The role of resilience in critical infrastructures: Findings from
an in-depth case-study
Final thesis
REPAIR
Master thesis, MSc Supply Chain Management University of Groningen, Faculty of Economics and Business
Author: Laura Hendriks
Student number: 3026663
Abstract
Content
Abstract 1 Introduction 3 Theoretical background 5 Resilience 5 Resilience in CI 9 Theoretical framework 13 Methodology 14 Research design 14 Case research 14 Data collection 15 Data analysis 16 Findings 17 Collaboration 18 Visibility 20 Flexibility 22 Velocity 23 Discussion 25Tools from literature 25
Tools from both literature and the research 26
Tools derived from research 29
Conclusion 31
Managerial implications 31
Limitations and research implications 31
Resources 33
Introduction
Critical infrastructures (CIs) are essential to society, because they can enhance quality of life, provide public services and spur economic growth (Boin & Mcconnell, 2007). However, CIs are also prone to disruptions. In addition to disruptions caused by natural hazards and human failure, CIs are also a target for terrorist attacks (Boin & Mcconnell, 2007). Effects of disruptions in CIs can amplify geographically on a large scale, and affect many external stakeholders like households, cities, other CIs and even other countries (Ouyang, 2014). CIs and other countries are more prone to being affected by a disruption because all CIs are interacting (interactive complexity); when one CI is disrupted, it will cause a ripple or domino effect (depending on the type of disruption), affecting other CIs and countries. There is no literature established about resilience in CIs, despite the importance of CIs, its vulnerability, and its impact when disrupted (Ouyang, 2014).
It is difficult to predict when disruptions occur, therefore, it is important that organizations (like CIs) adopt resilience (Boin & Mcconnell, 2007). Supply chain resilience (SCRes) allows organizations to solve disruptions as quickly as possible by proactively detecting risks in order to reduce vulnerability (Jüttner & Maklan, 2011), react to disruptions to help the organization in its recovery to its original state, and develop its adaptive capacity to make the supply chain more reactive to future disruptions and unexpected events (Jüttner & Maklan, 2011; Ponomarov & Holcomb, 2009). This requires the whole supply chain to look for capabilities to adapt, grow and survive when a (potential) disruption is detected (Scholten & Schilder, 2015).
There is a multitude of articles describing CIs and their characteristics, however, there is no consistency about those characteristics. This lack of consistency exists due to the different kinds of CIs and countries they are providing to. Therefore, there is no overarching research available about how to apply SCRes in CIs conform current articles. Nevertheless, many researchers recognize the need of resilience in CIs. In order to research how resilience is applied in CIs, it is important to establish which characteristics of CIs influence resilience. When the conceptualization of resilience is matched to the characteristics of CIs, we are left with only one question; How do CIs build resilience?
Theoretical background
This chapter will give an overview of the variables derived from SCRes found in literature. A literature review will provide characteristics applicable to all CIs. The relationship between these characteristics and the resilience variables are visualized in a theoretical framework.
Resilience
Events which hamper the flow of goods or services in supply chains are defined as supply chain disruptions (Ambulkar, Blackhurst, & Grawe, 2015). These disruptions can have a negative impact on firms in the supply chain and on society, such as economic consequences (Brandon‐Jones, Squire, Autry, & Petersen, 2014). The consequences of disruptions can be diminished by adopting SCRes (Christopher & Peck, 2004). While there are multiple definitions for SCRes (Ambulkar et al., 2015) the most general definition of SCRes is: the adaptive capability of a firm’s supply chain to prepare for unexpected events, respond to disruptions, and recover from them in a timely manner by maintaining continuity of operations at the desired level of connectedness and control over structure and function (Ponomarov & Holcomb, 2009).”
The conceptualization of SCRes is explained by multiple researches. For example, Christopher and Peck (2004) state that SCRes should derive from re-engineering, agility, collaboration and a supply chain risk management culture. The collaboration, visibility, flexibility and velocity elements from Jüttner and Maklan (2011) overlap the framework of Christopher and Peck (2004) partly, since Christopher and Peck (2004) conceptualize resilience regarding collaboration and agility (visibility and velocity). Melnyk et al. (2003) and Wieland and Wallenburg (2013) create a view which combines reactive and proactive frameworks. One states that resistance and recovery define SCRes (Melnyk, Closs, Griffis, Zobel, & Macdonald, 2014), whereas the other states that robustness and agility define SCRes (Wieland & Wallenburg, 2013). All conceptualizations overlap to a certain extend. Scholten and Schilder (2015) recognized this and did research into several SCRes conceptualizations. They found that the conceptualization of Jüttner and Maklan (2011) was most applicable compared to the other conceptualizations. In addition, literature review indicated that the conceptualization of Jüttner and Maklan seems well-defined compared to the other conceptualizations. This research will apply the SCRess conceptualization of Jüttner and Maklan (2011), due to its applicability and usability. The elements of this conceptualization are outlined in more detail in the following sections and in table 1. Collaboration
principle of collaboration regarding SCRes is that information sharing can reduce uncertainty (Christopher & Peck, 2004). Other concepts regarding elaborative supply chain collaboration include several constructs; namely goal congruence, joint knowledge creation among independent partners, decision synchronization, incentive alignment, collaborative communication and resource-sharing (Cao, Vonderembse, Zhang, & Ragu-Nathan, 2010). Collaboration applies to the reactive and proactive phase of resilience, hence the preparedness, response and recovery (Scholten & Schilder, 2015). Close collaboration with both upstream and downstream partners pro-actively reduces risk by sharing market trends and risk perceptions (Christopher & Peck, 2004). Synergies among partners, information sharing, and joint planning enable a supply chain to prepare for disruptions and quickly respond to disruptions (Christopher & Peck, 2004; Scholten & Schilder, 2015). Collaboration ratifies recovery from and the impact of disruptions by enabling joint planning and information-sharing (Christopher & Peck, 2004; Scholten & Schilder, 2015).
Close collaboration with both upstream and downstream partners pro-actively reduces risks by sharing market trends and risk perceptions (Christopher & Peck, 2004). Nevertheless, it is not always possible, desirable or efficient to establish long term supply chain collaboration. Collaboration is not always perceived as efficient, and some researchers even state that collaboration reduces the resilience of a system (Vugrin, 2016). Collaborating with a high number of external stakeholders increases internal complexity of the focal firm, making the firm less responsive to disruptions (Christopher & Peck, 2004). For example, for a small organization producing standard products it is difficult to collaborate with a large organization, since it is inefficient and time consuming for the large organization (Kraljic, 1983). Large organizations are often not willing to spend a lot of time in collaborating with small stakeholders since it simultaneously makes management more complex and expensive (Kraljic, 1983). Lack of or limited collaboration means that both organizations do not share information with each other, this can result in supply distortions (Lee, Padmanabhan, & Whang, 1997). In addition, potential risks will not be shared, which can lead to supply chain disruptions.
Visibility
Visibility is achieved by information sharing on internal and external levels regarding up- and downstream inventories, demand and supply levels, and production and purchasing schedules to enhance transparency in the supply chain (Brandon‐Jones et al., 2014; Christopher & Peck, 2004). Transparency includes (timely) information sharing between stakeholders in the supply chain about transportation, risks and events in the environment of the organizations (Jüttner & Maklan, 2011). Sharing this information with external partners can result in a reduction of the negative effect of demand distortion, strategic value creation, more organizational agility, and improvement of operational planning and efficiency (Brandon‐Jones et al., 2014). Supply chain intelligence is increased when supply chain partners share the abovementioned information, resulting in higher supply chain visibility (Christopher & Peck, 2004). Confidence and decision-making are improved by supply chain visibility, the same accounts for the reduction of interventions (Brandon‐Jones et al., 2014). Hence, visibility is closely related to collaboration due to information sharing and decision making (Brandon‐Jones et al., 2014).
Flexibility
Organizations are able to handle high levels of uncertainty due to flexibility (Scholten & Schilder, 2015), as flexibility is the ease with which an organization can adapt to instant changes caused by disruptions (Jüttner & Maklan, 2011). Therefore, flexibility provides the opportunity to adapt to disruptions and to ensure continuity as it allows several options during a disruption (Christopher & Peck, 2004). Flexibility makes an organization more resilient by enabling a quick and efficient response to market changes (Ponomarov & Holcomb, 2009).
Placing adaptive capacity and/or inventory at strategic points, can increase resilience by enabling extra production or delivery from stock when a disruption occurs (Christopher & Peck, 2004). Ways to achieve this kind of flexibility are; changeable delivery contracts, adaptive manufacturing facilities, and a multiskilled workforce (Scholten & Schilder, 2015). However, surplus capacity is more preferred than extensive final product inventory since adaptive capacity is more flexible (Christopher & Peck, 2004). This might not be the cheapest supply chain strategy, because maintaining extensive capacity or stock has high costs (e.g. storage costs and maintenance costs), and require multiskilled or temporary employees. These employees also require extra training, and additional suppliers or stakeholders require more management, which makes the supplier base more complex and costly (Dyer & Hatch, 2006). As such, while flexibility can increase effectiveness to deal with disruptions, it should be in equilibrium with efficiency (Scholten & Schilder, 2015). Flexibility can influence velocity, by speeding up recovery due to efficiency but can also lower velocity when (back-up) suppliers are located farther away compared to the original supplier (Jüttner & Maklan, 2011).
Velocity is defined as ‘end-to-end’ pipeline time, so the time it takes to create raw material into the final product (Christopher & Peck, 2004; Jüttner & Maklan, 2011). It is determined by loss per time unit or lead-time (Christopher & Peck, 2004; Jüttner & Maklan, 2011). Velocity is applied to all three stages of disruptions, namely preparedness, response and recovery (Jüttner & Maklan, 2011). Having high velocity makes a supply chain more responsive to disruptions and therefore more resilient. When consciously combining velocity with flexibility, organizations will be more prepared for disruptions, in addition, during a disruption, velocity can provide first mover advantages (Jüttner & Maklan, 2011). However, disruptions can also lower velocity because the flow of goods is hampered, which slow down processes (Christopher & Peck, 2004).
To increase velocity and the speed of recovery, an organization should focus on the ability to cope with short term changes and rapid responses in terms of delivery when selecting suppliers (Christopher & Peck, 2004). In addition, there are three approaches an organization can apply to increase velocity, namely reduce lead-times, streamline processes and reduce non-value-added time (Christopher & Peck, 2004). However, reducing non-value adding time (inventory or external unused capacity) can be contradicting to flexibility, hence an organization should find an equilibrium. This results in quicker responses regarding market changes and disruptions, making the organization more resilient (Jüttner & Maklan, 2011; Scholten & Schilder, 2015)
Recap
The concepts described above are briefly defined in table 1, the table also contains the corresponding tools organizations can adopt to increase SCRes regarding those elements.
Element Definition Tools
Collaboration Is about generating high quality and relevant information sharing which can be obtained through a supply chain community as it enables the exchange of information between members.
Information sharing Goal congruence Joint knowledge creation Decision synchronization Incentive alignment
Collaborative communication Resource sharing
Visibility The ability to see ‘from one end of the pipeline to the other end’ Information sharing Flexibility Ease with the number of possible states and changes a supply
chain can cope with. Adaptive capacity Inventory
Changeable delivery contracts Adaptive manufacturing facilities Multiskilled workforce
Velocity The pace flexible changes are adapted in all stages of a
disruption Reduced lead-times Streamlined processes
Non-value-added time reduction
Table 1 Elements of SCRes (Brandon‐Jones et al., 2014; Christopher & Peck, 2004; Jüttner & Maklan, 2011; Scholten & Schilder, 2015)
failure since their goods have limited shelf life. Inventory in this sector is very limited, and flexibility is difficult to achieve (Stone & Rahimifard, 2018). Another sector is the service sector, which is limited in creating inventory and therefore less prepared for disruptions due to limited flexibility (Boin & Mcconnell, 2007; Cho, Lee, Ahn, & Hwang, 2012). It appears that resilience is both positively and negatively influenced by characteristics of CIs (Ouyang, 2014), which will be described below.
Resilience in CI
The President’s Commission on Critical Infrastructure Protection (PCCIP) defines CIs: ‘a network of independent, most privately-owned, man-made systems and processes that function collaboratively and synergistically to produce and distribute a continuous flow of essential goods and services’ (Ouyang, 2014; 44). However, not all researchers agree with this definition and identify other characteristics of CIs, therefore, table 2 depicts unique elements which distinguish CIs from production supply chains identified in the literature (mentioned by three or more authors to ensure the elements are established and not specific for only one type of CI). Table 2 indicates that in addition to the PCCIP definition, CIs can be identified as vulnerable, essential, governmental services on which society is dependent (Johansson, Hassel, & Zio, 2013; Labaka, Hernantes, & Sarriegi, 2016; Urciuoli et al., 2014; Zebrowski & Sage, 2017). It appears that the definition of the PCCIP finds that CIs are privately owned, whereas literature indicates that CIs are governmental organizations. This indicates that the type of ownership of a CI depends on the type of CI and/or its country of origin (Zebrowski & Sage, 2017). The characteristics found in literature combined with the definition of the PCCIP provide a complete definition which is suitable for this research. Examples of CIs are highways, water transport, the electricity, gas network, and more (Johansson et al., 2013; Labaka et al., 2016; Zebrowski & Sage, 2017).
Zebrowski & Sage (2017)
Labaka, Hernantes & Sarriegi (2016)
Urciuoli, Mohanty, Hintsa & Boekesteijn (2014)
Johansson, Hassel & Zio (2013)
Reflect, reinforce, progress
divisions x x x
Reconfigure political, economic and social relations x x x Vulnerability x x x Society is dependent on it x x x x Essential services x x x Governmental x x x x
Table 2 CI literature review
A review of the literature allowed me to identify that CI supply chains are coping with a multitude of characteristics which influence the adoption of resilience. However, the characteristics of CIs influencing resilience are complex to define, as almost every research applies different characteristics of CIs. The method for defining CIs through a literature review is also applied to find overarching characteristics of CIs. Hence, characteristics which are mentioned in three or more articles are included in this research, since they are more established and not merely applicable to one type of CI but to (almost) all kinds of CIs. Research indicates that external stakeholders appear to be an important part of CIs. Literature shows there are a multitude of constructs (e.g. government, society, private owners, organizations and suppliers) influencing the adoption of resilience (Johansson et al., 2013; Labaka et al., 2016; Urciuoli et al., 2014; Zebrowski & Sage, 2017). Another characteristic entails the interdependencies of CIs and other systems (Johansson et al., 2013; Labaka et al., 2016; Urciuoli et al., 2014; Zebrowski & Sage, 2017), in this research referred to as ‘interactive complexity’. Impact at CIs is almost always large because CIs are interconnected with other CIs. For instance, when the electricity net is disrupted, internet is also disrupted, the same accounts for railroad crossings and traffic lights, which makes the impact and time till recovery grow (interactive complexity) (Johansson et al., 2013; Labaka et al., 2016; Ouyang, 2014). The final characteristic mentioned by several researches is the geographic spread, since CIs are often spread over at least a whole country (Johansson et al., 2013; Urciuoli et al., 2014; Zebrowski & Sage, 2017). This does not only entail the range in which a network is spread but also the physique of a network (e.g. cables, asphalt, above or underground) (Boin & Mcconnell, 2007; Ouyang, 2014). Summarizing, the overarching variables found and applied in this research are ‘external stakeholders’, ‘geographic spread’, and ‘interactive complexity’. A brief definition and the corresponding constructs of those characteristics are described in table 3.
Table 3 Variables and constructs of CI
According to Ouyang (2014), a complete overview of CI supply chains can be obtained by measuring the external stakeholders, geographical spread and interactive complexity. This research combines the findings of Ouyang (2014) with the findings of Labaka et al. (2016), Urciuoli et al. (2014), Johansson et
Variable Construct
External stakeholders
(Johansson et al., 2013; Labaka et al., 2016; Urciuoli et al., 2014; Zebrowski & Sage, 2017)
Are groups or individuals who influence an organization regularly from outside the organization (Franklin, 2001). Examples of external stakeholders are given in the next column. Government Society Private owners Organizations EU Suppliers Interactive complexity (Johansson et al., 2013; Labaka et al., 2016; Urciuoli et al., 2014; Zebrowski & Sage, 2017)
Is the complexity the supply chain experiences due to (obligated) interaction with the environment or different systems (Labaka et al., 2016; Urciuoli et al., 2014)
environment
Private vs. public sector Interdependent systems
Geographic spread
(Johansson et al., 2013; Urciuoli et al., 2014; Zebrowski & Sage, 2017)
Is the range and physique of the network, which differ for every type of CI and can even be influenced geographically by regulations.
al. (2013) and Zebrowski and Sage (2017) to create a comprehensive measure for measuring resilience at CI supply chains.
External stakeholders
External stakeholders are groups or individuals who influence an organization regularly from outside the organization (Franklin, 2001). There is a multitude of parties which could be external stakeholders in CIs, like the government, maintenance teams, private owners, and more (Ouyang, 2014). However, the external stakeholders are not the same for every CI. For example, when private owners are involved as external stakeholders it relates to networks which cross privately owned land (e.g. waterpipes) and not government land like highways. Other stakeholders influencing CIs are the government, because CIs have to follow the country’s regulations, and overarching legislation, for instance law created by the European Union (Urciuoli et al., 2014). Nevertheless, not all external stakeholders are unavoidable or obligatory. CIs can choose how many suppliers they prefer, they can select sustainable suppliers or suppliers from certain countries. The multitude of stakeholders and choices coming with them (e.g. sustainability or political preference) complicate decision making due to the multiple parties to take into account (Ouyang, 2014).
A high number of external stakeholders seems to increase internal complexity by creating more work and requiring more trained people compared to a low number of external stakeholders. Although a high number of external stakeholders can increase flexibility by providing multiple options to the organization (Kraljic, 1983), it decreases velocity, collaboration and visibility by the increased complexity (Ouyang, 2014).
Interactive complexity
(external) stakeholders (Truscott, 2009). This will decrease the internal and external complexity coming with those interdependencies (interactive complexity).
Visibility decreases when there is more interactive complexity, since it increases the possibility of a disruption by risks other CIs face (Ouyang, 2014). At the same time, not all interdependent relationships are visible or familiar to the focal firm but only emerge and become visible during disruptions (Ouyang, 2014). The previously mentioned commissions (e.g. PCCIP and European Union) provide standards and policies to increase velocity between interacting CIs to lower the risks and complexity coming with those ‘invisible relationships’ (Ouyang, 2014). Truscott (2009) states that Europe should apply legislation which (1) obliges CIs to communicate updates and (2) to share information to all CIs, as it appears to be effective in the USA. This and other legislation should increase visibility throughout the supply chain (Truscott, 2009).
Geographic spread
of CIs (Ouyang, 2014). The network can be processed in different ways, like above ground or underground (highways and electricity network), type of vehicle (emergency services), and more. For example, it is more difficult to repair an infrastructure underground than above ground due to buildings on top of it and due to digging to reach the disrupted part (Boin & Mcconnell, 2007), which decreases velocity and visibility. Hence, CIs should decide the most beneficial material and processing regarding (development of) the infrastructure to increase resilience.
Geographic spread decreases visibility and increases flexibility, velocity is either increased or decreased depending on how the network is managed. The way in which a supply chain is spread geographically, either centralized or decentralized, can also influence disruptions and resilience (Truscott, 2009). A centralized network has a direct negative relationship on velocity since distance decreases velocity and functionality (Ouyang, 2014). In addition, visibility is influenced by the number of nodes in a network, a high number of nodes and stations (decentralized network) decreases visibility. Visibility is important when a disruption occurs, since it will clarify how many customers are harmed but also clarify the unavailability of nodes (the higher amount of damaged nodes, the longer it takes to repair) (Ouyang, 2014; Truscott, 2009). Hence geographic spread influences resilience because it can influence visibility and velocity.
Theoretical framework
The literature describes resilience and its relation to CIs, research shows that there are three characteristics of CIs which influence the adoption of resilience (external stakeholders, interactive complexity and geographical spread). However, literature does not describe in what way resilience should be adopted according those variables. Therefore, a theoretical framework is conducted to visualize which elements influence resilience and which influence CIs. As figure 1 implies, resilience will have to adapt according the characteristics of CIs. Hence, CIs are the independent variable since the performance of resilience (dependent) is adjusted to the variables of CIs, resulting in the research question ‘How can CI providers build SCRes?’
Methodology
Research design
To find the relationship between resilience and CI empirically, a single case study is conducted at organization A. A case study is most suitable for this research because through case study an organization can be studied in its natural setting (Voss, Johnson, & Godsell, 2016). An understanding of how CIs cope with disruptions will develop current SCRes theory. In addition, a case study is very suitable for exploratory research where variables are still unknown (Voss et al., 2016). Case study allows the researcher to ask the how, why and what questions to increase understanding of the complexity and nature of a disruptions and its solutions. To find how CI providers build resilience, a single case study is conducted to get more in-depth information of one CI. Interactive complexity from the point of view of the focal firm can only be found by a single case study of the focal firm. Longitudinal research often applies to a single case study to obtain a clear overview of the event and its progress. The research conducted includes an interview entailing questions of previous disruptions and how they are solved. This enables measuring how the organization adapts over a long time period regarding disruptions, hence this research is longitudinal (Voss et al., 2016). In addition, this research is an exploratory research since no theory regarding resilience in CIs is established yet. Exploration of the topic begins by creating an in-depth broad overview of one supply chain (Voss et al., 2016). Hence, the longitudinal aspect and the exploratory focus of this research require a single case study (Voss et al., 2016). The CI of this research is suitable for a single case study since they have a clear supply chain and there are no similar electricity organizations operating at the level of this organization in the Netherlands.
Case research
influenced by its external stakeholders, interactive complexity and geographic spread, a single case study is conducted.
Interviewee Position of the interviewee Length of
interview (min) collaborations
Interviewee V Legal manager 87 Landowners, government
Interviewee W Civilian overseer 62 Landowners
Interviewee X Relationship manager 66 Clients, CIs
Interviewee Y Delivery manager 60 Suppliers, TSOs
Interviewee Z Team manager 59 Third parties (maintenance)
Table 4 Overview of interview participants
Data collection
Selecting employees, an emphasis was placed on the relationship with external stakeholders influencing or influenced by the network. Another selection criterion was that these employees manage the relationship with these external stakeholders (table 4) and hence have a close relationship with these stakeholders. In this way, more in-depth knowledge about how the external stakeholders are managed during a disruption, and how these stakeholders influence resilience, is obtained. These collaborations (table 4) also indicate the kind of knowledge the interviewees have about interactive complexity and geographic spread (e.g. interviewees collaborating with land owners know how the network is spread geographically). Since this research focuses on the three phases of resilience (preparedness, response and recovery), only interviewees with knowledge about past disruptions and the relationship with the external stakeholder(s) during that disruption, are selected.
addition, interviewing in their native language reduces the possibility of misunderstanding and misinterpretation (Voss et al., 2016). After the interviews, the recordings were translated and transcribed in to English by a C1 qualified person (CEFR) degree in English. The English transcripts were sent back to the interviewees for final corrections or clarifications. In addition to the literature review and the interviews, internal documents and websites of the CI are used to achieve internal triangulation (Voss et al., 2016).
Data analysis
The transcribed interviews were analysed according to the three steps of qualitative data analysis: data reduction, data display and conclusion (Miles & Huberman, 1994). First, quotes which are directly or indirectly related to the concepts of the independent variable (external stakeholders, interactive complexity, geographic spread), were selected (the first-order codes). Only data that is applicable for answering the research question is researched. Second, this data is categorized according the dependent variables (collaboration, visibility, flexibility and velocity) to see how the organization adapts the dependent variable according the independent variable, hence second-order coding. Third, this data is analysed regarding the third order codes ‘proactive’ and ‘reactive’, since this clarifies when actions have to be applied to increase resilience (Johansson et al., 2013). At this point, data is analysed according the variables of the dependent and independent variable, resulting into a deductive research up to this point. However, to find how the organization builds resilience, inductive research was conducted from this point onward to find patterns which result in tools to build resilience. This inductive research provides the tools organization A applies to build resilience. These tools are compared with the tools found in literature (Table 1) to find similarities and differences between production organizations and CIs in terms of building resilience. These comparisons will result in propositions regarding actions CI providers can take to build resilience. An example of the coding tree (appendix I) demonstrates how data reduction (first, second and third order codes) has progressed to descriptive codes (fourth order codes).
Findings
All interviews are coded, which resulted in an elaborative dataset and a complete coding tree. This coding tree is analysed regarding the independent (external stakeholders, interactive complexity and geographic spread) and dependent variables (collaboration, visibility, flexibility and velocity). Subsequently, these findings are, where possible, related to the resilience tools referred to in literature. An overview of the findings and their relation to the literature tools can be found in table 5. The abbreviations behind the resilience measures found in this research indicate the resilience measure established by literature (gc= goal congruence; is=information sharing; jkc=joint knowledge creation; r=redundancy; cdc=changeable delivery contracts; ac=adaptive capacity; sp=streamline processes; nvat=non-value-added time).
Collaboration Visibility Flexibility Velocity
Ex ternal stak ehol ders - Legislation (gc) - 3rd party involvement (gc) - Information sharing (is) - Monitoring - Legislation (is) - Information sharing (is) - Reflection (is) - 3rd party involvement - Monitoring - Action plan - 3rd party involvement (r) - Legislation - Legislation (sp) - 3rd party involvement (sp) - Monitoring - Training (nvat) - Action plan (r) Int era ct iv e co mplex it y - Information sharing (is) - Training (jkc) - Interconnectedness - Legislation - Interconnectedness - Action plan -Interconnectedness (cdc)
- Action plan (ac)
- Legislation (sp) - Interconnectedness - Action plan G eogr aph ic spr ead - Legislation (gc) - Information sharing (is) - Interconnectedness - Interconnectedness
- Action plan - Interconnectedness (cdc) - Action plan (ac)
- Interconnectedness - Action plan
Table 5 Overview of found measures combined with the measures found in literature
Collaboration
Data indicates that legislation can increase collaboration regarding external stakeholders and geographic spread. Government created a law specific for CIs to increase collaboration through goal congruency between CIs and its external stakeholders (e.g. landowners, construction organizations and government). Interviewee V states that “When the digging will happen near our cables, we are obliged to get in contact with the people who want to dig.” The same seems to account for the aboveground connections: “But also when land owners have an electricity pole on their land, they have to ask permission when they want to do something within those strips, like building a stable (interviewee V).“ Both parties are obliged to collaborate according to law, which will proactively decrease the risk of a disruption because the constructing party has to follow specific guidelines. This law increases resilience; when a disruption occurs at a connection with planned activities, it is easy to find where the disruption happened and what caused the disruption (interviewee V). Legislation forcing construction parties to communicate with the CI, increase resilience geographically because visibility is increased when applying the same law throughout the country (geographic spread). In addition to construction activities of external parties, organization A has to execute maintenance or reparation activities. The organization makes an agreement with the land owners about this. Interviewee W states: “This is land from someone else, so there are some arrangements about it. We aim to have unambiguous agreements with all land owners.” However, not all land owners are willing to collaborate with the organization despite agreements. Legislation exempts organization A to cross the land in those cases, to ensure that maintenance is conducted “When there is no private law agreement, there is always the double destination possibility (interviewee V)”. Although the land owners will not be pleased about it, they are obliged to collaborate with organization A (goal congruence). This collaboration makes the organization more resilient as they are able to reach the harmed connections. This also increases resilience geographically by goal congruency because all land owners throughout the Netherlands have by law the same goal, namely allowing organization A to cross their land when necessary.
climate goals. The government is directing us to transition as quick as possible to obtain the best results (interviewee V).” This will synchronize planning and development regarding interactive complexity, which increases collaboration with external parties (e.g. TSOs, water and gas companies). Organization A collaborates with important external stakeholders “which require electricity at all times, the organization will probably advise the stakeholders to install a generator (interviewee W)”. Although advising stakeholders does not increase the resilience of organization A, it does increase the resilience of its important external stakeholders. These stakeholders are enabled to continue operating when electricity is out. Information sharing can also increase resilience geographically. For example, when one TSO finds that a component is not sufficiently operable, they will inform other TSOs to check their network (and inventory) for the same component (interviewee Z) (geographic spread).
The data shows that monitoring influences collaboration regarding external stakeholders. Monitoring collaborations and connections will lower the possibility of disruption: “when everything is resolved I will talk with the landowner about what happened. We normally check everything every year to avoid things like this (interviewee V).” To contact and monitor the landowner of the land under the connection, insights in monitoring frequency, construction plans and more will be obtained. This kind of collaboration is both proactive and reactive since monitoring happens after a disruption and on a yearly base. Another stakeholder is the ACM (authority consumers market) which monitors how organization A performs. If organization A does not perform well, the ACM will send an advisory report to organization A which will also include advice regarding resilience measures (interviewee Y). In addition, Resilience will be increased when the organization either adapts management according the disruption or notices something during the yearly assessments: “So we learned that this component was not in a good condition and that also accounted for similar components at other connections (…). Then an international action is started about the same component from the same years or batch to replace them (interviewee Z).” Hence when the organization finds a point of improvement through reflection or monitoring, they can enhance SCRes by adapting their management.
In addition to collaboration with third parties and other external stakeholders, interconnectedness with other CIs is very important regarding interactive complexity and geographic spread. “All networks in Europe are connected with each other, meaning we can influence each other but can also use each other’s network (interviewee X).” This quote indicates that interconnectedness with other networks is often spread geographically. So according to interviewee X, resilience is increased by other CIs because CIs support each other in times of a disruption by sharing resources. This resource-sharing does not only involve interactive complexity but also geographic spread as the size of the disrupted area influences the need for interconnectedness. Interactive complexity does not merely increase resilience but also hampers resilience since a disruption in one CI can cause disruptions at other CIs through expanding failures (interviewee Y).
Data shows that training is the final way in which collaboration increases resilience regarding interactive complexity. Interviewee Y explains that European TSOs are in contact with each other and train every year to enhance their European protocols regarding disruptions and to synchronize goals. Those trainings entail previous and fictive cases and enable joint knowledge creation among interdependent partners.
Visibility
Data indicates that legislation increases visibility, which is influenced by external stakeholders and interactive complexity. Organization A is obliged to follow legislation, interviewee V indicates how visibility is enhanced by applying the law: “The government provides guidelines and norms and we have to follow those norms, like the minimum distance to keep from the connections.” These guidelines and norms are applied when an external party wishes to construct near the connection. Organization A shares information regarding the legislation with the external party. In addition, there is a law applied to all CIs entailing codes of conduct which “clarify what you may expect from each other, these codes work from both sides. So mainly the construction and delivery are stated in legislation and we have policies about how to handle disruptions (interviewee X) (interactive complexity).” Hence, legislation increases visibility because it lowers complexity regarding construction activities and behaviours through creating standardized protocols.
It is found that third party involvement increases visibility regarding external stakeholders. Organization A expects to build visibility in the relationship with third party maintenance teams through demanding a work plan and certificates from the third party (interviewee V). However, visibility is reduced when the supplier base is increased: “We used to have preferred suppliers (…). This increased because we needed more diversity (interviewee Z).” To increase visibility, third parties are selected on preference and certificates (interviewee V). This influences resilience positively because the organization can create agreements with third parties suitable for repairing and maintaining connections. Hence agreements ensure that well trained people, who have knowledge about the network, are available when a disruption happens.
Data indicates that visibility can be enhanced through information sharing, which is influenced by external stakeholders. When a disruption happens, all employees follow a protocol regarding information sharing with external parties: “In Arnhem is the crisis centre which clarifies what can be communicated and shared. We do not communicate externally but the communication department does, so the press will have an unambiguous story which is as honest as possible (interviewee W).” Clear protocols enhance visibility internally and externally because only well-trained internal members are occupied with information sharing. This increases resilience as people who should not be responsible for communication in times of disruption will not be occupied with it and are enabled to focus on solving the disruption (interviewee Y).
The research indicates that reflection increases visibility regarding external stakeholders. Organization A increases visibility after a disruption: “in the aftercare we get a report after 24 hours, 5 days and 10 days and we inform the clients about it. If we cannot find what caused the disruption, we involve a third-party research team to investigate the disruption (Interviewee X).” This information sharing does not directly increase resilience, however, interviewee Z states that “You can learn from this and you have to have a conversation about it (with external parties) when asked for.” This learning process increases resilience because the organization learns how to handle similar disruptions the next time and will be able to solve them quicker.
require a different approach in solving a disruption compared to above ground connections (geographic spread) (interviewee Z).
Data shows that interconnectedness increases visibility regarding both interactive complexity and geographic spread. This is mainly created through protocols clarifying how to restart parts of the network. Interviewee Y states that in times of a black-out, it is most important to restart CIs which can help restart organization A. Other interviewees emphasize the importance for organization A to solve disruptions as soon as possible to increase resilience in other CIs. For example, “we were active again in 50 minutes after the disruption of (…) happened. However, it took the train 1,5 day to be fully functioning again (interviewee Z) (geographic spread).” Thus, organization A knows how to restart the organization and is aware of clients being harmed by disruptions (interactive complexity). The protocols clarify that specific CIs can help restarting the network, this knowledge and corresponding protocols create visibility.
Flexibility
Data shows that flexibility can be decreased by legislation regarding external stakeholders. Nevertheless, interviewee W states: “When there is only some damage on the poles, we follow legislation but with bigger disruptions which can harm multiple consumers, we can act without being limited by legislation and no one will say anything about it.” Hence, limitations of legislation are diminished when a large disruption occurs as organization A is allowed to ignore legislation in times of a large disruption. Nevertheless, interviewee Z states that despite disruptions, safety is always the most important rule in the organization. Summarizing, the organization increases its flexibility by ignoring the law to some extent, but always considers regulations regarding safety.
According to the data, involving third parties increases flexibility regarding external stakeholders. When a disruption occurs, it is key to solve it as soon as possible. Certain external specialists are always available when required, but also less obvious parties like police, guards and traffic control can be summoned when needed. This type of redundancy regarding external stakeholders increases flexibility when a specialist is needed.
Another way organization A increases flexibility influenced by interactive complexity and geographic spread, is by creating action plans. The complete network of organization A consists of loops (adaptive capacity), like interviewee V explains “they created a cable connection (underground), when one connection fails, the other connection can take over.” Since the network has multiple ways to solve disruptions (e.g. looped network) (geographic spread and interactive complexity), flexibility is increased by adaptive capacity.
Velocity
The data indicates that legislation increases velocity regarding external stakeholders and interactive complexity. As mentioned before, it is important that disruptions are solved as quickly as possible and that maintenance happens in time. To ensure that organization A can always reach its connection and poles there is a law, interviewee V states “It includes deliverability, safety, reachability of the poles, and other essential regulations which require certain conditions of people.” Velocity is increased by this law because it streamlines processes of reaching a disruption. Another legislation influencing velocity is that disrupted TSOs can get support for 15 minutes of other TSOs as of the high cost of generating excessive energy. This international law increases velocity by allowing the organization 15 minutes to solve the disruption internally, which is often sufficient (interviewee Y). As a consequence of this legislation, the organization is able to continue providing its services during a disruption of generators.
It is found that external stakeholders influence training, resulting increased velocity. To reduce non-value-added time, all members of the supply chain are kept up to date through training conducted every year. The training involves fictive and real cases of disruptions, ranging from impact (e.g. by vegetation) to terrorist attacks. The training enables employees to act more quickly and results in protocols describing what should be solved first and how (interviewee W).
Another way in which external stakeholders influence velocity is through third party involvement. Organization A arranged contracts with suppliers and emergency services. These contracts state that when a large disruption happens, these parties should come in action directly. “When something big is disrupted, we have contracts with suppliers which have to deliver within a certain time span (interviewee Z).” These contracts streamline the processes of organization A to solve disruptions, hence increase velocity.
network operators: “Electricity is shut off when the frequency (50 hertz) gets below a certain point. (…) Initially 20% is switched off, when the frequency gets even lower, another 20% is switched off. When this appears to be insufficient, another 10% is switched off. So, half of the users are switched off, if this appears insufficient and the frequency continues to decrease, we have to switch off everything (interviewee Y).” The CIs try to stop a domino effect when applying this plan (interactive complexity), this will increase resilience because the amplifying disruption is stopped in this way (geographic spread). A similar plan exists for national disruptions (interviewee Y & Z). The network is built in loops; when one part of the network is disrupted, another part will directly take over. This velocity measure limits the external impact of a disruption, due to this system are many households not impacted by regular small disruptions (interviewee W) (geographic spread).
It is found that monitoring can increase velocity regarding external stakeholders by highlighting potential disruptions. Interviewee W mentioned a case in which ice created a dangerous situation. There was no disruption yet but there was a great potential, so organization A made everyone aware of it. In addition, organization A put 24/7 surveillance at that location so the organization could act directly when the disruption would occur. All external stakeholders were stand-by to solve the problem as fast as possible. In this case increased monitoring velocity because the organization could act directly when a disruption would happen and did not wait to find the cause and get everyone informed (interviewee W).
Discussion
CIs are very important to society, hence a disruption can have severe consequences (Brandon‐Jones et al., 2014; Urciuoli et al., 2014; Zebrowski & Sage, 2017). Nevertheless, there are no corresponding resilience measures identified yet which allow the organization to bounce back after a disruption (Boin & Mcconnell, 2007). This research provides multiple measures to adopt and enhance resilience in CI supply chains, whereas some are in accordance to literature, some are close to literature but research found that the CI applied a different approach, and measures which are not yet found in literature. This research aims to find how CIs build resilience, however, the research is only conducted at one type of CI. Therefore, this chapter starts a discussion on how the adoption of resilience might differ in other types of CIs.
Tools from literature
Literature established tools regarding SCRes. This research applies the measures found in relation to the conceptualization of Jüttner and Maklan (2011) (hence collaboration, visibility, flexibility and velocity). As table 5 implies, some measures found in this research have similarities to tools found in literature. The following tools are similar to the tools found in this research: goal congruence (collaboration), information sharing (collaboration and visibility), and joint knowledge creation (collaboration).
the infrastructure (Truscott, 2009). Literature does not elaborate on this kind of legislation regarding other types of CIs (e.g. water transport)
The US legislation emphasises the importance of information sharing (Truscott, 2009), which is confirmed by research. Sharing information increases collaboration and visibility. This allows the organization to find and adapt to potential threats, as it makes the organization more resilient (Jüttner & Maklan, 2011). The CI applies information sharing through collaboration and reflection with its external stakeholders. Information sharing can be either proactive, by informing stakeholders about the vulnerabilities of the infrastructure, or reactive, by informing the press about a disruption. The CI will only inform political leaders in case of a severe disruption. Boin and Mcconnell (2007) state that this kind of communication is applied in all types of CI to answer to the centralisation reflex. In this reflex expects society that the government can restore order. In case of a severe disruption, government will send a special research team to reflect on the disruption. In this way, the government obtains additional knowledge and the ability to reflect on the disruption and advice the CI in case of resilience. The research indicates that this allowed the government to create legislation in terms of information sharing between CIs and people who want to construct near an infrastructure. Informing CIs on forehand about potential threats, allows the organization to instruct the construction party about regulations according the infrastructure. In addition, for some CIs it is important to know about construction work at other CIs, for example emergency services thrive by knowing possible traffic jams and blocked roads caused by reparations at highways. Hence some legislation might differ per type of CI, but it is important for all CIs to keep informed about constructions and the possibility of disruptions to increase resilience.
The research shows that CIs practise worst-case scenarios (training) together to jointly create knowledge about large disruptions. This kind of exercises allow CIs to explore potential threats and corresponding solutions, which increases resilience by building mutual trust and understanding (Boin & Mcconnell, 2007). It appears that training with other CIs is not merely applied at electricity infrastructures but applicable is to all kinds of CI (e.g. emergency services) in order to prepare for large disruptions, for example terrorist attacks where the police, firefighters and hospitals have to collaborate (Boin & Mcconnell, 2007).
Proposition 1: goal congruence, information sharing and joint knowledge creation are applied in CIs in a similar way compared to other kind of organizations.
Tools from both literature and the research
measures is different. The tools found in literature are redundancy (flexibility), changeable delivery contracts (flexibility), adaptive capacity (flexibility), streamlined processes (velocity), and non-value added time reduction (velocity).
Literature describes redundancy as a tool to increase flexibility. An excessive amount of people (e.g. suppliers) or products (e.g. inventory) allows the organization to have back up in times of a disruption, making the organization more resilient (Truscott, 2009). Multiple researchers discuss the need of redundancy in relation to resilience, however, none describes it in terms of third parties. The research indicates that CIs have a multitude of third parties available when needed. Besides the excessive amount of maintenance and reparation organizations, which allow the organization to select only third parties of high quality, there are also other third parties available when needed. For example security to ensure safety to the employees and spectators in time of a disruption. Hence CIs do not merely have redundancy regarding one type of third party, but have redundancy because almost all (inter)national third parties are available when needed, which increases resilience. Although it is not described in literature, it is expected that this applies for almost all CIs as disruptions in most CIs can have life-threatening consequences (e.g. fire fighters without water). In addition, most parts of the physical network are specially created for CIs and require expertise and quality. These parts have to be directly available when a disruption happens. Therefore, the CI of the research created an inventory with parts to repair the connections. In some cases it can take about six months until new parts arrive, hence this inventory is necessary to ensure resilience (Truscott, 2009). Although literature cannot confirm, it is expected that this type of redundancy also accounts for some other CIs (e.g. gas transport), as every material comes with its own vulnerabilities.
In literature, adaptive capacity is described as ways in which an organization can adapt capacity; machines which normally work for 80% but can increase to 100% when needed or the use of seasonal employees can increase resilience (Ouyang, 2014). This allows the organization to be more flexible and adaptive to disruptions, hence the organization becomes more resilient (Jüttner & Maklan, 2011). In addition, research indicates another way of adaptive capacity, which is via loops. The complete network of the CI of this research exists of loops slightly differing from redundancy described by Zebrowski and Sage (2017). The additional processes in the description of Zebrowski and Sage (2017) mainly exist of processes that are not productive when there is no disruption. These processes will only be started when needed, hence in time of a disruption (Zebrowski & Sage, 2017). This is how they differ from the loops found in literature, because these loops are always operative. However, when one side of the loop is disrupted, the other will take over, hence it will double its capacity. This adaptive capacity should be monitored closely because when too much electricity is going through one connection or when capacity is increased for a long time, the connection can get disrupted. This would result in a black out for a (small) part of the connection. There is no literature available about this type of adaptive capacity, therefore it is also not known if this is applicable to other types of CI.
In this research it is found that legislation can streamline processes by providing overarching rules for both the focal firm and the stakeholders. Not only will this increase velocity, it will also enable CIs to respond quicker to disruptions and hence become more resilient (Jüttner & Maklan, 2011). The legislation found in the research is devised for CIs coping with public land (e.g. electricity, gas, telecommunication and water), hence land owners and the government. It is expected that this legislation not applies to CIs with private land (e.g. highways) or without land (e.g. emergency services). The law states that the focal firm and maintenance teams are always allowed to cross the land when necessary. By maintaining a close relationship with the land owners, the focal firm aims to make crossing private land as pleasant as possible for both parties. Truscott (2009) indicates that informing the private and public sector can increase velocity by creating more visibility. However, it seems that the government provides visibility and hence velocity by providing legislation which allows the CI provider to cross land when necessary. Although literature does not elaborate on this law nor on similar legislations regarding other types of CI, it is not certain if this legislation does apply for other types of CIs. However, it is expected that similar laws apply for other CIs so disruptions can always be solved when necessary.
P2: CIs do apply redundancy, changeable delivery contracts, adaptive capacity, streamlining processes, reducing lead times and non-value added time reduction but in a different way than described by literature:
- P2.1: redundancy is created by the high number of third parties available during a disruption and through inventory of parts to repair the connection.
- P2.2: EU legislation creates law to increase flexibility, for instance through changeable delivery contracts.
- P2.3: adaptive capacity is created by a loop system, which allow CIs to solve disruptions within a certain time span.
- P2.4: Legislation streamlines processes by enabling CIs to cross all lands when necessary, which increases velocity.
- P2.5: Training can reduce non-value added time to increase velocity.
Tools derived from research
Research provided a few tools which are not found in literature yet. These tools do increase resilience regarding the conceptualization of Jüttner and Maklan (2011) as shown in table 5, however, they do not match the resilience measures provided by literature. These tools are, in this research, described as monitoring, action plan and interconnectedness. Because these tools are not yet described by literature, not much information is available about their application at other CIs.
Often processes of production firms are in-house, however, this does not account for CIs. It is impossible to monitor the complete CI (Boin & Mcconnell, 2007), hence CIs partly account on society to detect potential disruptions. The CI informs the society about the vulnerabilities of the infrastructure, and since the society depends on those infrastructures, they are willing to report potential disruptions. The government has a special department which monitors highways. There is also a specific part of the government that monitors CIs. This part of the government will not only focus on electricity infrastructures but also other CIs, like gas and water. Hence in addition to monitoring by the focal firm, the CIs are also monitored by society and the government. Resilience will be increased through collaboration, visibility and velocity when being monitored by the focal firm, society and the government.
no literature about similar switch off protocols available, therefore it is unknown if similar approaches have been applied to other types of CI. These protocols will increase resilience by either avoiding a disruption or by stopping a disruption from evolving.
Literature does recognize interconnectedness, however, it is not described how interconnectedness influences resilience. CIs are interconnected as they are interdependent; for example telecommunication cannot operate without an electricity. Besides the fact that CIs are interdependent, some are also competitors. As described before, CIs have to bid on a market to buy electricity, with electricity companies from other countries as competitors. Zebrowski and Sage (2017) describe a situation in which an electricity organization has a monopoly, besides being expensive, the organization was not reliable. Hence the government of that organization created legislation to force the monopolistic market to a market with a pre-set maximum price. This not only lowered the price of electricity tremendously but also increased reliability as the organization was able to purchase electricity when necessary, it did not need to rely on its own energy generation anymore (Zebrowski & Sage, 2017). This example shows that disruptions can be prevented by creating interconnectedness with other CIs. P3: Resilience in all CIs can be increased through monitoring, action plans and interconnectedness.
- P3.1: Monitoring is not only performed by the focal firm but also by stakeholders.
- P3.2: Action plans help organizations to stop disruptions more quickly by providing protocols for certain events
Conclusion
This research was conducted to find how CI providers can build resilience in their network. After comparing the findings of this research with literature, results show that in addition to the regular resilience measures, CIs apply different measures. The research unravelled specific measures applicable for building resilience in CIs. In table 5, measures for CIs are recorded and classified in accordance to dependent and independent variables of this research. To conclude this research, it is found that CIs use tools found in literature and in this research simultaneously to build and reinforce SCRes; goal congruency, information sharing, joint knowledge creation, redundancy, changeable delivery contracts, adaptive capacity, streamlining processes, reducing non-value adding time, interconnectedness and action plans. These tools are influenced by the three main characteristics of CIs (external stakeholders, interactive complexity and geographic spread) and the variables of resilience (collaboration, visibility, flexibility and velocity). Once should understand that the applications of the measures differ per CI since the characteristics of every CI differ in addition to different emphasises on the conceptualization of resilience.
Managerial implications
This research shows that there are multiple measures available to adopt resilience in CIs. For management it is essential to estimate the type of CI they have to consort with. Management should analyse how the CI in question copes with its independent characteristics (external stakeholders, interactive complexity, and geographic spread) as they influence resilience. These characteristics influence the resilience concepts of the organization. Hence an understanding of how the concepts of Jüttner and Maklan (2011) apply to the organization is essential. When having an understanding how the characteristics apply to resilience concepts, measures to adopt or reinforce resilience can be found. Some measures are applicable to all types of CIs (like information sharing), whereas others are specified for a certain kind of CI (e.g. the legislation forcing changeable delivery contracts is applicable to electricity). Despite the type of CI, it is important to adopt and adapt information sharing, joint knowledge creation, streamlining processes through legislation, reducing non-value adding time, monitoring, action plans, and interconnectedness since they apply to all types of CI. These measures should be adapted regarding the resilience approach of the organization. In addition, there are more measures applicable for CIs, a literature review can provide more insights in these measures.
Limitations and research implications
Resources
Ambulkar, S., Blackhurst, J., & Grawe, S. (2015). Firm’s resilience to supply chain disruptions: Scale development and empirical examination. Journal of Operations Management, 33–34, 111–122. Boin, A., & Mcconnell, A. (2007). Preparing for Critical Infrastructure Breakdowns : The Limits of Crisis
Management and the Need for Resilience. Journal of Contingencies and Crisis Management, 15(1), 50–59.
Brandon‐Jones, E., Squire, B., Autry, C. W., & Petersen, K. J. (2014). A contingent resource-based perspective of supply chain resilience. Journal of Supply Chain Management, 50(3), 55–73.
Cao, M., Vonderembse, M. A., Zhang, Q., & Ragu-Nathan, T. S. (2010). Supply chain collaboration: Conceptualisation and instrument development. International Journal of Production Research, 48(22), 6613–6635.
Cho, D. W., Lee, Y. H., Ahn, S. H., & Hwang, M. K. (2012). A framework for measuring the performance of service supply chain management. Computers and Industrial Engineering, 62(3), 801–818.
Christopher, M., & Peck, H. (2004). Building the resilience supply chain. International Journal of Logistics Management, 15(2), 1–13.
Dyer, J. H., & Hatch, N. W. (2006). Relation-specific capabilities and barriers to knowledge transfers: Creating advantage through network relationships. Strategic Management Journal, 27(8), 701–719. Eisenhardt, K. M. (1989). Building Theories from Case Study Research. Academy of Management Review,
14(4), 532–550.
Franklin, A. L. (2001). Involving Stakeholders in Organizational Processes. International Journal of Public Administration, 24(4), 385–403.
Johansson, J., Hassel, H., & Zio, E. (2013). Reliability and vulnerability analyses of critical infrastructures: Comparing two approaches in the context of power systems. Reliability Engineering and System Safety, 120, 27–38.
Jüttner, U., & Maklan, S. (2011). Supply chain resilience in the global financial crisis: An empirical study. Supply Chain Management, 16(4), 246–259.
Kraljic, P. (1983). Purchasing Must Become Supply Management. Harvard Business Review, (September-October), 109–117.
Effect. Management Science, 43(4), 546–558.
Lund, P. D., Lindgren, J., Mikkola, J., & Salpakari, J. (2015). Review of energy system fl exibility measures to enable high levels of variable renewable electricity. Renewable and Sustainable Energy Reviews, 45, 785–807.
Melnyk, S. A., Closs, D. J., Griffis, S. E., Zobel, C. W., & Macdonald, J. R. (2014). Understanding Supply. Supply Chain Management Review, January/Fe, 34–41.
Miles, M. B., & Huberman, A. M. (1994). Qualitative Data Analysis : An Expanded Sourcebook. Journal of Environmental Psychology, 14(4), 336–337.
Ouyang, M. (2014). Review on modeling and simulation of interdependent critical infrastructure systems. Reliability Engineering and System Safety, 121, 43–60.
Ponomarov, S. Y., & Holcomb, M. C. (2009). Understanding the concept of supply chain resilience. The International Journal of Logistics Management (Vol. 20).
Rubin, H. J., & Rubin, I. S. (2012). Listening , Hearing , and Sharing Social Experiences (second edi). Thousand Oaks: SAGE Publicationis, inc.
Scholten, K., & Schilder, S. (2015). The role of collaboration in supply chain resilience. Supply Chain Management, 20(4), 471–484.
Stone, J., & Rahimifard, S. (2018). Resilience in agri-food supply chains: a critical analysis of the literature and synthesis of a novel framework. Supply Chain Management: An International Journal, 23(3), 207–238.
Truscott, P. (2009). Critical Energy Infrastructure. Whitehall Papers, 73(1), 76–83.
Urciuoli, L., Mohanty, S., Hintsa, J., & Boekesteijn, E. G. (2014). The resilience of energy supply chains: A multiple case study approach on oil and gas supply chains to Europe. Supply Chain Management, 19(1), 46–63.
Voss, C., Johnson, M., & Godsell, J. (2016). Research methods for operations management. In C. Karlsson (Ed.), Research Methods for Operations Management (second edi, pp. 165–197). New York: Routledge.
Vugrin, E. D. (2016). Critical infrastructure resilience. In M.-V. Florin & I. Linkov (Eds.), IRGC resource guide on resilience (pp. 236–239). Lausanne: EPFL International Risk Governance Center (IRGC). Wieland, A., & Wallenburg, C. M. (2013). The influence of relational competencies on supply chain
Yin, R. K. (2009). STUDY Design and Methods (Second edi). Thousand Oaks: SAGE Publications, inc. Zebrowski, C., & Sage, D. (2017). Globalisation and Intelligence. In R. Dover, H. Dylan, & M. S. Goodman
