The Introduction of the Central Planning Facilitator into the Physical Internet system, from a Strategic-Business Perspective.

(1)

The Introduction of the Central Planning

Facilitator into the Physical Internet system, from

a Strategic-Business Perspective.

February 10, 2020

Master’s Thesis

MSc Technology & Operations Management

University of Groningen, Faculty of Economics and Business

Sieuwke Elisa de Jong

Student Number: S2723727

Supervisor: Dr. N.B. Szirbik

(2)

Abstract

Purpose: In the current Physical Internet literature, a gap is identified between the design of the Physical Internet as a decentralized system and the optimization of the Physical Internet based on the transparency and information availability of a centralized system. The aim of this research is to explore if the Physical Internet can be designed as a hybrid system, by introducing the Central Planning Facilitator into the system. The Central Planning Facilitator is a central party, which provides services for the planning and control of shipments for multiple units.

Methodology: The purpose of this study is to explore the Central Planning Facilitator from a strategic-business perspective. This is done by performing a design science research. The problem is thoroughly investigated in theory and in practice. Based on this, critical requirements are established and three scenarios are designed for the Central Planning Facilitator.

Findings: The three scenarios designed in this research are: the Central Planning Facilitator operating as information facilitator, as forwarder and as capacity platform. Based on a validation, the Central Planning Facilitator as forwarder is identified as the most realistic scenario from a practical perspective to create a hybrid system for the Physical Internet. However, some changes in the design are required.

Contribution: This study contributes to the Physical Internet literature by making the first small steps in closing the gap between the design of the Physical Internet as a decentralized system and the optimization of the system as a centralized system. This is done by introducing the Central Planning Facilitator into the system, which enables the Physical Internet to act as a hybrid system. Furthermore, this study identifies multiple important future research topics for the Physical Internet.

(3)

1 Introduction

The Physical Internet increasingly gained attention the last ten years, which is driven by the fact that logistics systems are not environmentally and socially sustainable (Sternberg and Norrman, 2017). Current logistics systems are spatially and organisationally fragmented, leading to product flows crisscrossing through disconnected individual networks (Sarraj et al., 2014). Together with the growing environmental awareness, this inefficiency has led to the envisioned concept of the Physical Internet. The Physical Internet is considered as a collaborative and open logistics system, which aims to improve the efficiency of the way physical objects are supplied, moved and stored around the world (Crainic and Montreuil, 2016; Montreuil et al., 2010; Pan et al., 2017). It is envisaged that smart, standardized and modular containers are used, which can determine their own route in an efficient logistics system of shared transportation, cross-docking and storage, and which can be handled by any organization due to logistic protocols (Montreuil et al., 2010).

The Physical Internet is currently designed as a decentralized system regarding the planning and control of shipments. A decentralized system is a system that consist of independent self-coordinating units (L¨oh et al., 2000), which plan, control and make decisions based on the optimization of their own goals (Jung et al., 2009). For the Physical Internet, it is envisaged that all the different units, like containers, transporters or suppliers, act as self-coordinating units. They coordinate their own business processes, shipments, routes and schedules based on the optimization of their own goals. For example, the smart containers autonomously decide which optimal route to follow, and they autonomously optimize the handling and sorting of movements at the different hubs (Sallez et al., 2016). The Physical Internet as a decentralized system is desirable, since it can provide an open and global system, which can warrant its own reliability and resilience (Montreuil et al., 2012).

In contrast, many optimization models that are dedicated to the Physical Internet assume the transparency and information availability of a centralized system. For example, Pan et al. (2015) developed an inventory decision-making model for the Physical Internet and assume information availability between retailers and vendors as in a centralized system. A centralized system is a system where a central party manages the coordination of multiple units in the system, and plans, controls and makes decisions based on the optimization of the whole system (L¨oh et al., 2000). The central party has information from all units to base decisions on, which is why centralized systems are easier to optimize and more efficient (Ding and Kaminsky, 2019). Sternberg and Norrman (2017) and Ambra et al. (2019) recognize the contradiction between the design of the Physical Internet as a decentralized system and the optimization of the Physical Internet as a centralized system and therefore question if the Physical Internet should be designed as a decentralized system or as a centralized system.

(6)

sys-system should be created, which allows for more flexibility and independence of different units, while simultaneously retaining the simplicity and efficiency of a centralized system to a certain degree (L¨oh et al., 2000). This study explores if a hybrid system could be created for the Physical Internet, by introducing the Central Planning Facilitator into the system. The Central Planning Facilitator is a central party, which provides services for the planning and control of shipments for multiple units on a more local level. This allows the Physical Internet to act as a hybrid system, where decisions on a global level are made as in a decentralized system and decisions on a more local level are made as in a centralized system, which is supported by Vargas et al. (2019). By designing the Physical Internet as a hybrid system, this study tries to make the first small steps in closing the gap between on the one hand the design of the Physical Internet as a decentralized system, and on the other hand the optimization of the Physical Internet based on the transparency and information availability of a centralized system.

This study takes a strategic-business perspective while exploring the Central Planning Fa-cilitator. A strategic-business perspective focuses on the stakeholders, the provided service, the way money can be made, the ownership and the scale of the Central Planning Facilita-tor (Oktaei et al., 2014). The reason for this perspective is because the Central Planning Facilitator plays a key role in the planning and control of shipments, which impacts multiple stakeholders in the Physical Internet (Li et al., 2012). In addition, the Central Planning Fa-cilitator is yet a non-existing party in the current envisaged Physical Internet. Therefore, it needs to be explored how the Central Planning Facilitator can exist as a business and what its added value is for the stakeholders. The critical requirements for the Central Planning Facilitator are established in this study, which is the input for the design of the Central Planning Facilitator (Hevner, 2004). Critical requirements can be defined as the essential conditions to enable the Central Planning Facilitator to be valuable in the Physical Inter-net system. By building on the available literature about planning and control in logistics systems and the Physical Internet, and by conducting a practical analysis, a conceptual design of the Central Planning Facilitator in the Physical Internet system is introduced. The research question that is addressed in this research is:

How can the Central Planning Facilitator be designed in the Physical Internet system, and what are the critical requirements for such a party from a strategic-business perspective?

The following sub-questions are developed to help answer the research question.

1. How are central parties, that could be similar to the Central Planning Facilitator, currently designed in logistics systems according to the academic literature?

2. How is the Physical Internet currently envisaged in the academic literature regarding the planning and control of shipments?

3. What are the main stakeholders of the Physical Internet in the academic literature and what are their goals and requirements for the planning and control of shipments? 4. How is the current logistics system designed in practice and what are the requirements for the planning and control of shipments and a potential Central Planning Facilitator? 5. How should a Central Planning Facilitator be designed in the Physical Internet

(7)

6. How valid is the design of the Central Planning Facilitator in the Physical Internet and what are the future opportunities for this design?

In parallel with this study, Wissink (2019) takes an operational perspective while designing the Physical Internet system with a Central Planning Facilitator. In her study, focus is more on the actual operations undertaken by the Central Planning Facilitator. This is further discussed in Section 3.1.

Not much literature about the Physical Internet as a hybrid system is available. Ambra et al. (2019) conclude their research by stating that some Physical Internet planning, like often used routes, flows and capacity, could be more feasible if they are handled as in a centralized system. Lafkihi et al. (2019) strengthen this by comparing the planning of shipments of a decentralized Physical Internet system with a centralized Physical Internet system. They found that in a centralized system, the planning of shipments is more effective and efficient. However, in a decentralized system it is more flexible (Lafkihi et al., 2019). Therefore, Lafkihi et al. (2019) state that the performance of the Physical Internet as a decentralized system needs to be further studied, especially by asking the question if a decentralized system is sufficient for the planning and control of shipments. The research that comes closest to this study is the one of Vargas et al. (2019). They studied the combination of decentralized and centralized systems in a Physical Internet context and made an operational design of an independent party that regulates a gain-share business algorithm. This study builds on their research by designing a party which focuses on planning and control of shipments rather than gain-sharing. In addition, this study takes a strategic-business perspective and therefore it also focuses on how the party should be owned and make money in the Physical Internet system.

(8)

2 Theoretical Framework

In this chapter, a theoretical framework is established about the planning and control of shipments within logistics systems and about the Physical Internet. The chapter starts with an overview of the academic literature about the current planning and control of shipments in logistics systems and about how central parties, comparable to the Central Planning Facilitator, operate in these systems. This study takes a strategic-business perspective and therefore, focus is on what services central parties in current logistics system provide and on how they are owned. This literature serves as an important base for the conducted interviews in the practical analysis, which is described in Chapter 3. Afterwards, the chapter continues with a broad explanation about the Physical Internet and its stakeholders. Requirements for the Central Planning Facilitator are identified from the Physical Internet literature and are used in Chapter 5.

2.1 Planning, Control and Collaboration in Logistics Systems

To get a deeper understanding about the planning and control in logistics systems, the current literature is explored. A logistics system can be controlled as a decentralized system, a centralized system or as a hybrid system (L¨oh et al., 2000). A decentralized system is a system that consist of independent self-coordinating units (L¨oh et al., 2000), which plan, control and make decisions based on the optimization of their own goals (Jung et al., 2009). A decentralized system allows for independence and fast and well-informed decisions in a logistics system (Carney, 1995), like real-time routing (Sallez et al., 2016). However, a decentralized system also has disadvantages. For example, all the independent parties first decide based on the optimization of their own goals instead of the common goal of the system (Jung et al., 2009). Therefore, a fully decentralized system can lower the overall quality of decision-making, and it can lead to a distorted distribution of benefits amongst stakeholders (Carney, 1995; Cimon, 2014). In addition, a lack of coordination and communication can lead to an increase in wasteful efforts (Carney, 1995).

A centralized system is a system where a central party manages the coordination of the different units, and plans, controls and makes decisions based on the optimization of the whole system (L¨oh et al., 2000). Centralized systems can minimize the total system’s costs and can optimize shipments based on the goals of the whole system (Ding and Kaminsky, 2019). The central party has to efficiently integrate the different units that are necessary to get products at the right quantities, at the right location, at the right time, with minimal costs (Fu et al., 2014). However, a completely centralized system is in many situations not possible, since units could be independent or competitors, and therefore not willing to collaborate (Heeswijk and la Poutr´e, 2019; Rached et al., 2016). In addition, if the system is operating internationally, it is challenged by differences in politics, norms and culture (Hudnurkar et al., 2014), which makes a completely centralized system difficult to implement.

(9)

A way to plan and control shipments in a hybrid system could be by introducing a central party to the system, which could be similar as the envisaged Central Planning Facilitator in the Physical Internet system. Such a party can support the planning and control of shipments in the whole system, and leave the local planning and control to the individual units in the system (L¨oh et al., 2000; Sahay and Ierapetritou, 2014). However, it should be taken into account that hybrid systems require much data and information exchange, which does not only require networks that are able handle the amount of data, but also collaboration between the units in the system and the central party (L¨oh et al., 2000).

Even though the Physical Internet is considered as a collaborative logistics system (Mon-treuil et al., 2010), not much literature is available about how collaboration is established within the Physical Internet. Collaboration in current logistics systems can be seen as an approach to increase the efficiency of self-coordinating units in a decentralized system (Pom-poni et al., 2015). Many companies realised that it is not sufficient to work alone if they want to move towards a greener supply chain. Therefore, collaboration is already a leading business strategy in today’s world for planning, forecasting, production and replenishment (Ramanathan et al., 2014). For units to participate in a collaboration relationship, it is important that a win-win situation for all the collaborating units is created, which provides them with superior performance, that they would not have reached individually (Barratt, 2004; Pomponi et al., 2015). Most of the current collaboration relationships in logistics systems regarding planning and control of shipments are based on vertical collaboration. Vertical collaboration is a partnership or an alliance between units that operate in differ-ent stages of the supply chain (Ding and Kaminsky, 2019). However, in order to become as efficient as possible, all but the largest firms also need horizontal collaboration (Pom-poni et al., 2015). Horizontal collaboration is a cooperation between different independent units, which provide comparable logistics functions for similar suppliers or customers (Ding and Kaminsky, 2019; Pomponi et al., 2015). Despite the effort, horizontal collaboration is still barely used within decentralized systems, which is due to the lack of trust (Ding and Kaminsky, 2019; Pomponi et al., 2015).

Trust plays a key role in any successful collaboration relationship (Barratt, 2004; Sheu et al., 2006), and a lack of trust is seen as the biggest resistor for collaboration relationships (Salam, 2017). To increase mutual trust, partners must interact continuously and have matching objectives (Lindawati et al., 2014; Pomponi et al., 2015). In addition, Salam (2017) states that internal interests, external forces and a lack of commitment might decrease the degree of collaboration. Furthermore, an unfair distribution of power might decrease the level of trust and subsequently the degree of collaboration between units. Especially when a unit uses it for the purpose of dominance (Pomponi et al., 2015).

2.2 Services of a Central Party

(10)

in a system are willing to collaborate with a central party. Information-sharing is recog-nized as a minimum collaboration relationship between units in a system (Sheu et al., 2006). Sharing information within a system can lead to real-time information for all participating units and therefore an improved decision-making (Minson and Mueller, 2012; Prajogo and Olhager, 2012). However, individual units could not make optimal use of this information because they are too focused on their own views (Minson and Mueller, 2012). Therefore, making decisions together or having one central party making certain decisions on behalf of a selection of units in the system about the planning and control of shipments, could be more efficient for the whole system (Minson and Mueller, 2012). Units in the system can also outsource the decision-making about the planning and control of shipments to a central party in order to get an increase in the efficiency of the planning or to increase for example on-time delivery or customer service (Govindan et al., 2016; Li et al., 2012).

2.3 Ownership of a Central Party

Central parties in logistics systems are established and owned in multiple ways. This could be similar for the envisaged Central Planning Facilitator in the Physical Internet system. Horizontal and vertical collaboration relationships regarding the planning and control of shipments could arise between certain units in the system as a strategic alliance. A strategic alliance is an agreement between two or more units in a system to collaborate or cooperate in an activity, so they gain mutual benefits from each other’s strengths (Isoraite, 2009) and subsequently a competitive advantage (Haeussler et al., 2012). In a strategic alliance, the units can remain independent and in competition (Isoraite, 2009). Central parties can be the result of a strategic alliance, where every unit has a stake in the party. Central parties that arise from a strategic alliance can facilitate the information-sharing. Furthermore, they can facilitate a fair risk- and cost-distribution between the units (Isoraite, 2009). Even though a strategic alliance offers many advantages, there is also a big risk of knowledge leaking beyond the scope of the strategic alliance, intentionally or unintentionally. Contracts could to some extent undermine this knowledge leaking, but it is not the most effective safeguard (Jiang et al., 2013). Therefore, for a strategic alliance to be successful and for units in the system to share information, there should be mutual trust between the units (Jiang et al., 2013).

Another way for central parties to be owned and established is as an independent party, where the units in the decentralized system can outsource activities to. This allows units to optimize their own profit, while collaborative activities with respect to global interest and optimization can be managed by a central party (Heeswijk and la Poutr´e, 2019; Lafkihi et al., 2019). Outsourcing activities are rapidly growing in the current economy, since it allows companies to focus on the core activities of the firm (Globerman and Vining, 2004). A requirement for a unit in the system for such an outsourcing decision is that the net benefits of the planning and control of shipments are maximized with a central party, compared to the planning and control of shipments in-house (Globerman and Vining, 2004). In the current economy, planning and control of shipments can be outsourced to freight forwarders or third or fourth party logistics providers.

(11)

resources from different carriers (Krajewska and Kopfer, 2006). Freight forwarders arrange shipments as a service, by for example booking space on a ship, providing all required documentation and arranging customs (Shang and Lu, 2012). The strategic-business goal of a forwarder is to make profit, by generating the difference between the price the customer has to pay for their service and the costs of arranging the shipment (Krajewska and Kopfer, 2006). Therefore, forwarders should be able optimize the costs of the total routes and schedules off all the transport they have to arrange (Krajewska and Kopfer, 2006). The freight forwarding market is in the current logistics a competitive market, and therefore it is important to manage relationships with customers and carriers, in order to survive and gain competitive advantage (Shang and Lu, 2012).

A third party logistics provider is an independent unit that performs logistic activities for a customer. These activities are often the management of transport and warehousing (Hertz and Alfredsson, 2003; Jung et al., 2009). A freight forwarder arranges the transport, while a third party actually performs the transport. The strategic-business goal of a third party logistics provider is similar to the freight forwarder, but the costs consist of providing the transport instead of arranging the transport. A third party logistics provider adds value to the processes of the customer, by developing skills, competencies and scale and scope ad-vantages (Hertz and Alfredsson, 2003). In addition to third party logistics providers, fourth party logistics providers are emerging in the market, and can be defined as independent, non-asset based integrators that bring together the needs of customers and the resources of third party logistics providers (Win, 2008). Third party logistics providers are asset-based companies, and therefore try to maximise the return of these assets, while the fourth party logistics provider tries to maximise the value for the customer (Win, 2008).

2.4 Scale of a Central Party

Central parties can operate on different scales. By operating on a global scale, a central party can benefit from economies of scale and scope (Meyer and Su, 2015). However, according to Vargas et al. (2018), the incentive to increase efficiency decreases when giving one central party the authority about the planning and control of shipments. Furthermore, in order to adapt to all the different cultures and needs, it is important to be responsive on a local level (Hudnurkar et al., 2014; Meyer and Su, 2015). Therefore, on a global level, the planning and control of shipments should be performed as in a decentralized system. However, on a local level, it is most efficient that one party performs the planning and control of shipments, similar as in a centralized system (Vargas et al., 2018).

2.5 Physical Internet

(12)

extent similar, but instead of using packets encapsulating information and data, it uses modular containers with a container-bound-digital-agent that encapsulates products. The development of these containers, the container-bound-digital-agent and an interconnected logistics system is very dependent on technological, infrastructural and business innovation, since the logistics system has to change severely, in order to become open, global and interconnected (Montreuil et al., 2012; Pan et al., 2017).

Montreuil et al. (2010) defined three key components, which enable the Physical Internet Shipments (PI-Shipment) within the Physical Internet system. These components are the Physical Internet Containers (PI-Containers), the Physical Internet Nodes (PI-Nodes) and the Physical Internet Movers (PI-Movers). Naturally, the owners of these components are three of the main stakeholders of the Physical Internet, which are discussed in Section 2.6. The three main components, and their envisaged role within a PI-shipment are described below.

PI-Containers PI-Containers are used for storage and transportation of products and are specifically developed to efficiently flow through the Physical Internet system (Sallez et al., 2016). They are designed in an universal way, so any organisation is able to handle any PI-Container (Pach et al., 2016). PI-Containers are modularly designed in multiple sizes, so they can be composed of and decomposed to smaller PI-Containers (Montreuil, 2011). Every PI-Container has a unique identifier accompanied with a representing container-bound-digital-agent (Montreuil, 2011; Montreuil et al., 2010). This container-bound-container-bound-digital-agent should be able to communicate and manage a decision-making process (Meyer et al., 2009), which could be achieved by applying the Internet of Things and its technologies on the container-bound-digital-agent (Kim and Tran-Dang, 2019). By improving the decision and communication intelligence capabilities of the agent, the PI-Containers can play an active role in the Physical Internet system (Sallez et al., 2016) and therefore act as a decentralized unit.

PI-Nodes PI-Nodes are the locations designed to perform some kind of operation on a PI-Container, like storing, docking or testing (Montreuil et al., 2010). Especially cross-docking PI-Nodes are crucial in the functioning of the Physical Internet (Chargui et al., 2018; Pach et al., 2016). PI-Containers are transported past multiple PI-Nodes, which support intermodal transport (Montreuil, 2011). A PI-Node needs to be open for any PI-Container from multiple sources and destinations, in contrast to the traditional nodes, which are restricted to a number of partners of the organisation (Pach et al., 2016). Furthermore, Physical Internet aims for universal interconnectivity. Therefore, it is important that high performance PI-Nodes arise that exploit standard protocols, which make interconnecting PI-Containers through different modes and routes fast, flexible, cheap, easy and reliable (Montreuil, 2011; Pach et al., 2016).

(13)

Internet supports this goal, by offering the opportunity for an intermodal system, where rail and sea transport are connected to the road freight, in order to utilize the current capacity as much as possible (Ambra et al., 2019).

PI-Shipment The management of all the shipments within the Physical Internet is a complex subject (Sarraj et al., 2014). The current envisaged process for a shipment within the Physical Internet has several steps. The process starts with the customers, who order products at the suppliers. To ship the products to the customer’s destination, the suppliers request one or more PI-Containers. It is currently envisaged that this is done via a sort of matchmaker platform, which brings together suppliers and PI-Container owners (Meyer and Hartmann, 2019). This platform has to match the requirements of the supplier to the most suitable and available PI-Container. The container-bound-digital-agent of the PI-Container requests transport via a PI-Carrier and then the PI-Container is delivered to the supplier, who loads his products or pallets in the PI-Container (Sarraj et al., 2014). Thereafter, the container-bound-digital-agent of the PI-Container itself decides the best route within the Physical Internet system to arrive at the destination (Sallez et al., 2016). Initially, this is based on the compromise between the costs, time, environmental impact, and the objectives set by the supplier (Sarraj et al., 2014) or the PI-Container owner. Then the route is determined by making use of scheduled transportation (Sarraj et al., 2014). The PI-Container can go through several PI-Nodes and it can be cross-docked several times (Montreuil, 2011). When the PI-Container arrives at its last PI-Node or destination, the products are unloaded from the PI-Container.

2.6 Physical Internet Stakeholders

The stakeholders of the Central Planning Facilitator are important to take into account when introducing a Central Planning Facilitator to the Physical Internet system. Especially since this study takes a strategic-business perspective. Based on the three key components of Montreuil et al. (2010), the study of Meyer and Hartmann (2019) and the study of Hofman and Dalmolen (2019), multiple stakeholders of the Physical Internet and their requirements for a new system are identified.

The Suppliers and the Customers The Physical Internet system enables the trans-portation of physical objects from suppliers to customers. Therefore, the suppliers and the customers are an important stakeholder. Because the supplier often arranges the trans-portation, they forward the customer’s wishes, which are on time delivery and visibility (Meyer and Hartmann, 2019). According to Meyer and Hartmann (2019), suppliers have the highest interest in the Physical Internet and they have to force the carriers and hubs to adapt their business model to the Physical Internet. Even though social and ecological sus-tainable transport is a positive addition, the requirement from the supplier is a short-term cost benefit (Meyer and Hartmann, 2019).

(14)

mar-and Hofman, 2019; Sarraj et al., 2014). One vision, which is assumed in this study mar-and differs from the current market, is that PI-containers are provided by multiple owners in the Physical Internet (Djurica and Hofman, 2019). It is assumed that these owners provide specific containers with a high variety of sizes and purpose. The goal of the PI-Container owner is to maximize its profit, which is associated with carrying many products, having the smartest container-bound-digital-agents and minimizing the depreciation of the PI-Containers (Theofanis and Boile, 2009). Therefore, the PI-Container owner wants to get the opportunity to ship as much products as possible in the PI-Container, and to offer the most efficient routes to customers or suppliers compared to the competition.

The PI-Carriers The PI-carriers are the owners of the PI-Movers. Similar to the goals of the PI-Container owners, the PI-Carriers want to maximize their profit. The PI-Carriers should focus on minimizing transportation and handling costs and maximizing load factors (Theofanis and Boile, 2009). Therefore, they should get the opportunity to optimize their own routes, according to their own needs, assets and schedules.

The PI-Node Owners Important stakeholders, associated with the PI-Node, are the owners of the PI-Nodes. The goal of the PI-Node owner is to maximize profit gained from providing services to the PI-Containers. In addition, the owner wants to have the best and most efficient PI-Node, since it is envisaged that the PI-Nodes are rated on a number of key performance indicators, like service level or speed (Montreuil et al., 2010). Therefore, a requirement from the PI-Node owner is to get the opportunity to optimize the processes in the PI-Node as much as possible.

The Matchmaker The matchmaker platform should bring at least suppliers and PI-Container owners together (Meyer and Hartmann, 2019), which makes it a relevant stake-holder. The objective of the matchmaker is to match as much PI-Containers and suppliers as possible and therefore a requirement from the matchmaker platform is that there is a balance between the number of available PI-Containers and the requested shipments.

(15)

3 Methodology

In this chapter, the research methods applied to this study are described and explained. To answer the research question, a design science research is conducted. A gap is identified within the current research about the Physical Internet between on the one hand the design of the Physical Internet as a decentralized system, and on the other hand the optimiza-tion of the Physical Internet as a centralized system. Therefore, this study explores the Physical Internet as a hybrid system, by introducing the Central Planning Facilitator. A design science research is an appropriate method to answer the research question, since the objective of this study is to improve the current Physical Internet by introducing a new and innovative unit, the Central Planning Facilitator (Hevner, 2004). In this chapter, the re-search setting is first shortly explained. Afterwards, the steps of the design science rere-search are described, which are based on the framework of Wieringa and Heerkens (2007). The first step is Problem Investigation, the second step is System Design and the third step is Design Validation. According to Wieringa and Heerkens (2007), the next steps in a design science research are Design Implementation and Implementation Evaluation. However, the designed system in this study cannot be implemented, since the Physical Internet is still a non-existing system. Therefore, only the first three steps of the framework of Wieringa and Heerkens (2007) are carried out in this study. A rough overview of the steps taken in this study are given in Figure 1.

(16)

3.1 Research Setting

This study takes a strategic-business perspective. A new unit in the Physical Internet system is introduced as the Central Planning Facilitator. This makes it important to have a strategic, top-down understanding of where the Central Planner Facilitator should operate in the system relative to other stakeholders and how it influences these stakeholders (Li et al., 2012). In addition, it is explored how the Central Planning Facilitator can add value to the system as a business, like which service it should provide, how should it make money, how should it be owned and on what scale should it operate (Oktaei et al., 2014). Data for this study is collected in collaboration with another master student, Wissink (2019), who takes an operational perspective while designing the Central Planning Facilitator in the Physical Internet system. Focus in the study of Wissink (2019) is on the actual operations of the Central Planning Facilitator, whereas this study focuses more on the general service it should deliver and how the Central Planning Facilitator could participate in the Physical Internet system as a business. It is recommended that the reader is following both texts to gain a vertical understanding of the problem and the suggested design.

3.2 Problem Investigation

To make a valid design of the Central Planning Facilitator, the problem is first thoroughly investigated. The theoretical framework in Chapter 2 focuses on the literature about the current logistics system designs and on central parties operating in these systems. These central parties could be similar to the Central Planning Facilitator and are analysed from a strategic-business perspective. Furthermore, the theoretical framework presents the avail-able literature about the Physical Internet. The theoretical framework is used to establish the critical requirements for a Central Planning Facilitator from literature, and it is used as a base for the conducted interviews.

Furthermore, a practical analysis is conducted by means of a multiple case-study, since a good design science research identifies opportunities, problems and critical requirements in an actual application environment before making a design (Hevner, 2004). For example, in practice, initiatives for introducing central parties to a decentralized system might already exist, which could be used for the Physical Internet. The cases are selected based on purposive and convenience sampling (Karrlson, 2016). Purposive sampling is used, since this study takes a strategic-business perspective and therefore the main stakeholders of the Physical Internet are desirable to analyse. Convenience sampling is used due to the limited time-span of this study. Six different cases are selected, by taking into account the potential stakeholders of the Central Planning Facilitator in the Physical Internet system. Hubs, road carriers, shipping companies, container owners and forwarders are selected. In addition, two experts are selected that could give valuable insights about designing the Physical Internet as a hybrid system. First, Birgit Hendriks, works at Eco2City and Goederenhubs Nederland and is because of her work for the city hubs highly involved in the developments of the Physical Internet. Second, Patrick Fahim is a PhD student, researching ports in the Physical Internet. The last case that is selected serves as a benchmark for a hybrid planning system. This is Case 7, which combines planning as in a decentralized system with planning as in a centralized system for its operating rooms. The description of the cases can be found in Table 1.

(17)

Case Company size Industry

Case 1 Big International hub

Case 2 Medium Local hub, forwarder and carrier

Case 3 Medium Road carrier: plans and controls

shipments of their customers

Case 4 Medium Road carrier: focuses mainly

on resource planning

Case 5 Big

Shipping carrier, road carrier, container owner, forwarder and

terminal owner

Case 6 Small Forwarder

Birgit Hendriks

Goederenhubs Nederland Small

Small hub and Physical Internet expert

Patric Fahim N.a. Physical Internet Expert,

research in ports

Case 7 Big Plans operating rooms

as in a hybrid system

Table 1: Short Description of the participating cases

way of planning and control of shipments, rather than on the Physical Internet. This is because the Physical Internet is a very complex subject, and hard to explain in an one hour interview. A specific protocol is developed for every interview. This is displayed in Appendix D. In addition, every case is studied thoroughly beforehand. At the beginning of every interview, a short presentation is given about the research and the schedule of the interview. At the six cases that represent the stakeholders, questions are asked about their goals regarding the planning and control of shipments, and about how and when a Central Planning Facilitator could add value. At the two interviews with the experts, questions are asked about the current logistics and about how a Central Planning Facilitator could add value in the Physical Internet system. Case 7 is used as a benchmark for the combination of hybrid systems, and questions are asked why such a system is used and why it is successful or not. All interviews are held in Dutch, since this is the native language of every interviewee. To assure the quality of the data, permission is asked to record the interviews. This reduces the bias of the researcher and it makes sure that the interviews can be transcribed. In addition, the interviews are cross-checked by the cases to assure that all the answers are interpreted correctly.

The collected and cross-checked data is analysed by means of coding, since coding can result in the discovery of different patterns in the data (Karrlson, 2016). The relevant data is la-beled with inductive first-order codes, to get a good overview of the data. Afterwards, these codes are divided in second-order, third-order and aggregated codes, to discover patterns. The aggregated codes are partly deductive codes gathered from the theoretical framework in Chapter 2, but also contain some inductive codes.

3.3 System Design

(18)

used to design the Central Planning Facilitator in the Physical Internet system. Both crit-ical and desired requirements are established, to make a distinction between the essential conditions for the Central Planning Facilitator to be a valuable party in the Physical In-ternet system and the conditions that would increase the Central Planning Facilitator’s value in the system, but which are not necessarily essential. Because this study takes a strategic-business perspective, the design of the Central Planning Facilitator is described partly verbally. Furthermore, the general service which the Central Planning Facilitator should deliver, is modelled in Business Process Modelling and Notation (BPMN). For more detail in the operations of the Central Planning Facilitator, the reader is referred to Wissink (2019). Because this study is an exploratory study, three different scenarios are designed for the Central Planning Facilitator.

3.4 Design Validation

(19)

4 Findings from the Practical Analysis

This chapter describes the findings from the practical analysis, which is discussed in Chapter 3. Different stakeholders, two experts and a benchmark company are interviewed about planning and control in logistics systems and about how and when a central party can add value to a logistics system. Because this study takes a strategic-business perspective, focus is on what services current or future central parties in logistics systems, similar as the Central Planning Facilitator, deliver and how they are owned. The findings are discussed based on the coding tree in Table 2 and Table 3 in Appendix A. Furthermore, the findings are used to establish requirements, which are further described in Chapter 5.

4.1 Planning, Control and Collaboration in Logistics Systems

Systems can be controlled as hybrid systems. This is the case for Case 7, which serves as a benchmark in this study. The different departments within the hospital plan and control their processes as in a decentralized system. However, the operating rooms plan and collaborate with the different departments more as in a centralized system. Planning, controlling and making decisions for the departments as in a decentralized system is more successful, since the system is too complex for a completely centralized system. This is supported by Case 1, Case 2 and Case 3, where it is explained that the planning and control of shipments as in a centralized system is also too complex and decreases the flexibility of the planning. However, planning and control as in a decentralized system also has disadvantages. Case 7 argues for example that it does not allow units to take each others planning and decisions into account, which is why many units have to often change their planning. Therefore, Case 7 states that it would be better to share information as in a centralized system. Birgit Hendriks supports this and states that good decisions in a decentralized system can only be made when the right information is in place. Therefore, it is best to share information as in a centralized system, while simultaneously make decisions as in a decentralized system.

Another example of a hybrid system is Case 3. Case 3 is part of a system with 13 different units, which are all road carriers, and one central party. The central party is the result of a strong horizontal collaboration relationship. All 13 units are located in a different region in the Netherlands. By sharing the shipments from all customers, all units are able to ship products for their customers all over the Netherlands, in a more efficient way. The central party is introduced to provide every unit with the necessary information. In addition, it is allowed to make decisions about for example opening-times or time-slots, which are necessary to share the shipments. All units plan their shipments individually. The goal of the central party is to make the whole system more efficient. Every unit has a stake in the central party and has a representative in the board. The central party can exists because every units pays a yearly amount.

(20)

For some cases, there is a desire to increase the degree of collaboration, especially horizontal collaboration. Case 4 for example argues that they are too small to plan their shipments efficiently and minimize the empty kilometers. Case 5 also states that to be optimal overseas, all shipping companies would have to collaborate. Even though most cases agree that they, and the whole supply chain, would become more efficient when there is more collaboration, they also state that they do not want to collaborate more. According to Case 1 and Case 4, the complexity is in the fact that not everybody is able to win on the short-term. Case 2 for example argues that the risk of increasing the efficiency of the competitor is high when they would collaborate horizontally, which could lead to losing market share. In addition, by increasing the collaboration and information sharing between units regarding the planning and control of shipments, some cases mention the fear of losing the possibility to differentiate themselves on price, quality and efficiency. Case 2 also states that they do not want to collaborate with units they do not know or trust, even though this increases their efficiency.

During the interview with Birgit Hendriks, collaboration is also discussed in combination with the Physical Internet. She argues that collaboration is important in order for the Physical Internet to work. However, in the current literature about the Physical Internet, the human side is often neglected, which is very important for collaboration. Birgit Hendriks states for example:

“Carriers often drive back empty. However, they will not change this, because in order to change it, they would have to collaborate”

4.2 Services of a Central Party

(21)

according to Case 6, Case 5 and Case 2. Furthermore, Patrick Fahim mentioned that it is important that the shared information is reliable and Case 6 stated that the sharing of information needs to be fully secured and good contracts need to be in place.

“I firmly believe that you have to go there [sharing more information], but information that is not characterized as operational, is not information for another party”. (Case 5)

According to Case 1, forwarders could be seen as the current central party in logistics systems. Birgit Hendriks and Case 6 also mention that a future Central Planning Facilitator could provide a similar service as a forwarder. By combining different shipments, a forwarder could collaborate on a higher level with the units and make decisions for the planning and control of shipments. In addition, Birgit Hendriks and Case 6 mention that a Central Planning Facilitator could handle finances, contracts and customer service as additional services. There are a few cases that are open to having a future Central Planning Facilitator make decisions. Case 4 state that they would be open for an independent party to plan their shipments, if it reduces the empty kilometer significantly. As explained in Section 4.1, Case 3 is already collaborating with a central party that has some power to make decisions. However, Case 5 and Case 2 think that decisions about the planning and control of shipments should be made by all the individual units.

Another initiative in the current logistics, which is mentioned by multiple cases, is that there is a platform that allows carriers to collaborate based on capacity. Case 3, and Case 4 both stated they use this platform when they have only a few pallets to ship. They offer these pallets to the platform, and another carrier can ship these pallets for them for a certain price. In addition, when they have to ship a full container to a specific location, but they do not have any shipments in return, they check if another carrier offered pallets on the platform, which they can efficiently ship. Even though this platform makes their planning sometimes more efficient, Case 4 also mentioned that supply and demand are a major downside of such a platform. In busy times, there are no carriers that take the shipments of the platform and in quiet times, there are many carriers looking on the platform, but there are no shipments on it.

4.3 Ownership of a Central Party

The way a central party or a future Central Planning Facilitator is owned, is an important condition for units to start a collaboration relationship with such a party, which is also indicated in Section 4.2. Regardless of the specific ownership type, it is mentioned by all the stakeholders that the owners of a future Central Planning Facilitator should be a neutral and transparent unit in the system and they should not have too much power. Different ownership types are discussed for a future Central Planning Facilitator during the interviews. The Central Planning Facilitator could be similar to a strategic alliance, it could be owned by the government or hubs, or it could be an independent unit. These are discussed below.

(22)

mentioned that decisions of the central party in their system are usually accepted quite fast because every unit is involved in the decision-making. According to Case 1, this kind of ownership can also lead to a future Central Planning Facilitator that is not a neutral unit, since one stakeholder might be more powerful than the other. Furthermore, Case 2 indicate that even though they would represent a part of the board, they are still not willing to share information with such a party, because they are afraid of losing customers.

“When the stakeholders are owners, the party might be a bit more neutral. However, you always have that one stakeholder that is more powerful than the others” (Case 1)

According to Patrick Fahim and Case 1, hubs can also be the owner of a future Central Planning Facilitator. All the logistics flows come together in the hub, and therefore they already have much information from all the different units in the system. However, Case 5 argues against this type of ownership, since a hub is not a neutral unit and not transparent. In addition to hubs, the government is also named as a possible owner of the Central Planning Facilitator. Some cases argue that they are most neutral, but according to Case 6 and Case 5, the market should find a solution, since that is where most knowledge is located.

A future Central Planning Facilitator can also be an independent unit in the system and act as a forwarder according to Case 6, Birgit Hendriks and Patrick Fahim. By combining different shipments, the Central Planning Facilitator can plan and control shipments more efficiently, and by getting discounts, it can make money and provide shipments against a lower price.

4.4 Scale of a Central Party

All cases agree that central parties, or a future Central Planning Facilitator, should operate on a local level. A reason is that on a local level, there is more local and relevant knowledge on how to improve efficiency. However, Case 1 also state that a future Central Planning Facilitator should not be too local, since this decreases the efficiency. Furthermore, a few cases mention that different parts of the supply chain require different ways of planning and control of shipments. Therefore, they think a future Central Planning Facilitator should focus on one part of the chain.

4.5 Stakeholders in the Current Context

It is important to identify the stakeholders and their goals in practice, since this study takes a strategic-business perspective. Furthermore, Case 5 argues that it is crucial for the success of a future Central Planning Facilitator that all the different stakeholders are taken into account. Below, the stakeholders and their goals are discussed.

(23)

frame. This leads to inefficient routes and many empty kilometers and therefore, customers have to be re-educated according to Birgit Hendriks. In addition, Case 5 argues that the requirements from the customer and supplier will shift to real-time information regarding the time of arrival of the shipment instead of having the shipment delivered in a limited time-frame.

Container-Owners According to Case 5, the goal as container-owner is to transport as much products in the container as possible to gain maximum profit. Therefore, the more efficient the shipment moves through the supply chain, the more products can be transported in one container per year. Because container-owners like Case 5 are often also a shipper, they rather use their own containers than the containers of others, leading to more inefficiency.

Carriers There are different types of carriers, for example shippers or road carriers. Case 6 argues that there are a few big shippers, which makes them very powerful and less willing to collaborate. According to Case 5, the goal is to make as much money as possible. However, they also recognize they are in the lead for creating a better world, which is why they are to a certain degree open for collaboration, as long as they are involved in the decision-making. From the cases representing road carriers, it is made clear that road carriers want to make as much money as possible by driving minimal empty kilometers. Therefore, they would be willing to collaborate with central parties or a future Central Planning Facilitator if it reduces the empty kilometers.

Nodes There can be different types of Nodes. In this study only a port is interviewed and therefore their perspective is taken. The goal of ports is to make sure that companies can do a good job and are connected with each other in the port. When a future Central Planning Facilitator can improve the efficiency of the shipments that go through the port, they would be open for collaboration, since this improves the reputation and competitive advantage of their port.

Authorities The authorities are mentioned by multiple cases as stakeholder of a future Central Planning Facilitator. Most cases do not want the authorities to interfere with the planning and control of shipments. However, they could give incentives or change laws to stimulate units to work with the Central Planning Facilitator.

(24)

5 Discussion about the Requirements

This chapter describes the requirements for the Central Planning Facilitator. The Central Planning Facilitator is a central party in the Physical Internet. It is envisaged that the Central Planning Facilitator ensures that the Physical Internet acts as a hybrid system, by delivering a service for the planning and control of shipments for multiple units. The requirements are established from the problem investigation, which contains a theoretical framework in Chapter 2 and a practical analysis in Chapter 4. The requirements are based on a strategic-business perspective. Therefore the requirements regarding collaboration, service, ownership, scale and the stakeholders are discussed below. A distinction is made between the critical and desired requirements. Critical requirements are the essential con-ditions for the Central Planning Facilitator to be a valuable party in the Physical Internet system. Desired requirements are conditions that could increase the value of the Central Planning Facilitator, but are not necessarily essential for the Central Planning Facilita-tor to be a valuable party in the Physical Internet system. All the requirements are also summarized in Table 4 in Appendix B.

5.1 Requirements for Collaboration with the Central Planning Facilitator

In order for the Central Planning Facilitator to be an additional valuable party in the Physical Internet system, it is important that other units in the system collaborate with it. Even though the established requirements for collaboration with the Central Planning Facilitator are all complex requirements, it is important to consider them while designing a Central Planning Facilitator, since the Physical Internet is envisaged as a collaborative logistics system (Montreuil et al., 2010). A first critical requirement is that the Central Planning Facilitator should be able to create a win-win situation on the short-term for each unit. This is mentioned by multiple cases and it is in agreement with the academic literature about collaboration, where Pomponi et al. (2015) states that it is required that units are provided with superior performance when they collaborate, which they are not able to reach individually. A second critical requirement, which is mentioned by almost every case in the practical analysis, is that units should trust the Central Planning Facilitator in order to collaborate with it. In addition to the first critical requirement, this is also in agreement with the academic literature, where a lack of trust is described as the biggest resistor for collaboration (Barratt, 2004; Salam, 2017; Sheu et al., 2006). A third desired requirement is that the Central Planning Facilitator gives units in the Physical Internet system the possibility to differentiate themselves on for example price, quality or efficiency. Therefore, it is important that the Central Planning Facilitator does not become too powerful, and supports the competition between the units. A fair distribution of power might also increase the level of trust between the units in a system according to Pomponi et al. (2015). This is not explicitly named during the interviews in the practical analysis. However, it is still considered important. Therefore, a fourth desired requirement is that the Central Planning Facilitator does not become too powerful.

5.2 Requirements for the Service of the Central Planning Facilitator

(25)

handled, stored and transported in the Physical Internet (Pan et al., 2017). In addition, there are multiple requirements established from the practical analysis regarding the service the Central Planning Facilitator should provide. Regardless of the type of service, a second critical requirement for the service is that the Central Planning Facilitator is transparent about its decisions and actions during the service. This is mentioned by a few cases as very critical in order to use the service of the Central Planning Facilitator. Furthermore, a third critical requirement, mentioned by Case 5, is that the Central Planning Facilitator should be open for every unit in the system. This is an important requirement, since the Physical Internet is envisaged as an open system, where for example PI-Nodes, PI-Carriers, and thus Central Planning Facilitators, should not be restricted to a few units (Montreuil, 2011; Pach et al., 2016). A fourth desired requirement, which is also indicated by Case 5, is that every unit should be able to decide themselves if and when they want to start using the service of the Central Planning Facilitator.

The Central Planning Facilitator can deliver multiple services, ranging from only informa-tion sharing to making decisions for units regarding the planning and control of shipments. A fifth critical requirement for the service of the Central Planning Facilitator is that it provides the different units in the system with information and a sixth critical requirement is that this information only contains operational information. It is mentioned by almost every case in the practical analysis that they would not be willing to share more than op-erational information with any unit. In the Physical Internet, much attention is already paid to information flows. It is envisaged that information is protected by an encryption key, and only operational is visible without this key (Montreuil et al., 2010). The Central Planning Facilitator should comply with this and make sure it tries to limit its information sharing service to operational information. A seventh desired requirement, mentioned by a few cases in the practical analysis, is that the Central Planning Facilitator should pro-actively share real-time information with the different units in the system as a service. An eight desired requirement is that the Central Planning Facilitator creates data standards in the system. This is in line with the Physical Internet, where using standard protocols is one of the foundations of the envisaged Physical Internet system (Montreuil et al., 2012). A ninth desired requirement is that the Central Planning Facilitator fully secures the sharing of information and that it puts contracts in place to prevent information from leaking.

(26)

5.3 Requirements for the Ownership of the Central Planning Facilitator

From the practical analysis can be concluded that ownership is a very important aspect to take into account while designing the Central Planning Facilitator, since many cases indi-cated that this influences their decision to collaborate with the Central Planning Facilitator severely. A first critical requirement is that the owner of the Central Planning Facilitator is a neutral unit in the system. It should for example not be owned by a hub, since mul-tiple cases in the practical analysis believe a hub is not a neutral party. A second critical requirement, which is also stated in Section 5.2 for the service of the Central Planning Fa-cilitator, is that the owner should be a transparent unit in the system. Furthermore, there are some requirements which are specifically mentioned for when the Central Planning Fa-cilitator is the result of a strategic alliance and therefore owned by all the stakeholders. All stakeholders should be able to correct each other in a strategic alliance, which is a third desired requirement. In addition, a fourth desired requirement is that the Central Planning Facilitator behaves towards the common interest.

5.4 Requirements for the Scale of the Central Planning Facilitator

A first critical requirement regarding the scale of the Central Planning Facilitator is that it should operate on a local level. This is indicated by multiple cases in the practical analysis. The reason for this is that it is too complex to provide a service for the planning and control of shipments on a level that is more global. In addition, this allows the Central Planning Facilitator to be responsive, which is in agreement with the academic literature. By providing a service on a local level, the Central Planning Facilitator can adapt to all the different cultures and needs (Hudnurkar et al., 2014; Meyer and Su, 2015). However, it is also mentioned in the practical analysis that the service of the Central Planning Facilitator should not be too local. Therefore, this is classified as a second desired requirement for the scale of the Central Planning Facilitator. A third desired requirement is that the Central Planning Facilitator should focus on one part of the chain. This requirement is identified during the practical analysis and dependent on the service the Central Planning Facilitator provides. Especially when the Central Planning Facilitator makes decisions regarding the planning and control of shipments, it should focus on a part of the chain. Otherwise, the processes of the Central Planning Facilitator might get too complex.

5.5 Additional requirements from the Stakeholders

During the interviews in the practical analysis, it is stated that the Central Planning Fa-cilitator should take into account all its stakeholders. This is classified as a first critical requirement regarding the stakeholders of the Central Planning Facilitator. In addition, there are many specific requirements from the different stakeholders, which are to a cer-tain degree also dependent on the service the Central Planning Facilitator provides in the Physical Internet system. Therefore, the requirements below are all classified as desired requirements.

(27)

sustainable benefits. Furthermore, from the practical analysis, it is established that the Central Planning Facilitator should support shipments to be delivered in a certain time frame. Even though this decreases the efficiency of shipments, it is highly valued by cus-tomers and suppliers. A third desired requirement is that the Central Planning Facilitator should support the real-time information sharing with customers and suppliers. One case mentioned that this is getting increasingly important.

PI-Container Owners/PI-Carriers/PI-Node Owners The Container owner, PI-Carrier and PI-Node owner are important stakeholders to consider, since these are the owners of the three main components of the Physical Internet (Montreuil et al., 2010). A desired requirement from the PI-Container owners is that the Central Planning Facilitator enables the PI-Container owner to ship as much products in its PI-Container, so they gain maximum profit. A desired requirement from the PI-Carriers is that the Central Planning Facilitator gives the PI-Carriers the opportunity to maximize the load factors of their vehicles. This allows the PI-Carriers to earn as much money as possible. Both requirements are established from the practical analysis in Chapter 4 and the theoretical framework in Chapter 2. A desired requirement from the PI-Node owners is that the Central Planning Facilitator gives the PI-Node owners the opportunity to optimize the processes in their PI-Nodes as much as possible, since this can create a competitive advantage. This requirement is established from the practical analysis in Chapter 4 and it is in agreement with the academic literature about the Physical Internet, since it is envisaged that PI-Nodes are rated on a number of key performance indicators (Montreuil et al., 2010).

Matchmaker A desired requirement from the matchmaker is that the Central Planning Facilitator supports a balance between the number of available PI-Containers and the re-quested shipments. The matchmaker is an envisaged party in the Physical Internet. There-fore, this requirement is established from the theoretical framework in Chapter 2, and no requirements are collected from the practical analysis.

(28)

6 Design

This chapter describes the design of the Central Planning Facilitator. It is envisaged that the Central Planning Facilitator ensures that the Physical Internet acts as a hybrid system, by delivering a service for the planning and control of shipments for multiple units. Three scenarios are designed, which are the Central Planning Facilitator as information facilitator, forwarder and capacity platform. The scenarios are designed from a strategic-business perspective. The functions of the service that the Central Planning Facilitator provides and the way the Central Planning Facilitator can make money are modelled in Business Process Modelling and Notation (BPMN). In addition, it is described for every scenario how the Central Planning Facilitator is owned and on which scale it should provide the service. The scenarios are designed while carefully taking into account the stakeholders and their goals. The scenarios are based on initiatives in the current logistics systems, which are discussed in Chapter 4, and on the critical requirements, which are discussed in Chapter 5. At the end of this chapter, the three scenarios are shortly compared and discussed.

6.1 The Central Planning Facilitator as Information Facilitator

The first scenario that is designed and described in this study is that the Central Plan-ning Facilitator provides a service where it facilitates the information sharing between the different units in the Physical Internet system. In the practical analysis in Chapter 4, it is described that there are already initiatives in the current logistics system where parties act as information facilitators, which makes this scenario a realistic possibility for the Cen-tral Planning Facilitator. A CenCen-tral Planning Facilitator as information facilitator could be a valuable addition to the Physical Internet system, since this allows for information availability as in a centralized system, while all the units in the system are still able to act as self-coordinating units and are allowed to make their own decisions regarding the planning and control of shipments. This is a desired requirement. In addition, the Central Planning Facilitator as information facilitator supports the goal of the Physical Internet, to improve the way physical objects are handled, stored and transported and it aims to create a win-win situation for every collaborating unit.

It is important that the Central Planning Facilitator only facilitates operational informa-tion, since this is a critical requirement for the service of the Central Planning Facilitator. Therefore, the Central Planning Facilitator should only distribute information about for ex-ample schedules. In addition, the Central Planning Facilitator could also share information pro-actively, like real-time arrival times. This is a desired requirement. The functions of the service that the Central Planning Facilitator delivers are modelled in BPMN and described step by step. Afterwards, the ownership and the scale of the Central Planning Facilitator are discussed.

(29)

In order for the Central Planning Facilitator to act as an information facilitator within the system, it is important that the Central Planning Facilitator continuously collects the right information. It is envisaged that the Central Planning Facilitator does this by requesting it from the right stakeholders as in Figure 2. This process goes as follows:

1. The Central Planning Facilitator requests information from the stakeholder.

2. The stakeholder receives the request and discusses within the firm if it wants to share certain information with the Central Planning Facilitator.

3. After the discussion within the firm, the stakeholder comes to an answer, which it sends to the Central Planning Facilitator. This can be either a rejection of the request, or the stakeholder can accept the request.

4. When the stakeholder rejects the request, the process stops after it informs the Central Planning Facilitator.

5. When the stakeholder accepts the request for sharing certain information, the stake-holder and the Central Planning Facilitator start negotiating the terms and conditions under which the stakeholder is willing to share the information, and what the Central Planning Facilitator is willing to give for the information. It should be discussed if the information sharing is only once, or if the stakeholder will provide the Central Planning Facilitator with a certain type of information on for example a daily basis. Even though stakeholders need to have a say in the terms and conditions under which they are willing to share information, it is also important that the Central Planning Facilitator standardizes this process as much as possible with protocols, since this is a desired requirement. It is assumed that the Central Planning Facilitator is transpar-ent to every stakeholder about what happens to the information and how it facilitates the safe information sharing between the right units in the system, since this is a crit-ical requirement. Furthermore, it is important that the Central Planning Facilitator treats all the stakeholders equally.

6. After the negotiation, the process can go in two ways. When the negotiation is unsuccessful, the stakeholder still rejects the request and the process stops after the stakeholder informs the Central Planning Facilitator.

7. When the negotiation is successful, the Central Planning Facilitator sends a contract where all the terms and conditions about the information sharing are discussed in detail. In this contract, it is for example specified who can receive certain information and who cannot. This contract is important, since it assures the safe handling of the information, which is a desired requirement.

8. After both parties signed the contract, the stakeholder sends the information to the Central Planning Facilitator, who stores it in its database and subsequently uses it to provide other stakeholders or units in the system with valuable information that increases the efficiency of their processes.

(30)

The Introduction of the Central Planning Facilitator into the Physical Internet system, from a Strategic-Business Perspective.