.11: Secure Transport Mode

(52)

(53)

Inspection The transport mode back office has to inspect the safety mechanisms of their vehicles to make sure that they all work.

Inspection Results The inspection results of the vehicles are shared with the back office so that the transport mode can be used for secure transportation.

Background Check Background checks have to be conducted by the transport mode back office to ensure each of their workers is cleared.

Background Infor-mation

The transport mode employees have to share their background information with the back office to clear them for working in the vehicles.

Alarm In Vehicle The PI alarm request enters the vehicle, which notifies the driver of the vehicle. The driver then has limited amount of time to confirm or deny the alarm. If this is not done within the specified time, the PI alarm will automatically be declared as confirmed and contingency actions will be performed.

Route Information Visualization

The vehicle receives the route information and visualizes this information to the driver in the form of navigation instructions. If the driver deviates from the route, the PI alarm will be triggered and the driver will be asked to confirm the security of the vehicle.

(54)

5.2.6 Perform PI container security functions

The four different sub-functions that make up the 1 function ”Perform PI container security functions” can be seen in Figure 5.13 below.

Figure 5.13: Perform PI container security functions

(55)

(56)

Sensor Information The sensors and locks installed in the container will update the container about their status at all times.

Lock Release Order A lock release order can only be done when the container has either reached its final destination, or a verified inspection request is being processed.

Fortification Re-quest

The fortification request is done based on the details from the shipment request and asks the fortify container function to plan the needed fortifications based on the available data.

Fortifications Fortifications are used in the container to aid in the deterrence of perpetrators.

Fortification De-tails

After the fortifications have been planned by the fortify function, the fortifi-cation details are send to the use information function to request the correct container.

Deterrence Request A deterrence request is send based on the shipment requests details so that the deter perpetrator function can plan their requirements for the container. It also sends information needed to conduct the route plan.

Deterrence Re-quirements

The deterrence requirements are sent to the fortify function so that the correct fortifications for the situation can be planned.

Route Plan The deter function plans the route with all the information it receives in order to secure the container fully.

Triggered Alarm When the use information function detects inconsistencies within the system or tampering, a triggered alarm is sent to the contingency function of the system. This alarm then has to be either confirmed or denied by the responsible functions.

Lower Risk The deter function lower the risk of contingency actions needing to be per-formed as it obstructs and deceives perpetrators.

Threat of Action The threat of action from the contingency actions is a factor that deters per-petrators from attempting to tamper with the PI containers.

(57)

5.2.7 Use Information

The 11 function ”use information” is divided into three different functions as can be seen below in Figure 5.15.

Figure 5.15: Use Information

The sub-function ”use strategic information” is responsible to plan out possible hubs and transport modes that could be used by the PI container. It would contain updates on roads and long term construction. This information is stored in a database for all of the PI containers to access and plan accordingly. ”Use tactical information” receives the details of the shipment and plans the transport of the PI container initially. It creates a tactical plan, which is used by ”use real time information” as a baseline for comparing real time data. This tactical plan is updated if needed with real time information in case of changes to it. The sensor information from the container is used by the function ”use real time information” to compare and communicate the route or plan to the necessary stakeholders.

PI Database The strategic information from all information sources are collected to for a PI database with which the security of the upcoming transports can be planned. Database Update When the tactical information function receives new information from all

sources, the strategic information’s database is constantly updated.

Tactical Plan The tactical information is used in a tactical plan which provides the baseline for the real time information and allows the container to compare the plan with the real time information.

Tactical Plan Up-date

The tactical plan is updated based on the real time information it is processing.

(58)

(59)

5.2.8 Fortify Container

The 12 function ”fortify container” is divided into three sub-functions as can be seen in figure 5.17.

Figure 5.17: Fortify Container

The ”plan and implement fortified asset” sub-function is responsible for taking the fortification requests as well as the deterrence requirements into consideration and plan the level of PI container fortifications that are required. The ”operate fortified asset” sub-function is responsible for maintaining the fortifications and monitoring devices. The ”monitor fortified assets” sub-function is responsible for making sure the forti-fications are intact and the information is send to the information function. The interactions of the ”fortify containers” function can be seen in figure 5.18 with the explanations of it in table 5.8.

Feedback The monitor fortified assets function constantly provides feedback to the oper-ate fortified assets function to make sure the fortifications are working correctly.

(60)

(61)

5.2.9 Deter Perpetrators

The ”Deter Perpetrator” function of the model is based on the two principles of obstruct and deceive the perpetrators. There are three alternatives outlined below that all obstruct and deceive the perpetrators and thus lower the risk of security breaches and subsequently the risk of contingency actions. The main differences between the options below is the level of security of each option and the cost of each option. These alternatives do not have to be used exclusively but rather can complement each other in different phases of the transport depending on the needed security and the cost of such security. Considerations on which of these alternatives are to be used can be made for each part of the journey the PI container undertakes. The risk factors of the route between two PI hubs along the journey heavily influence the decision on which of the alternatives to employ. The cost of the transport between two PI hubs therefore has to be weighed against the security risks as well as the value of the goods.

Single PI container

The single PI container alternative is based on the principle that the inherent security functions of the traditional PI along with the added functionalities of the security model will be sufficient to securely transport the cargo from origin to destination. The single PI container will be indistinguishable from other PI containers and therefore provide security through not being able to identify the high value PI container. Even though the containers will look identical to the other containers, they will be able to obstruct perpetrators through a higher structural integrity of the container itself, making it more difficult to penetrate. The deter function will also route the transport mode in such a way to avoid known security risks and high risk areas in advance and be able to adjust the routing with real time information from PI hubs, other transport modes, other containers as well as law enforcement. The single PI container alternative to transporting high value cargo is by far the cheapest of the three alternatives, however it also provides the least amount of security.

Multiple PI container

(62)

PI container using special transport

The third alternative to deterring the perpetrators is through using PI containers on special transports. These can either be through using air transport with additional security or sending armed escorts to ensure the security of the goods. Even though these transports could easily be identified as carrying goods of high value, the extra security measures added increase the overall security of the shipment. Using special transports on any of the journeys between two PI hubs is of course based on the willingness of the goods owner to pay for extra security if the value of the goods is sufficiently high enough. Especially for high value transports from remote production facilities for high value goods these special transports can be particularly useful. Similarly, if the distance from the goods owner facility to the nearest PI hub is too large to use standard trucks, or if the trucks have to take breaks in order to complete the journey between two PI hubs, these special transports are a valuable alternative.

5.3 Validation

Through the fully envisioned nature of this model, and even of the PI itself, validation through imple-mentation was not feasible. Therefore, the validation of this model had to be done through interviews with industry experts in high value goods transports from several industry leading special goods shippers. These interviews were all conducted after the design of the model and its functionalities was completed. The in-terviewed experts were from two different companies that operate mostly Europe wide but are also capable of delivering high value goods internationally. All of the interviewed companies have a the highest possible security certification from TAPA. Additionally to the validation through the transport companies, an expert from TAPA itself was also interviewed to validate this thesis.

The basic principles of the PI had to be explained to the interviewees beforehand. After that, the main optimistic scenario was illustrated through the use case scenario of a one time order within the PI. Added to this, the specific security considerations and alternatives that are available to the goods owner for each part transport was further explained to the experts. Based on this information, the researcher questioned the experts for feasibility of the high value transport within this system.

The basic concept of digitizing the full transport and using electronic means to track and trace high value goods was very well received by the first interviewee. He identified a ”real need to get interactions digitized”. Even though the sensors are already in place for high value security transports, he acknowledged that is ”sensors already in place but can be optimized”.

(63)

The concept of encapsulating the high value cargo into smaller containers, which in turn will be loaded onto bigger containers was very well received by the second interviewee. The current high value cargo transport deals with full truck load high security transports. ”On high security [we are] focusing on FTL [full truck load] but we see that more and more the questions of customers come in on LTL [less than truck load]”.

Regarding the three alternatives for the deter function, the first interviewee mentioned that ”extra security trucks could be an idea, but quite expensive”, and that they would only be used for very high value cargo. The routing of the high value containers through multiple hubs to increase efficiency was also well received during the first interview as the expert stated that ”all the hubs and the interactions [is] where it needs to go”.

The TAPA experts main concern for the first deter alternative was that trucks leaving a manufacturing site that only produces high value goods that ”all the goods are going to be high value”. Furthermore, he suggested adding refrigerated high value containers into the system to allow for transport oh pharmaceuticals and other high value refrigerated goods. The other concern from this interview was about the insurability of the items and the concerns of goods owners about this: ”As long as the system takes into consideration those concerns (insurability) of the industry I think you will be fine”.

As most of the concepts of securing the high value cargo used within this research is based on industry standards, the interviewed experts were very receptive towards the new system. They all agreed that this system was feasible as they are currently already employing a lot of sensors within their transport. The main outcomes of the validation interviews were:

1. The system could work if legal issues with responsibilities of the driver are sorted

2. There’s a real demand for LTL high value secure transports

(64)

6 Discussion

6.1 Novelty

Managerial

During the validation phase, it became clear that developing a system for less than truckload high security transports was a definite business case and consumers are asking for this already. Since most high value transport companies operate with full truck loads, sending a less than truck load high value shipment becomes more difficult. Since this system operates on the H-containers, i.e. current unit loading devices, it grants a higher flexibility for managers to request such transports and smaller loads can be shipped cost effectively. Additionally to that, the current way high security transports are working is within a closed network of highly specialized transportation companies. In order to access this network customers have to negotiate the terms with the transportation companies who plan the route according to the customer specifications. Through the addition of high value transports within the PI, customers can order high value transports much quicker and cheaper. Most high value transporter currently employ the third option within the deter function, meaning special transports. The cost reduction for customers therefore can be achieved by employing either the single or multiple PI container options to the partial routes wherever it is deemed sufficient and the special transport option if it is necessary.

Theoretical

The main contribution of this thesis to the literature is the addition of high value transports in the PI using the deter function with three alternatives. This function supplements the PI literature with the aspect of removing criminal opportunity from the high value transport, influenced by the cargo theft literature. Yang et al. (2018) mentioned that the PI container itself has impact on the security, however this research adds deception of perpetrators in order to not only rely on the physical barriers of the PI containers as the sole source of security for the cargo.

A security-driven design for high value cargo transport within the Physical Internet