S UPPLY C HAIN R ESILIENCE ORGANISATIONAL INFORMATION SHARING ON T HE INFLUENCE OF INTER -

(1)

ORGANISATIONAL INFORMATION

SHARING ON

S

UPPLY

C

HAIN

R

ESILIENCE

by

MICHEL BRUINS

Supervising Teacher: prof. dr. J. de Vries Second Assessor: dr. H. Balsters

2015

FINAL VERSION

Master Thesis TOM, MSc Technology Operations Management University of Groningen, Faculty of Economics and Business

(2)

ABSTRACT

Supply chain disruptions like a flooding or the migrants in Calais can have a major influence on supply chain. The effect of supply chain disruptions on the supply chain can be reduced by improving its resilience. Therefore, there is an increasing interest on supply chain resilience in recent research. Information sharing is important in improving supply chain resilience. Therefore in this paper, the influence of inter-organisational information sharing on supply chain resilience is studied in-depth. It is suggested that the constructs ‘reason to share information’, ‘tangibility’, ‘job position fit’, ‘supply chain positions involved’ and the ‘means by which information is shared’ are constructs of information sharing in developing supply chain resilience. Furthermore, accuracy, relevance and timeliness form the drivers of this relationship as well as trust. Finally, the job position of the persons and the supply chain position of the firm involved in information sharing form antecedents of information sharing. In addition, this research has been performed by a case study in a full supply chain context. Up-to-know, research in a full supply chain context about either supply chain resilience or inter-organisational information sharing is scarce.

Keywords: inter-organisational, information sharing, supply chain resilience, full supply chain

ACKNOWLEDGEMENT

(3)

1. INTRODUCTION

A severe natural disaster can be devastating for your company. For instance the flooding in Thailand in 2011 crippled the local hard drive suppliers (Fuller 2011). Worldwide, hard drivers immediately raised at least ten percent in price (Fuller 2011). These large scale natural disasters clearly show the impact of a disruption on the whole supply chain. These so-called supply chain disruptions are therefore described as an event that disrupts the flow of goods or services in a supply chain (Craighead et al. 2007). Noteworthy, a supply chain is described as “a set of three or more entities (organisations or individuals) directly involved in the upstream and downstream flows of products, services, finances, and/or information from a source to a customer” (Mentzer et al. 2001, pp 4).

Not only are severe natural disasters a source of supply chain disruptions, smaller disruptions like migrants trying to break into trucks in Calais (Capon 2015; Pullman 2015) have an impact on the supply chain as well, at least when a supplier does not take action to become more resilient to it (Kim et al. 2015). Therefore, organisations are increasingly interested in ways to become more resilient. This resilience can be described as the capability to anticipate and overcome supply chain disruptions (Pettit et al. 2013) or the ability to recover from supply chain disruptions quickly (Blackhurst et al. 2011).

Consequently, in recent literature there is an increasing interest in finding characteristics of supply chains that influence its resilience. Thence, collaboration is found and stressed out as an important factor for a resilient supply chain (Jüttner & Maklan 2011; Pettit et al. 2013). Collaboration can be described as “two or more firms working together to jointly achieve a greater success than can be attained in isolation” (Daugherty et al. 2006, pp. 61) (as in Daugherty 2011). Collaborative activities comprise information sharing, joint relationship efforts and dedicated investments (Nyaga et al. 2010). All these activities are empirically proven to increase supply chain resilience (Scholten & Schilder 2015)1_{, with information}

sharing as a critical factor to realise the benefits of collaboration (Zhou & Benton 2007; Wei et al. 2012; Zhou et al. 2014; Ambulkar et al. 2015) as information sharing is encouraging parties to commit to the relationship (Anderson & Weitz 1992). Also, information sharing is argued to be essential in creating flexibility in the supply chain (Silveira et al. 2001), which in addition is an important construct of supply chain resilience (Ponomarov & Holcomb 2009; Jüttner & Maklan 2011; Pettit et al. 2013; Johnson et al. 2013; Scholten et al. 2014). In detail,

(5)

information sharing can be described as the “inter-organizational sharing of data, information and/or knowledge in a supply chain” (Kembro & Näslund 2014, pp 181).

Yet, while the positive effect of information sharing on supply chain resilience is empirically proven (Scholten & Schilder 2015), research about the underlying structure of information sharing on supply chain resilience (i.e. what aspect of information sharing affects which construct of supply chain resilience) is still lacking. It is therefore of scientific interest to research the influence of information sharing on supply chain resilience in more detail. From a managerial point of view, this topic is of interest, because the outcome of this research can give insight in how the sharing of information can help organisations to anticipate on or recover from supply chain disruptions.

Furthermore, Kembro & Näslund (2014) found that the majority of research about information sharing in supply chains has been performed on a dyadic relationship. Following Mentzer et al. (2001, pp 4) however a supply chain is: “a set of three or more entities…”. Kembro & Näslund (2014) therefore suggests that research should focus on the supply chain as a whole. This research will follow up on that suggestion and will consequently add to the scientific world as one of the few researches on information sharing in the context of the full supply (Kembro & Näslund 2014; Kembro et al. 2014).

In short, this paper contributes to the literature in adding insights on information sharing as an activity to affect supply chain resilience, thereby facilitating new insights on the underlying structure of the effect of information sharing on supply chain resilience in the context of the full supply chain.

(6)

2. PROBLEM STATEMENT

2.1. Main research question

What aspects of information sharing between organisations in a supply chain influences supply chain resilience?

2.2. Research objective

To get a better understanding of how information sharing between organisations affects supply chain resilience.

3. LITERATURE AND RESEARCH FRAMEWORK

3.1. Supply chain resilience

As mentioned in the introduction, supply chain resilience can be described as the capability to anticipate and overcome supply chain disruptions (Pettit et al. 2013). Specifically, supply chain resilience assists in dealing with those supply chain disruption(s) by building adaptive capacity to prepare for the disruption(s), by reducing the impact of the disruption(s) and by solidifying the capability to recover quickly form it (Sheffi & Rice 2005). Jüttner & Maklan (2011) described this as the three phases of supply chain disruptions: readiness, responsiveness and recovery. The supply chain disruptions studied in this research will be expand on in the method section (see Events).

(7)

to maintain high situational awareness. These items are all present in recent literature and will respectively be referred to in this research as the following constructs; flexibility (Jüttner & Maklan 2011; Pettit et al. 2013; Johnson et al. 2013; Scholten et al. 2014), adaptability (often referred to as the ability to (re)engineer the supply chain) (Tomasini & Van Wassenhove 2009; Ponomarov & Holcomb 2009; Zsidisin & Wagner 2010; Blackhurst et al. 2011) velocity (Ponomarov & Holcomb 2009; Jüttner & Maklan 2011; Johnson et al. 2013; Scholten et al. 2014) and situational awareness (Blackhurst et al. 2011; Bode et al. 2011; Pettit et al. 2013; Scholten et al. 2014). Furthermore, following Ambulkar et al. (2015, pp. 112) the definition of supply chain resilience used in this research is: “the capability of the firm to be alert to, adapt to, and quickly respond to changes brought by a supply chain disruption”. In order to get a better understanding of the concept supply chain resilience, the above mentioned constructs and its relation to supply chain resilience will be elaborated in more detail in the following.

Flexibility

First, flexibility is an important element as flexibility enables an effective response to supply chain disruptions (Pettit et al. 2013) because it allows a supply chain to react to an unforeseen change (Tummala et al. 2006) by for instance modifying the delivery schedules or production planning (Sheffi & Rice 2005). Therefore, the definition of flexibility as “being able to bend easily without breaking” (Jüttner & Maklan 2011, pp. 247) will be used, in the sense that a supply chain is resilient to a supply chain disruption when it can easily adapt to a disruption without breaking the flow of goods.

Adaptability

(8)

it is of great importance that the necessary processes are prepared for. In this way, the supply chain is capable of enabling the execution of those processes when needed for (Scholten et al. 2014). So, the difference between flexibility and adaptability is that flexibility is about easily altering the process when a disruption happens and adaptability is about the extent to which the process can absorb a disruption when it happens.

Velocity

Third, a high velocity leads to a faster response to market changes or events (Christopher & Peck 2004) thereby facilitating an improvement in the speed of recovery from disruptions (Jüttner & Maklan 2011), which as a result improves supply chain resilience. Accordingly, velocity can be described as the ability to reduce the timeframe where the disruption impacts the supply chain, which determines the loss that happens per unit of time (Jüttner & Maklan 2011).

Situational awareness

Finally, situational awareness could be described as the firm’s recognition and awareness of pending disruptions (Kembro et al. 2014). In this sense, Bode et al. (2011) indicate that firms can improve their response to a supply chain disruption by encouraging a supply chain disruption orientation, thereby reducing the impact of the supply chain disruption, because firms that spend time on scanning and learning from the environment are better able to develop abilities that improve their responsiveness (Ramaswami et al. 2009).

3.2. Effect of information sharing on supply chain resilience

(9)

Also, Whipple et al. (2002) showed that the exchange of information between organisations improves the speed and quality of decision making within those organisations, which suggests that information sharing could increase velocity. This is in line with recent research by Scholten & Schilder (2015) which suggests that information sharing increases velocity and hence supply chain resilience. Hence, it seems logical to vindicate a direct relationship between information sharing and supply chain resilience.

3.3. Information sharing

In order to get a better understanding of this relationship, first, information sharing has to be defined more thoroughly. Following Wacker (2004, pp 631): a good (operational) definition must be “a concise, clear verbal expression of a unique concept that can be used for strict empirical testing”. A concept can be considered clear if its definition can replace the term used in a sentence without changing the meaning of this sentence (Wacker 2004).

Although the importance of information sharing is extensively elaborated (e.g. Lamming 1996; Whipple et al. 2002; Malhotra et al. 2005; Sheu et al. 2006; Cao & Zhang 2011), definitions of information sharing differ. For instance Mohr & Spekman (1994, pp 139) defined information sharing as “the extent to which critical, often proprietary, information is communicated to one’s partner”, whereas Whipple et al. (2002) defined it in conjunction with timeliness, accuracy and relevance. Nevertheless, Sheu et al. (2006) concluded that information sharing is about level (i.e. job position) and quality. Furthermore, Wadhwa & Saxena (2007) argued a critical distinction has to be made between data, information and knowledge, while Zhou & Benton (2007) advocated the inclusion of content. More recently, Cao & Zhang (2011) promoted to include confidentiality to the definition of information sharing and Ha et al. (2011) relates information sharing to the willingness to share that information.

(10)

3.4. Research framework

Recall from the introduction that the majority of research on information sharing was done on dyadic relationships. Altough this does not mean findings of those researches are all invalid for a full supply chain, it does mean that those findings are not suitable for empirically testing the relationship of information sharing on supply chain resilience (Wacker 2004). Therefore, Kembro & Näslund (2014, pp 184) identified four themes, based on a systematic review of literature on information sharing in supply chains:

 Why: benefits of/reasons to implement information sharing in supply chains.  What: what data, information and knowledge should be shared.

 How: the means of facilitating information sharing in supply chains.

 Antecedents, barriers and drivers: factors that may prevent or promote the implementation of information sharing in supply chains.

Contributing to these four themes, Kembro et al. (2014) added the ‘to whom’ theme because different information needs to be shared on different job positions (Sheu et al. 2006) and positions in the supply chain (Whipple et al. 2002; Nyaga et al. 2010; Porterfield et al. 2010). As most extensive and considerate, this framework is used in this research.

Why

The question ‘why’ information is shared should be placed in a socio-economical context (Kembro et al. 2014). More specifically, do managers for instance share information because they belief this increases performance and reduces uncertainty in the supply chain (a transaction cost economic (TCE) view) or because they believe this strengthens their relationships with other entities in the supply chain (a relational governance theory (RGT) view). The assumption that just one of these views fits inherently results in an incomplete understanding of the complex structure of information sharing (Chu & Wang 2012). This lies at the heart of why different studies have different, and sometimes opposing, findings on information sharing (Kembro et al. 2014). In essence, this means that different views/reasons to share information will generate different outcomes. Therefore, they were included as a construct of information sharing.

(11)

production schedules or adapting its facility capacity, the flexibility and adaptability of this supply chain will be lower (Sheffi & Rice 2005; Johnson et al. 2013), which indicates that the supply chain resilience will be lower as well (see 3.1. Supply chain resilience).

In contrast, people with a more relational governance theory (RGT) view share information in order to improve inter-organizational cooperation (Wei et al. 2012). Because cooperation is suggested to have a positive influence on supply chain resilience (Scholten & Schilder 2015), it is theorised that supply chains in which people share information from a more RGT view will have a higher supply chain resilience. Still, it should be minded that people do not fit solely to one view but only have a tendency towards one of them (Chu & Wang 2012).

What

For using information sharing in a supply chain context a relevant distinction is the tangibility of information, i.e. data, information or knowledge (Wadhwa & Saxena 2007). For instance, mutually creating knowledge increases velocity (Scholten & Schilder, 2015) and therefore influences supply chain resilience positively. In contrast, data has to be interpolated first and therefore increases the time needed for decision making, thus decreasing velocity (Wadhwa & Saxena 2007). Also, the question with whom to share this content is advocated in recent literature, referred to as the job position (Sheu et al. 2006). Sheu et al. (2006) states that an intense relationship between people who share information is crucial for this information sharing to be effective and as different job positions involve different intensities of the relationship, certain job positions lead to a more effective information sharing then others. Also, the different position of an organization in the supply chain is of influence on the effectiveness of information sharing (Whipple et al. 2002; Nyaga et al. 2010), because benefits of information sharing can only be reaped when it is shared with different persons in different position in the supply chain (Fearne & Taylor 2006). More specific, this will result in a better capacity and resource planning across the whole supply chain which in turn increases adaptability and flexibility (Ponomarov & Holcomb 2009; Wieland & Wallenburg 2012) and therefore supply chain resilience.

How

(12)

Also, organisations can fear to become dependent on specific entities in the supply chain, because of dedicated investments in specific information systems (Wong et al. 2012). Hence, the means of information sharing has to fit to the context, otherwise people can find it too costly or inconvenient to share information (Tallon 2012; Kembro et al. 2014). At the end, actually sharing the information is essential in creating flexibility in the supply chain (Silveira et al. 2001), which suggests that the convenient means and those which require low investments have a more positive influence on supply chain resilience.

Antecedents, barriers and drivers

In order to address information sharing accurately, recent research argues that the impact of antecedent, barriers and drivers also have to be covered (Hernández-Espallardo et al. 2010; Kembro et al. 2014). Certainly, it is quite obvious that it doesn’t make sense to share irrelevant (i.e. too much) information with other entities in the supply chain or for instance to share the dawn of a supply chain disruption when it already happened. Therefore, information must be relevant, accurate and timely in order to have effect on supply chain resilience (Whipple et al. 2002; Malhotra et al. 2005; Zhou & Benton 2007; Cao et al. 2010; Zhou et al. 2014). These three indicators at a more aggregate level could be described as information quality (Kembro et al. 2014), in other words, information quality could be described as the degree of fit between the exchanged information between organisations and the needs of those organisations (Zhou et al. 2014).

Also, confidentiality of information (Cao & Zhang 2011; Wei et al. 2012; Kembro et al. 2014) as well as trust (Gaur et al. 2011; Ha et al. 2011; Wei et al. 2012) and mutual dependency (Klein & Rai 2009; Scholten & Schilder, 2015) are often referred to in recent literature. The latter increases the willingness to share information and therefore positively influences supply chain resilience (Scholten & Schilder 2015). Confidentiality of information and trust, in turn, motivates people to act and actually share information (Bode et al. 2011), which positively influences supply chain resilience (Silveira et al. 2001).

3.5. Conceptual model

(13)

antecedents, barriers and drivers as moderators of information sharing. More specific the ‘why’ theme is defined by the reason to share information. Here, a RGT view is suggested as a better view than a TCE view, so a RGT view positively influences supply chain resilience where a TCE view negatively influences supply chain resilience. The ‘what’ theme comprises tangibility of information (data is most tangible and knowledge least tangible), job position fit between people involved in information sharing and the different positions of the firms in s supply chain that are involved in the information exchange. Here, it is expected that if firms from more different positions in a supply chain are involved, supply chain resilience will be higher. Furthermore, the ‘how’ theme includes the means by which the information is shared, where a better mean is described as a more convenient mean which requires lower investment. Lastly, antecedents, barriers and drivers include that the information is relevant, accurate and timely, it includes confidentiality of information and trust and mutual dependency. This is visualised in the following conceptual model.

Information is relevant, accurate and timely Tangibility Reason to share

Job position fit

Supply chain positions involved Means by which information is shared Confidentiality of Information Trust and mutual dependency Flexibility Velocity Situational awareness Adaptability

-+

+

(14)

4. METHOD

4.1. Philosophy and archetype

Before elaborating on the method choice, a short notion is made to the reader on the relative perspective of the researcher. It is fundamental for the methodological consideration because it touches the heart of the philosophical debate and reveals the research preferences of the researcher (Karlsson, 2008). In this study the researcher will take on a post-positivism/critical theory viewpoint (Meredith et al. 1989; Easton 2010; Järvensivu & Törnroos 2010) and finds most recognition in the archetype of a dialectician and an empiricist archetype (Karlsson 2008).

4.2. Method choice

From a technical perspective, there is no clear division between quantitative and qualitative research methods within this study (Karlsson 2008). Being aware of certain preferences, the most appropriate method was chosen and elaborated here after.

Choice for qualitative research

First, recall from the previous chapter that in-depth research on the influence of information sharing on supply chain resilience is relatively new to the academic world. Specifically, research that addresses the a full spectrum of supply chains. Therefore, this research describes the relationship between the concepts of information sharing and the resilience of supply chains. More specifically, it puts the focus on the relationship between constructs of information sharing and supply chain resilience. This is done by explaining the complex phenomena of information sharing and the reason why it influences supply chain resilience in more detail. By doing so, this research contributes to the academic world in an existential way (Cooper & Schindler 2008; Yin 2014).

(15)

Choice for case study

Not only the type of research questions influence the appropriateness of a certain research method. Two additional conditions are the degree of control over actual behaviour and the degree of contemporary focus (opposed to historical elements) (Yin 2014). In this research the degree of control over behaviour was limited since supply chain disruptions were unplanned (Craighead et al. 2007). This research focused on contemporary elements because supply chain disruptions have to be resolved quickly (Jüttner & Maklan 2011).

Because this research focussed on supply chain disruptions (i.e. events that disrupt the flow of goods), it is important to take in mind that information about past events is often available only through surveying or interviewing people who remember the events (Cooper & Schindler 2008). Besides, a case study allowed the researcher to research the phenomena of information sharing while maintaining holistic and meaningful characteristics of real-life events (Yin 2014). Therefore, to understand the complex phenomena of information sharing, a case study was an eminent choice (Ellram 1996; Yin 2009).

4.3. Research design

The following section provides an elaboration of the research design, which provides the plan to get from questions to answers (Rowley 2002). Because the research questions were elaborated before this section will expound on the unit of analysis (Yin, 2009; Kembro et al., 2014) and the case selection (Runeson & Höst 2008; Barratt et al. 2011). Dubé & Paré (2003, pp. 610) suggested that a clearly definition of the unit of analysis “is critical if we want to understand how the case relates to a broader body of knowledge.” This correlated to the finding of Kembro & Näslund (2014), who argued that current research on information sharing is (incorrectly) generalised to the full supply chain, while a dyadic relationship was researched.

Unit of Analysis

(16)

Focal company

Because this research is about supply chain resilience, the case has to relate to supply chain disruptions. This required that the focal company has suffered from supply chain disruptions. Therefore, in contrast to the random sampling (Seawright & Gerring 2008), this study uses a purposive approach for selecting an appropriate number and type of cases (Voss et al. 2002). This inferred that findings can only be generalised to a general theory via a process of analytical reasoning (Yin 2009). Moreover, understandings can only be generalised to those cases that are theoretically identical to the ones used in this study Therefore, to improve the external validity of this research, a focal company with a broad assortment of products was selected. One which had a large number of different suppliers. From within this focal company, it was then possible to randomly select a supply chain partner in order to maximise the generalisability to the boundaries of this focal company. Because this research was limited to within The Netherlands, the number of firms with a wide range of products and a large number of suppliers was narrowed down significantly. Specifically, ecommerce firms are offering the widest amount of products because they have no maximum amount of physical store locations (Chang et al. 2003). Known to the research, bol.com is offering the widest amount of products in the Netherlands as they were offering over ten million unique products. For this, they had relations with over eight hundred suppliers. Therefore, this company was selected as focal company.

Events

(17)

and the supply chain returns to the original - or even to an improved phase (Sheffi & Rice 2005; Jüttner & Maklan 2011).

Case A Case B Case C

Sector FMCG FMCG FMCG

Main products Books Books Toys

Disruption Trucks could not drive because of ice on the roads (in south east USA)

Immigrants in Calais want to enter trucks, so shipments were delayed

No response on placed orders, so orders could temporarily not be delivered

Effect Important products (lots of consumers affected when delayed because of cross docking)

Important products (lots of consumers affected when delayed because of cross docking)

More expensive products and also a high amount of crossdock orders

Entities involved Wholesaler, Transport, Retailer, End-consumer

2*Transport, Retailer Wholesaler, Retailer, End-consumer

Impact minimized? Limited Yes Limited

Information Sharing

Why? RGV RGV TCE

What? Data, information, (knowledge)

Data, information, (knowledge)

Data, information, knowledge

Who? Multiple job positions, multiple positions in supply chain

Multiple job positions, multiple positions in supply chain

How? Telephone call, E-mail, Advanced in-house build system(s)

Telephone call, E-mail, Advanced in-house build system(s)

E-mail, EDI

Figure 2 - Case Selection Criteria

4.4. Data collection method

In order to get a better understanding in how data should be collected, a data-information-need framework was developed (see Appendix B). During the research, this framework helped to recall whether collected data and information indeed relate to the specified needs and therefore increases the purposefulness. Subsequently, the data collection methods were fitted to the data and information needed. Here, the combination of multiple methods increased the richness of the results of a research (Mingers 2001) also referred to as triangulation.

Triangularity

(18)

there was enough fundament to start with a questionnaire (Meredith et al. 1989) to validate the existence of relationships between the constructs of information sharing and constructs of supply chain resilience mentioned in the conceptual model. Now, this more rational nature of theory also facilitated semi-structured or structured interviews (Meredith et al. 1989). As mentioned above the addition of knowledge to the scientific world of this paper lies in describing the relationship between the concepts of information sharing and supply chain resilience. And in addition, a start has been made to bring the research from descriptive to explanatory (Cooper & Schindler 2008). Therefore semi-structured interviews were superior, as this gives a deeper understanding of the complex phenomenon (Mingers 2001; Voss et al. 2002) Furthermore, the interviews were used to verify the results of the questionnaire, which increased the validity of this research, because of triangulation (Mingers 2001; Voss et al. 2002).

Technique selection

A relevant phase for the method section is the technique selection (Pratt 2007; Yin 2009). In this research, the questionnaire was web based. The advantages of a web based questionnaire includes improved quality due to pre-programmed validation checks, lower data entry error because of electronic answers, which could automatically by transferred into the analysis and the opportunity to skip irrelevant questions (based on earlier answers) or add visuals and audio to improve the comprehensiveness (Van Gelder et al. 2010). Also, Van Gelder et al. (2010) state that web-based questionnaires are returned more rapidly and unforeseen problems could be immediately be adjusted, which are both important advantages for a master thesis. The most important disadvantage for this research was the possibility of reluctance because of safety and confidentiality concerns (Van Gelder et al. 2010). Therefore, in the introductory text a notion was made to the confidentiality of this research.

(19)

4.5. Data collection procedures and questions

Next, after the data collection methods were determined, the procedures and questions had to be worked out (Miles & Huberman 1994; Cooper & Schindler 2008).

Questionnaire (web based)

The questionnaire was used to collect information about the characteristics of information sharing before, during and after supply chain disruptions. Also, the questionnaire was used to validate the items from the conceptual model (which will be expounded in 4.6. Data analysis). Here, the European guidelines and standards for measuring Information Society data collection were used, from the second OECD Guide to Measuring the Information Society (OECD 2011). This enhanced the comparability of data and the quality of the questionnaire (Bryman & Bell 2011; Rowley 2014). See Appendix C for the actual questionnaire.

The questionnaire was sent to 362 different e-mail addresses of 276 unique companies. In order to increase the response rate, a separate email was sent to pre-announce the invitation for the questionnaire. Two reminders were sent and after the first reminder a control message was sent to validate the arrival of the invitation (i.e. to check for spam issues). In total 47 respondents (13% response rate), from 44 different companies (16% response rate) filled in the questionnaire. The demographic characteristics were well dispersed (see Figure 3).

Type of firm Number of respondents

Manufacturer/Publisher 13

Distributor/Wholesaler 24

Retailer 8

Third-party provider 1

Transportation 1

Number of employees Number of respondents

1 0 2-9 13 10-50 13 51-100 4 101-250 5 251-500 2 501-1000 6 1000+ 4

(20)

Job title Number of respondents Intern 2 Engineer 1 Professional 5 Planner/Scheduler/Analyst 6 Buyer/Sales representative 13 Section manager/Supervisor 12 Senior manager 3

Vice president/director/general manager 2

CEO/CFO/CIO/President 3

Years of experience Number of respondents

1 6 2 8 3 5 4-5 6 6-7 6 8-10 6 11-15 4 16-20 4 21-25 0 25+ 2

Figure 3 (continued) - Demographic characteristics of the questionnaire

Unfortunately, the appropriate sample size was 86 with an margin error of +/-10%, a confidence level of 95%, a standard deviation of .5 and the population as all suppliers of bol.com (see formula 1 and 2) (Berenson et al. 2012). By lowering the population to 362 (i.e. all suppliers of which an active e-mail address was available) still, a sample size of 76 was needed (see formula 2).

Formula 1 – Deriving the sample size (Berenson et al. 2012)

=

96 

=

86 

=

76

Formula 2 – Filled in formulas

(21)

1994; Kovács & Spens 2005; Walton 2004; Cooper & Schindler 2008). The latter will be explained in the data analysis section.

Interviews (semi-structured)

The questionnaire was send in advance and filled in before the interview took place, so the interviewees had in indication what the interview was about. Also, before the actual interview the interviewees were informed in more detail about the interview. To improve the reliability an interview protocol was used (see Appendix D) and in order to make the interviews as highly comparable and less biased as possible the order of questions was follow as much as possible (Turner 2010). In the following figure the interview details are stated.

Type Case A Case A -B Case B Case C Case C

Core activity Wholesaler Retailer Carrier Retailer Retailer

Job Position Section Manager Section Manager Planner/ Scheduler Section Manager Professional Time recorded (min) 42:54 43:04 37:54 44:36 32:57

Figure 4 – Interview details

4.6. Data analysis

A helpful starting point in evaluating the obtained data was to analytically manipulate the data (Yin 2009). Therefore, after the collection of the data, heat maps and frequency tables were made. After some first insights were found, the three steps suggested by Miles & Huberman (1994) were followed. These three steps comprised at first data reduction, followed by data display and finished with conclusion.

(22)

Figure 5 – Categories and subcategories

During the data display phase of the questionnaire frequency tables and heat maps were used to get a better understanding of the data (Yin 2009) (See Appendix F for a sample of the analysis). Thereafter, following Ambulkar et al. (2015), the substantive validity coefficient of the reasons to share information were measured, which is a method to validate whether constructs are ranked important enough to be included in the research. Other reasons were added in order to improve the ranking of those reasons (Cooper & Schindler 2008). This was also to secure that other reasons were covered in the questionnaire in case they were referred to during the interviews. The substantive validity coefficient is measured as Csv = (nc - n0) / N

(Lawshe 1975; Anderson & Gerbing 1991), where nc is the number of respondent confirming

the item and n0 the number of respondents refuting the item. Large positive values indicated a

Disruption

E Event

I Impact

Information Sharing

IS1 Why

IS1T Transactional cost economic (TCE) IS1R Relational Governance Theory (RGT)

IS2 What and Who

IS2J Job Position

IS2S Supply Chain Position IS2T Tangibility(+ what specific)

IS3 How

IS3M Mean

IS4 Antecedents, barriers, drivers

(23)

substantive validation of the theorised item, whereas large negative values indicated a substantive validity for a construct other than the theorised item (Ambulkar et al. 2015). Items with a Csv value greater or equal to 0.5 were included in this research (Anderson & Gerbing

1991; Ambulkar et al. 2015). Initially the ‘means by which information is shared’ would have been measured by this method as well (see Appendix F for the constructs which initially had to be measured by ranking), however, due to a limited amount of questions and technical limitations of the questionnaire tool this construct could not be validated by this method. Therefore, a free entry field was added to the questionnaire to improve the evaluation options of this item (see Appendix C).

Next, further analysis was performed including correlation analyses and t2 tests in order to determine relationships between data, as Likert scale type of data are considered robust for using parametric statistics (Norman 2010). These outcomes were used as input for the interviews, in order to increase reliability and improve the process of inductive, deductive and abductive reasoning (Arthur 1994; Kovács & Spens 2005; Williams 2007; Cooper & Schindler 2008; Heracleous & Lan 2012). Subsequently, after the interview and the coding of quotes, sentences and paragraphs, following Scholten & Schilder (2015) the data was juxtaposed in order to interpret the outcomes. Also, axial coding was used to improve this process (Kendall 1999; Charmaz 2006; Amsteus 2014; Charmaz 2014). The findings are stated in the next chapter, after the following paragraph of limitations.

Limitations

A natural limitation of qualitative research is that the scope and level of analysis is limited (Kim et al. 2015) because understandings could only be generalised for those cases that are theoretically identical to the ones used in the study (Runeson & Höst 2008). Hence, there is a trade-off between selecting highly different cases to gain rich data or selecting similar cases to be able to generalise results. Nonetheless, this effect was reduced because the questionnaire cases were randomly selected and the answers from the interviews were compared to the questionnaire results during the interviews.

(24)

starting with them. Moreover, because of the choice for a web-based questionnaire, major errors could have been corrected, which ultimately was not needed.

5. FINDINGS

This chapter elaborates on the main findings of this research by expounding the results found during the questionnaire and the interviews. The findings are categorised in the same order as used throughout this paper, starting with the reason why people share information. Every subcategory is worked out following the roadmap used during the research. In other words, starting with the results from the questionnaire followed by the interviews.

5.1. Reasons to share information about a supply chain disruption Questionnaire

The TCE view as well as the RGT view were substantively validated (see Figure 6 and Appendix F for a sample of the analysis). Recall from the method section that other reasons were added in so they could be ranked by the respondent. Remarkable are the high scores on ‘improve cooperation’ (RGT view). Every respondent agreed on the question whether they shared information about the disruption to improve cooperation. In accordance, the RGT view was considered as the most important reason (see Figure 6).

Reasons to share Likert Average

St.de v.

Rating score (sum)

Ranked 1 Ranked 6 Agreed Disagree d Relational (RGT) 6.50 0.83 34 6 0 16 0 Trust 5.67 0.80 63 1 2 13 0 Secure critical resources 5.69 1.50 49 3 0 12 1

Reduce cost (TCE) 5.50 1.68 53 3 2 12 2

Dependency 5.20 1.44 64 2 7 11 2

Contingency 5.07 1.46 73 1 5 9 2

Figure 6 – Reasons to share

Interviews

(25)

disruption. Logically, this raised the question why these interviewees marked this item as important. The interviewees reacted comparably that you always want to try to improve the cooperation, which was reinforce by one interviewee, by; “especially when your relationship is not good.”

All except two interviewees pointed out that during a disruption it is most important to have a TCE view, in the sense that “impact is the most important”, “it is most important to know how much certain choices will cost in order to make the right decision” and “if costs or benefits are not made visible, the carrier from case A will not take action.”

One of the other interviewees indicated that “not taking the cost now will probably cost more in the long run, because customers will buy less products if you don’t deliver them in time”. Also a striking answer was “we take the extra costs for rerouting and do not charge them to our customers. We want to keep up our service level agreement.” The interviewee however also admitted that it is important to be aware of a possible increase in transactional costs during supply chain disruptions, because “During the disruption it was prime time for carriers to take advantage of the situation, because the capacity was scarce.” However, she added “Luckily, this was not the case, as we have a good relationship.”

The other reasons to share (i.e. contingency and secure critical resources) were not agreed on as being most important, during the interviews. This is in congruence with the theoretical framework (Kembro et al. 2014).

(26)

In contrast, when a major disruption strikes the supply chain the shipments cannot arrive within the margin. Then, from a RGT view, an interviewee indicated that “there was nothing more we could have done to reduce the impact of this disruption.” This concerned the major disruption in case A. At all cost, the firm searched, together with its supply chain partners, for the fasted option to reroute the shipment. From a RGT view, this was justifiable, as the interviewee did wanted to show how well they performed in reducing the impact of the supply chain disruption and therefore interacted as much as possible. From a TCE view however, those additional costs may not be justifiable. Regard from above that this is substantiated by the quote; “It is most important to know how much certain choices will cost in order to make the right decision.” Hence, from this viewpoint an e-mail indicating that the shipments are delayed because of icy roads would have been sufficient.

5.2. Tangibility, job positions and positions in the supply chain Questionnaire

The data from the questionnaire signified substantial differences between the data, information and knowledge shared about disruptions (see Figure 7). On order to signify differences on individual level t-test were used (i.e. averages can diminish those insights Devore 2006) (see Appendix G). Before a disruption, respondents share more data and information than knowledge. During the disruption, more data is exchanged than knowledge, there is however no substantial difference between information and data or knowledge. After the disruption there is no apparent difference between either data, information or knowledge. Furthermore, there is substantially more information shared before a disruption than after a disruption.

Data exchanged Likert

Average

St.dev. Agreed Disagreed

Before 5.56 1.03 13 0

During 5.63 1.15 12 0

After 4.81 1.56 9 3

Information exchanged Likert Average

Before 6.00 1.15 14 1

During 5.50 1.32 11 1

(27)

Knowledge exchanged Likert Average

Before 4.75 1.39 8 2

During 4.69 1.54 7 2

After 4.25 1.61 6 5

Figure 7 – Tangibility

With regard to the job position a difference was found: professionals indicated to get too little information at all phases of the disruption, while section managers indicated to receive exactly enough information during and after the disruption. However, before the disruption they did not agree (see Figure 8). However, one section manager did not agree, because he indicated to get too much information (though, in all others disagree was answered as receiving to little information). Job position fit will be elaborated in more detail at the findings of the interviews.

Professionals Agreed Disagreed

Before 1 2

During 0 3

After 1 2

Section manager Agreed Disagreed

Before 4 4

During 5 3

After 7 1

Figure 8 – Job position differences on answering the statement “I received exactly enough information”

Similarly, a distinction was noticed concerning the position in Supply Chain: distributors and wholesalers indicated to receive enough information before, during and after the disruption. Also, one respondent indicated to receive too much information before the disruption. In contrast, retailers indicated to receive too little information at all phases (see Figure 9.

Distributors and wholesalers Agreed Disagreed Before 5 3 During 6 2 After 7 1

Retailers Agreed Disagreed

Before 1 3

During 0 4

(28)

Figure 9 – Supply chain position differences on answering the statement “I received exactly enough information”

Regarding the amount of different positions involved in the exchange of information also difference where found. The outcomes ranged from only informing one partner to informing the full supply chain. However, there was no relation found with either constructs of supply chain resilience, or other constructs of information sharing.

Interviews

Data/information/knowledge

The difference between data, information and knowledge was indeed confirmed during the interviews. With regard to what was shared, data mostly comprised sales data and shipment data like digital packing slips, which were used to determine safety stock and track the delayed shipments. Information included volumes, destination addresses, standard operating procedures and daily reports about the shipments. Knowledge was limitedly shared.

All interviewees agreed that impact was the most important information, which was in one case a crisis report the other cases a telephone call in which costs and risks of the disruption were discussed. Finally, what let a disruption to happen, was marked as important by most interviewees.

Job Position

Except for one interviewee, everyone agreed that there is a clear difference in the level of information shared with persons on different job positions. Consensus was found in “only disruptions with a high impact reach high job positions”. One interviewee substantiated this by: “When our director is involved you know we face a serious issue”.

Furthermore, agreement was found in the statement that specialists from one company should have contact with specialists from the other company (e.g. planners with planners and business analysts with business analysts), because “otherwise information is muddled up” and because “specialists react more quickly because they recognise what’s going on and know how to react”. In addition, one interviewee revealed that occasionally “information ends up at the wrong person first which leads to slack in the processing time and hinders a quick response to the disruption”.

(29)

indicated there are special cases in which they had contact with a supplier, however the cases were limited to major disruptions when the expertise of the logistical section managers was needed.

Position in Supply Chain

There is a considerable difference between the information exchanged with people in different positions in the supply chain. More specific, on the one hand there is an objective difference in what information is shared when and how. On the other hand, however, only in the eyes of the persons involved in information sharing, referred to as the difference in need for information.

Two interviewees indicated that retailers receive and share too little, one interviewee specified “We can improve our preparation for upcoming disruptions if more information was shared with us. Also, about disruptions which do not hit us, so we can foresee if they are going to strike again and hit us the next time.” In addition, another interviewee indicated that suppliers were indeed genuinely surprised when retailers pro-actively share information.

It was already indicated that some people only have contact with the suppliers in case of disruptions, but also the supplier indicated a difference; “We start calling carriers first.” The interviewee also indicated that; “If we have to reroute I’ll try get the best contact first and have contact with that single person. Thereafter, I inform everyone from the distribution list to inform them with one single e-mail, so everyone knows what’s going around.”

In accordance, most interviewees indicated that there is indeed a significant difference between firms on different positions in the supply chain. For instance one interviewee indicated “There is a vast difference between wholesalers and manufacturers. Manufacturers know better what they are doing, there are in better control over the supply chain. Wholesalers mostly don’t know themselves what they are getting when. Information from manufacturers is more accurate and timely.” Another interviewee described this in more detail as; “The entity on the beginning of the supply chain has much effect on every following entity. It is a chain reaction and therefore you have to intervene as high as possible in the supply chain.”

The latter quote was in line with other interviewees indicating that in case of a disruption the information flows downstream, but entities are searching for information upstream. This is explained, because “entities at the beginning of a supply chain have more accurate information.”

(30)

other, the following differences were mentioned, which marked the differences. If there is a disruption, the suppliers want to know whether and when the shipment is going to arrive, in order to know when the shipment is financially completed. The carriers want to know if, how much and where they can collect so they know when to reserve which capacity. The people in the warehouse of a retailer want to know exactly when the shipment is going to arrive so they know how to reschedule their personnel planning. Finally, the retailer wants to know when the shipments is coming, which products are included and why they are delayed, so they can inform the customer.

5.3. Means to share Questionnaire

From the questionnaire there was a clear difference between the preferences for certain means; e-mail and telephone where favoured above the other. Between those two, there was one difference as e-mail was favoured after a disruption (see Figure 10).

In the free entry field most reasons to use these two means comprised ‘fast’, ‘easy’, ‘convenient’ and ‘availability’.

Means used Before During After

E-mail 15 16 16

Telephone (call) 12 12 9

Supply chain (web) portal 6 2 2

EDI 4 2 2

Collaborative Software 3 2 1

Advanced in-house build

system(s) 2 1

Customer/Vendor Relationship

Management software 1

Fully integrated supply chain

management software 1 1 1

LIS offer file 1

Plaza Ahold 1 1

Telephone (text messaging) 1 1 2

Figure 10 – Means by which information is shared

Interviews

(31)

notify that the disruption is over, how much the delivery is delayed, or that products are rerouted via a certain route. A large advantage of e-mail called by every interviewee, is that you can open it at the first moment you are available. Three interviewees indicated; “it handles with time difference.” Where another interviewee indicated; “I get a lot of phone calls so people cannot always reach me, on e-mail I’m always available”.

However, a distinction was made between how large the disruption was; “Get the phone immediately when there is a big disruption, you want to know for sure that the message is received, and mostly you want to have more details. For a small disruption e-mail is perfect”. Another interviewee agreed on this that it is sometimes “faster and easier to take the phone, because e- it can be demanding to explain all details about at a disruption in an e-mail”. There were also other means mentioned, however they were mostly called devious. One interviewee said; “All day we are behind are desk to answer phones and e-mails, a website should be checked next to it. That is annoying, especially if you have to check the websites of all suppliers separately. That is impossible.” One interviewee lucidly described it as; “Other means than e-mail and telephone are often acclaimed by the deviser, however supply chain partners mostly disapprove it.” He continued with an example (i.e. the supply chain portal) which is renowned at the focal company, yet infamous at the supply chain partners for the additional, sometimes burdensome effort which has to be made. Nonetheless, there is an advantage of a website or home-made tool, as “it provides additional information after the supply chain disruption has been announced to the supply chain partners.”

5.4.Accuracy, relevance, timeliness, confidentiality, trust and dependency Questionnaire

Moreover, accuracy, relevance and timeliness were substantively validated. Confidentiality was more agreed on than disagreed, but was not substantially validated. Also, the standard deviations were high which signified disagreement between respondent (see Figure 11).

Accuracy Likert

Average

Before 5.60 1.35 14 0

During 5.40 1.18 12 1

After 5.40 1.18 11 3

Relevance Likert

Average

St.dev. Agreed Disagreed

Before 5.73 1.03 14 1

During 5.40 1.24 12 1

(32)

Timeliness Likert Average

St.dev. Agreed Disagreed

Before 5.27 1.44 13 2

During 4.80 1.47 10 3

After 5.40 1.06 12 0

Confidentiality Likert

Average

Before 4.27 2.05 8 5

During 4.07 2.02 7 6

After 4.27 2.09 7 5

Figure 11 – Accuracy, relevance, timeliness and confidentiality

Furthermore, trust and dependency were validated, although dependency was ranked lowest (see Figure 6).

Interviews

Accuracy, relevance, timeliness

The interviewees agreed that the information they received about the disruptions was mostly accurate, relevant and timely. Also, it helped them handling the disruption. However, there were some exceptions where information was not relevant, accurate and timely, which symbolises the process of information sharing and its impact on resilience. The following event deliberately described how important accurate, relevant and timely information is; “Two weeks ago a priority shipment should arrive on Thursday, everybody in the warehouse was lined up. However, the shipment did not arrive. After contacting the supplier, they again gave wrong information about where it was. At some point the next day, the section manager from the warehouse called and told us, the shipment had arrived. At that point, we still had not clarified where it was. At the end, the exchanged information had increased the impact of the disruption on our supply chain, because it increased the uncertainty about where the shipment was and let us make fallacious decisions.”

In contrast, with regard to the disruption in Calais, interviewees indicated that they were immediately informed when a disruption happened and that information was very accurate. The interviewees indicated that the disruption was well handled and decisions where easy and quickly made.

(33)

During the interviews it was corroborated that confidentiality of information is not influencing supply chain resilience. Essentially, one interviewee indicated that there is no relevant confidential information about supply chain disruption at all “it is not confidential as it is just shipping information”. Also one interviewee indicated “Confidential information is not relevant for supply chain disruptions. We bring goods from A to B there is no confidentiality in that.” One interviewee refined this to “Sometimes we explicitly share confidential information, in order to create goodwill in order to ensure that a partner will help you out when you have problems.” This signifies an indirect relationship where sharing confidential information is an antecedent of trust, which is in line with the findings of Wei et al. (2012). Nonetheless, this also indicates no influence on either information sharing or supply chain resilience directly.

Trust

All interviewees indicated to have trust in their supply chain partners. They mindfully confirmed that it is important for supply chain resilience. Mindfully, because they also indicated that “trust does not really play a role, as we have no reservation in the sincerity of our partners.” Similarly one interviewee indicated that there is a high amount of trust, but also added “if there is no trust between partners it would have been more difficult. We would not have such a high service agreement if we did not had such a profound trust in our partners.”

Mutual dependency

Finally, dependency was also substantiated during the interviews. Although, there was no agreement on how it influences the relationship of information sharing on supply chain resilience. On the one hand, two interviewees indicated that it helps to be less dependent, because “You have a stronger voice in the case, which relaxed the process of broaching a supply chain disruption at a supply chain partner and therefore increases the change of preventing it in the future.”

(34)

6. DISCUSSION

In this chapter, the stated findings from the previous chapter are related the research question. In order to answer this research question, first, the subcategories are approved, rejected and refined by relatively inductive, deductive and abductive reasoning and linking the findings to the theoretical framework. Subsequently, this is aggregated to an empirical model, which answers the research question.

6.1. Logical reasoning

A transaction cost economic (TCE) view for small disruptions and a relational governance theory (RGT) view for major disruptions.

First, in the theoretical framework of this research, a RGT view was suggested to positively influence supply chain resilience, because cooperation is linked to a higher supply chain resilience (Scholten & Schilder 2015). However, despite the proposition that information sharing improves collaboration (Nyaga et al. 2010) during this research this suggestion is only approved for major disruptions.

A possible explanation may be found in the findings by Kembro & Näslund (2014) that previous research has incorrectly generalised findings of a dyadic relationship to a full supply chain, like the research of Nyaga et al. (2010). For example, it may not be observable for a buyer and a supplier that costs are present at a different entity in the supply chain. Or, the other way around, for a buyer and a supplier it may not be distinguishable that for small disruptions costs should be made visible in order to get every entity involved in the process. Moreover, it could also be overlooked, because in the research on a dyadic relationship the focus may only be on the buyer and the supplier.

Noteworthy, this is a suggestion which holds for a full supply chain, however which was only found from the interviews. Therefore, as triangularity could not be applied (Mingers 2001; Voss et al. 2002) and qualitative research is naturally constraint (Kim et al. 2015), this suggestion is proposed as a new direction for further research on information sharing within a full supply chain during disruptions.

Tangibility influences supply chain resilience

(35)

found as most important. More specific, it is suggested that determining the impact is most important for creating an effective response to a supply chain disruption, which in turn increases flexibility (Ponomarov & Holcomb 2009; Wieland & Wallenburg 2012). This impact was either shared by means of a crisis report or by telephone.

Furthermore, as shared data was used to determine safety stock, it is suggested that this could also increase supply chain resilience. Because, an improved safety stock increases adaptability, as indicated in the theoretical framework (Zsidisin & Wagner 2010).

Lastly, knowledge is not substantiated to increase supply chain resilience from the data obtained in this research.

Different job positions in the supply chain influence what and how information is shared. Job position fit suggested to influence supply chain resilience.

Third, considering the job position, it was confirmed that job position fit influences supply chain resilience. If specialists share information with other specialists it is suggested that velocity is increased as the information is interpreted more quickly. In accordance with Sheu et al. (2006) it is confirmed that the information sharing was more effective, which indeed resulted in a better capability and resource planning (Ponomarov & Holcomb 2009; Wieland & Wallenburg 2012). Therefore, adaptability and flexibility were increased.

Furthermore, during the analysis it became apparent that the job position of a person on itself influences what information is shared or is (whether or not consciously) withhold from a certain job position. Also, how information was differently preferred, because of a difference in availability (i.e. some interviewees were less available by phone then others). Altogether, this suggests that job position on itself is an antecedent of ‘what’ and ‘how information is shared.

Different positions in the supply chain influence what and how information is shared. Positions involved influence supply chain resilience.

(36)

disruption risks)(Whipple et al. 2002; Fearne & Taylor 2006; Ramaswami et al. 2009; Nyaga et al. 2010).

Similar to the job position, it was found that the supply chain position influenced ‘what’ information is shared and ‘how’ it is shared. Therefore, it is suggested that the supply chain position should also be included as an antecedent.

Means has to be easy, fast, convenient and available

Fifth, it is confirmed that the mean has to fit the context of information sharing during a supply chain disruption (as suggested by Tallon (2012) and Kembro et al. (2014)). More specific, that recommends that the mean should be easy, fast, convenient and available. Therefore, e-mail and telephone are suggested as most suitable for sharing information about supply chain disruptions. This is congruent with the research framework where it was proposed that the mean should not require dedicated investments (Tan et al. 2010), people should not fear to become dependent on it (Wong et al. 2011) and it should be convenient (Tallon 2012; Kembro et al. 2014).

Furthermore, ‘easy’ fits to ‘flexibility’ which is defined as being able to bend easily without breaking and ‘fast’ fits to ‘velocity’ which is defined as the ability to reduce the timeframe where the disruption impacts the supply chain. Finally, ‘availability’ can rationally be linked to ‘situational awareness’ as this, in accordance with the theoretical framework, increases the amount of time which could be spend on scanning for supply chain disruptions (Ramaswami et al. 2009).

Accuracy, relevance and timeliness suggested as drivers

Sixth, accuracy, relevance and timeliness were evidently approved in this research. From the findings of this research and in congruence with the theoretical framework, it is suggested that if those factors are not achieved supply chain resilience could not be reaped from information sharing (Whipple et al. 2002; Malhotra et al. 2005; Zhou & Benton 2007; Cao et al. 2010; Zhou et al. 2014).

(37)

Trust and mutual dependency influence relationship of information sharing on supply chain resilience

Finally, trust and dependency were confirmed of influencing the relationship between information sharing and supply chain resilience. First, trust was confirmed in agreement with the theoretical framework. However, mutual dependency was only partially confirmed in agreement with the theoretical framework, as higher mutual dependency positively influences the relationship between information sharing and supply chain resilience (Klein & Rai 2009; Scholten & Schilder 2015). In contrast, some interviewees indicated the opposite and notified a negative influence of dependency, which could be explained by the findings of Scholten & Schilder (2015) that mutual dependency also increases dedicated investment, which decreases flexibility. However, as investments were out of scope for this research this could not be verified. Concluding, it is apparent that mutual dependency influences the relationship of information sharing on supply chain resilience, however, it is suggested that this relationship should elaborated in more detail in a following research in order to get a deeper understanding of how relationship could be explained.

6.2. Empirical framework Scale of a disruption Information is relevant, accurate and timely Tangibility, except knowledge Reason to share

Job position fit Supply chain positions

involved Means by which information is shared Trust Mutual dependency Flexibility Velocity Situational awareness Adaptability

-+

+

Easiness, rapidity, convenience and availability Job Position Supply Chain Position

S UPPLY C HAIN R ESILIENCE ORGANISATIONAL INFORMATION SHARING ON T HE INFLUENCE OF INTER -

ORGANISATIONAL INFORMATION

SHARING ON

S

UPPLY

C

HAIN

R

ESILIENCE

MICHEL BRUINS

2015

ABSTRACT

ACKNOWLEDGEMENT

Table of Contents

1. INTRODUCTION

2. PROBLEM STATEMENT

3. LITERATURE AND RESEARCH FRAMEWORK

-+

+

+

+

+

+

+

4. METHOD

=

=

=

5. FINDINGS

6. DISCUSSION

-+

+

+

+

+