The voice of a refugee: how information sharing influences demand
management in a social responsible service network
University of Groningen
Faculty of Economics and Business
MSc Supply Chain Management
Szymon Idzikowski
s.a.idzikowski@student.rug.nl
s2750147
Supervisors: Carolien de Blok & Kirstin Scholten
Abstract
Purpose – This paper aims to explore how information sharing influences demand management within the context of a social responsible service network. Information sharing factors and their underlying mechanisms in relation to demand forecasting and demand segmentation are investigated. Design/approach/methodology – An exploratory single in-depth case study of the asylum network in the Netherlands was conducted. Data were gathered from 11 semi-structured interviews as well as document analysis.
Findings - Key findings show how the flow of information is contingent on environmental factors that set the stage in terms of the quality (i.e. reliability, timeliness and usability) of shared information. Consequently, the quality of shared information and environmental factors rule how and to which extend demand management practices are applicable.
Originality/value – This is one of the first explorative investigations to provide deeper insights into the topic of information sharing linked to demand management in the context of social responsible service provision. A series of arguments and propositions explain the specific influence of information sharing activities on demand management practices beyond a single company perspective.
Keywords – Social responsible service network, demand forecasting, demand segmentation, information sharing.
Table of content
1. Introduction ... 3
2. Theoretical foundation ... 5
2.1 Demand Management in Service Networks ... 5
2.2 Demand Management Practices... 6
2.3 Information sharing ... 8
2.4 The social responsible service network ... 9
3. Methodology ... 11
3.1 Case selection ... 11
3.2 Interview protocol & data collection ... 12
3.3 Data analysis ... 14 4. Findings ... 17 4.1 Demand forecasting ... 17 4.2 Demand segmentation ... 20 4.3 Additional findings ... 22 5. Discussion ... 24
5.1 The flow of information ... 24
5.2 Coping with demand requirements ... 26
6. Conclusion ... 27
7. Managerial implications ... 28
8. Limitations & future research ... 29
Acknowledgements ... 29
Reference list ... 30
Appendix I (Interview guide) ... 34
1. Introduction
The dramatic increase of asylum seekers seeking refuge throughout the European Union, caused major challenges for refugee service organizations in relation to the management of the excessive demand requirements induced by this crisis (De Volkskrant, 2015). Thereby difficulties arise in establishing what exactly the needs of these asylum seekers are, as their backgrounds vary in terms of nationality, age, gender, and etcetera (Arnold, Kantona, Cohen, Jones & McCoy, 2015). Mainly due to service requirements (e.g. healthcare and legal needs) in most cases being contingent on customer characteristics. While in a commercial service network1 the customers give specific demand inputs such that the required service can be delivered, the customers (asylum seekers) in this network are usually passive in nature and do not give explicit input regarding their demand/ need requirements. This uncertainty challenges service providers in planning an effective and efficient service delivery as it is difficult to determine the specific needs such as how many and which services are required when and where (Jung, Lee, & White, 2015). Efforts in coping with these difficulties ask service providers to communicate and share demand related information with network members such that demand requirements can be managed.
In its broadest sense, demand management can be interpreted as the ability of an organization to understand customer demand requirements and balance these against the capabilities of the network (Croxton, Lambert, Garcia-Dastugue, & Rogers, 2002; Lambert & Cooper, 2000). Practices like controlling demand, influencing demand (Crandall & Markland, 1996), demand forecasting (Croxton et al., 2002), and demand segmentation (Childerhouse, Aitken, & Towill, 2002; Rexhausen, Pibernik, & Kaiser, 2012) help service providers to effectively balance demand and supply throughout the network (Baltacioglu, Ada, Kaplan, Yurt And, & Cem Kaplan, 2007). In order to make use of these practices, however, service providers require demand related information. In services, production and consumption of the service take place simultaneously hence, one source of information are the customers (Baltacioglu et al., 2007). However, when customers do not give explicit input regarding their demand/need requirements, sharing information between organizations is crucial as little information is available. Therefore, service network management is of paramount importance. To that end, literature suggest to coordinate information among network members to synchronize their demand management activities (Lee, 2002; Zeng & Pathak, 2003). Thereby, it is argued that the quality of shared information is of major importance (Yallof & Morgan, 2003), e.g. in the context of the refugee network; the specific number of the asylum seekers in need of medical assistance.
1
Within public service networks, such as the asylum network, interorganizational information sharing can include several types of entities (e.g. non-profit, commercial, and governmental) which makes interorganizational information sharing more diversified and complex (Yang & Maxwell, 2011). Mainly because public networks are often not consciously formed but entities are rather forced to collaborate. Such a network can be seen as a social responsible service network, in which different types of entities together provide a public service.
While diverse fields like service operations management and service marketing have been established to increase service productivity comparably, still little research is done on service networks (Cho, Lee, Ahn, & Hwang, 2012). In addition, despite the fact that demand management is one of the preliminary functions in service networks (Cho et al., 2012), very little is known about how service providers can enhance value by integrating processes that extend organizational boundaries (Ellram, Tate, & Billington, 2007) i.e., reducing demand uncertainty by sharing information. In particular, literature lacks studies which explore mechanisms concerning information sharing and demand management within the context of social responsible service provision. This paper addresses this gap in literature by answering the following research question: how does information sharing within and outside a social responsible service network influence demand management?
In order to answer this question, we adopted a single in-depth case study of the Dutch asylum procedure. By answering this research question, this study makes three important contributions. First, it allows us to progress the understanding of the complexities that hold in social responsible service networks. Secondly, it extends current knowledge by deepening understanding of the prior mentioned mechanisms (i.e. interorganizational information sharing and demand management) in social responsible service networks. Especially by giving valuable insights into factors that set the stage in terms of information sharing activities and how these rule demand management practices. And finally, from a managerial perspective this research enables service providers to understand the potential benefits of sharing demand related information with network partners in order to manage demand.
2. Theoretical foundation
2.1 Demand Management in Service Networks
A service network has been defined as “a network of suppliers, service providers, consumers, and other supporting units that perform the functions of transactions of resources required to produce services; transformation of these resources into supporting and core services; and the delivery of these services to customers” (Baltacioglu et al., 2007, p.112). This network of service providers can be seen as a team of individual entities who establish relationships among entities with a common business objective i.e., to provide part of a specific service (Barros & Oberle, 2012). This offers networking benefits as group members can communicate and share information with each other (Tella & Virolainen, 2005), such as to identify market developments and react accordingly to these changes (Clarkson, Jacobsen, & Batcheller, 2007). This is particularly important as services, unlike products, are characterized by intangibility, heterogeneity, inseparability and perishability (see table 1). Therefore, challenges arise in balancing capacity with demand for service networks as a result of difficulties in predicting demand patterns. For example, the delivery of many services is usually not possible unless the customer is present in the system, mainly because the customer gives the required input to the service delivery process. This in turn makes it difficult to identify necessary resources, plan required capacity, understand customer needs, and etcetera.
Service
specific feature
Definition Implication on demand
management Intangibility The service cannot be touched, tasted, or
seen.
Enhances uncertainty in managing demand (e.g. the service provider must render and generate the service at the specific request of a customer). Heterogeneity The service cannot easily be standardised
since every customer experiences a different service i.e., a variation of services from customer to customer.
The same service is different for each customer, which increases variability in demand.
Inseparability It is impossible to divorce the supply or production of the service from its consumption.
The service provider and service consumer must both participate in the service delivery at the same moment in time, which increases difficulties in reconciling supply and demand. Perishability Service is lost whenever they are not used,
services cannot be stored.
Unlike safety stock in products, services cannot be stored to cope with unexpected increases in demand. Services must be available when demand occurs.
Table 1 service characteristics adapted from (Fitzsimmons & Fitzsimmons, 2013)
management is one of the preliminary functions as it allows an organization and its network to understand customer demand requirements (i.e. knowing when customers will have which kind of service demand) and balance these against the capabilities of the network (Croxton et al., 2002; Lambert & Cooper, 2000). Demand management focuses on forecasting and managing customer requirements, with the objective of facilitating information that can help shape service network operating decisions (Cho et al., 2012). Good demand management can enable an organization and the network it operates in to become more reactive to unanticipated demand, and more proactive to anticipated demand (Croxton et al., 2002), and in turn serve customers more effectively and efficiently. According to Baltacioglu et al., (2007) the success of all other network operation decisions (e.g. in relation to capacity and resources management) depend on determining demand.
2.2 Demand Management Practices
Different practices such as controlling demand, influencing demand (Crandall & Markland, 1996), demand forecasting (Croxton et al., 2002), and demand segmentation (Childerhouse et al., 2002; Rexhausen et al., 2012) are discussed in literature that, when implemented adequately, determine how well an organization is able to balance customer demand with the capabilities of the firm/network. Although we are aware that there are more practices which enable demand management, we will solely focus on demand forecasting and demand segmentation. The reason being that demand management practices can generally be divided into two groups; anticipatory (pro-active, reaction to internal stimuli) and reactive (reaction to external stimuli). Therefore, we choose demand forecasting and demand segmentation as one being anticipatory and the other more reactive in nature. Table 2 provides an overview of the demand management practices of interest.
Demand management practice
Definition according to literature
Demand forecasting The process of converting available historic and current demand information from internal and external sources in order to predict future demand (Croxton et al., 2002; Mentzer, 2006).
Demand segmentation Demand segmentation bundles practices related to the identification of key (groups of) customers and product (service)-specific requirements (Childerhouse et al., 2002; Lambert & Cooper, 2000). Table 2 demand management practices
reduce or even eliminate management practices that induce variability by enabling consistent planning (Croxton et al., 2002), thereby reducing demand variability throughout the network. Demand forecasting can either be qualitative (i.e. judgmental) or quantitative (i.e. statistical) (Caniato, Kalchschmidt, & Ronchi, 2011). Qualitative forecasting is preferred when (i) the phenomenon under investigation is rapidly changing (Fildes, Goodwin, Lawrence, & Nikolopoulos, 2009), (ii) when no historical data are available, or (iii) in case of rare events (e.g. manmade disasters). Quantitative forecasting in turn is more accurate when (i) the phenomenon under investigation is stable and (ii) when a decent amount of historical data are available (Caniato et al., 2011). Therefore depending on the environment in which the service network operates one or a combination of both might be preferable. Consequently, we argue that forecasting is an anticipatory demand management practice as it helps to plan ahead in order to manage or avoid problems.
Another way by which demand can be managed is by carrying out different policies (e.g. per services offering) for different customer groups. As each service offering has its own customer base, by segmenting these customers into separate groups characterized by customer needs (i.e. being reactive to external stimuli) and/or service-specific requirements, firms can acquire specific skills (e.g. specialization in service provision) and knowledge to better understand the current and future demand and manage it accordingly (Lambert & Cooper, 2000). Understanding the demand variability and service volume in turn can enhance forecasting practices by guiding the selection of the appropriate forecasting method (Croxton et al., 2002). In line with this reasoning, the practices described above are considered as the practices enabling service providers to manage their demand.
2.3 Information sharing
The flow of information has shown to be key in relation to service networks, and is regarded as critical in terms of sharing information and identifying and managing demand (Ellram et al., 2004). Throughout literature it is well-recognized that information sharing2 can lead to better supply network performance (e.g. Cachon & Fisher, 2000; Wu, Chuang, & Hsu, 2014), and enhance demand management practices. Shared information can lead to better forecasts (Lee & Whang, 2000), enable segmentation based on customer characteristics, enhance value by better planning and managerial decisions (Sahin & Robinson, 2002), and long-term relationship improvements (Premus & Sanders, 2008). In order to achieve these benefits, information should be shared between members in the network (Wadhwa & Saxena, 2007). Wu et al. (2014) found that information sharing among channel members positively mediates service network performance and collaboration, particularly information sharing enables collaboration in achieving service network performance. Throughout literature, different factors are discussed that influence and/or determine whether information is shared within and outside a network and hence help to improve performance. One such factor concerns the source of information. Kaye (1995) proposes a typology for classifying information sources, arranged by location (i.e. internal or external to the network) and status (i.e. formal or informal). The location of the source can be argued to be contingent on the boundaries of the network. Whereas the status of the source is determined by the relation the source has with the recipient. Formal sources are constituted either in a legal manner or have a regulating position in relation to the recipient of information (Kaye, 1995), whereas informal sources do not have such relation.
Additionally, Denize & Young (2007) suggest that the quality of the source of information has implications on whether the information is valuable or not. For example, an organization may only use information for demand management and particularly forecasting if it is regarded as timely, useful, correct in terms of quantity and if the source of information is seen to be reliable (Kaye, 1995; Lee, Strong, Kahn, & Wang, 2002). On the other hand it is suggested that information may have quality in its own right (e.g. believability, objectivity) dependent on the source providing this information. The quality of the shared information therefore depends on the receiving entity’s perception of, and the attitude towards the source of information. This suggests that different factors (e.g. quality and source) of the shared information can enhance or complicate demand management activities. Nevertheless, difficulties in sharing and coordinating information among network members might arise due to several reasons. Kapucu (2006), suggests that this might be due to the dynamic, unpredictable, and complex nature of the environment in which organizations need to collaborate. In light of this research, this complex environment is a social responsible service network.
2
Information sharing is defined as “interorganizational sharing of data, information and/or knowledge in supply
2.4 The social responsible service network
The prior described mechanisms (i.e. interorganizational information sharing and demand management) are mostly based on the for-profit organizations and networks. Generalization towards a social responsible network is difficult because of the variety of objectives within this network. The social responsible service network (SRSN) consist of different individual entities (e.g. governmental, non-profit-, public-, and private entities) which together provide the “public service” to the ultimate customer (Osborne, Radnor & Nasi, 2013). Providing a public service makes the SRSN subject to political rather than economic control, as Bozeman & Bretschneider (1994) argue; political authority may affect some of the processes and behaviors (e.g. in terms of information sharing and demand management) as public services pertain to the effects of political authority. Public services are often not consciously formed but rather forced to collaborate, and in turn less concerned with the maximum satisfaction for their customers, but rather with their minimums needs (Jung et al., 2015). For example a temporary housing facility for causalities of manmade-disasters might only focus on the provision of basic needs e.g., food, shelter, etcetera. Berman (1998) distinguished three output goals applying for public services; equity (“the need to provide services to all citizens or equal access to those who require the specific services provided” Karwan & Markland, 2006, p.349), efficiency, and effectiveness. In general, for-profit organizations mainly focus on efficiency, less on effectiveness, and hardly on equity. On the contrary, public organizations focus mainly on equity and effectiveness, but less on efficiency (Karwan & Markland, 2006). Not being efficient has consequences, it implies that scarce resources (e.g. human resources) are not being put to their best use, and more money is spent than is needed in order to attain the same result (e.g. underutilization of transportation services). In traditional services, customers are regarded as the key participants of the service; they are deliberately and voluntarily looking for services and in most cases have predefined expectations of these services. In turn, service providers can segment these customers according their expectations (Jung et al., 2015) such to enhance effectiveness. The SRSN has a different focus, the main target group are non-voluntary citizen, these customers of aid hardly ever pay for supplies, and seldom enter into a commercial transaction with the service providers (Oloruntoba & Kovács, 2015). Customers in need are passive in nature and usually only have lateral expectations of the service (Jung et al., 2015), and therefore are usually unable to express what they need or want. This causes a lack of demand information for the service providers which increases difficulties in (i) determining demand requirements, (ii) adequately segmenting customers according to their needs, and (iii) sharing information with other network members. One could argue that having little input from customers, urges the need to generate demand requirements (by making use external and internal information sources) and share this information with other network members.
provide their services in name of the government, or governmental agencies, interrelationships emerge between the government and these service providers. Therefore, the functioning of these organizations does not only depend on their internal interactions, but also on interactions with the other network members (Bharosa et al., 2010), i.e. internal sources. Although this network perspective offers networking benefits as group members can communicate and share information with each other (Tella & Virolainen, 2005) to synergize their activities. Sharing information and managing demand might be more difficult since service providers need to take into account the different roles, stakes and responsibilities of the various network members. Thereby challenges might arise due to the differing nature of the organizations (e.g. in terms of their perception of the customer), and environmental factors (e.g. being subject to political control). Which in turn may result in conflicting goals and/or interests, pose challenges for effective information sharing (Sahin & Robinson, 2002) in terms of quality, and have implications on how demand is managed throughout the network.
Therefore this paper proposes a conceptual model to aid understanding on how information sharing within and outside a social responsible service network influences demand management (see Figure 1).
3. Methodology
With the aim of gaining insight into the relationship between information sharing and demand management in the context of a social responsible service network, this research uses case study methodology (Eisenhardt, 1989; Eisenhardt & Graebner, 2007; Yin, 1994). Case study methodology is explicitly suitable for exploratory research where the aim is development of theory rather than generalization of the results, and the primary objective to be investigated is difficult to quantify, not well understood and needs to be studied in-depth within its natural setting (Yin, 1994), as is the case for social responsible service networks. Further, it is argued that case study research is an excellent method for studying emerging concepts (Voss, Tsikriktsis, & Frohlich, 2002), and therefore well suiting the nature of the European refugee crisis. In order to gain insight into the above mentioned relationship, a single in-depth case study was conducted analysing a social responsible service network.
3.1 Case selection
Figure 2 case selection; asylum stage
We selected 11 entities of this network based on (i) the entity being a partner in the network that provides services, or enabled provision of a service to the ultimate customer (i.e. asylum seeker), and (ii) on their interdependency with the other entities in the network. In order to guide against respondent bias the selected entities had to represent a fair amount differing entities (e.g. governmental agencies, nonprofit -, private entities). By studying a network in which all different types of entities were present, it was possible to explore the network as a whole, isolating the influence of dominant parties in relation to core and absolute network members.
3.2 Interview protocol & data collection
Each of the entities that agreed upon participating in the research was visited once by one or both researchers conducting the data collection (an overview is presented in Table 4). At each of the entities the interviewer(s) met with one or multiple individuals which in most cases were in a position with a form of authority, so either in an operational position with decision mandate or on a management level. Noting that two of the interviewed entities did not hold such positions, as them being politicians. The completion of the interview guide took on average 45-60 minutesacross all the entities that were visited. All the interviewed entities filled in a consent form to maintain common interests and values, thereby among other things gave permission for the researchers to record the interview. These recordings, along with the notes made by the interviewer which was not posing any questions at that time, helped to accurately transcribe the interviews verbatim, and in turn validate the gathered data and achieve triangulation (Meredith, 1998). After each interview, a follow-up e-mail was sent to the entity enabling the transcripts to be validated and verified. This allowed the researchers to correct any inaccuracies or mistakes made in transcribing the answers.
No. Type of entity Position in the network Position of the interviewee Length of the interview (minutes)
1 Governmental agency Core Case manager 45
2 Commercial Absolute Refugee lawyer 58
3 NGO Absolute Manager 60
4 Governmental entity Core Program manager 50
5 NGO Absolute Team leader 64
6 Political Absolute Local politician and
spokesman
38
7 Political Absolute Counselor in an
municipality
28
8 Governmental entity Absolute Project manager 50
9 Commercial Absolute Advisor public affairs 38
10 NGO Absolute Policy advisor 46
11 Governmental agency Core Deputy director Asylum and Protection
40 Table 3 interview details
Further, due to the explorative nature of this study we utilised data triangulation by decisively searching for, and making use of as many diverse data sources as possible (Denzin, 1978). In other words, supplementary information was reviewed, concerning information that was either provided by one of the network members or publicly available (Table 5). This supplementary material was coded in a similar way as the interviews (see section 3.3 Data analysis).
Supplementary material reviewed Link to entity #
1. Naar een Visie 2015 5 (5.1) 31
2. Programmaplan Keteninformatisering Vreemdelingenketen 2014-2016
4 (4.1) 50
3.3 Data analysis
The interviews were recorded and transcribed verbatim, thereafter the collected data were reduced into categories by coding for the aim of interpretation and organization (Miles, Huberman, & Saldana, 2014). Figure 2 shows the underlying process of this analysis oriented on Miles et al. (2014).
Figure 2 process of coding
During coding we looked for significant themes in the text that were expected to explain the topic of this research. We started data reduction by fragmenting sentences and/or paragraphs into quotes that were truly relevant for answering the research question (First-order indicators). Afterward, we analyzed data in relation to demand management and information sharing activities. We descriptively coded all first-order indicators into several sub-categories such as “procedural steps”, “dependency on other entities” or “based on verdict” (Second-order codes). This allowed us to get a first indication of the various types of demand management and information sharing activities that were taking place in the network. In addition, it enabled us to deduce third-order themes. Dependent on the context, some of the first-order indicators were linked to show the influencing mechanisms between the theoretical themes of interest, thereby causal relationships were established through explanation building (Yin, 2009).
or directly influence demand management practices. Table 7 gives an example of coding regarding the additional findings (see appendix II for the entire code tree).
Link to DM practice Interpretive code Descriptive code
Data reduction (first-order indicators) Connected IS mechanism Demand segmentation Dependency on policy Five track system
"(X) has tried erm, to make a distribution for five different types of applicants, so someone comes here and can be divided into five different groups" [2, 4 & 5]"
Quality: Not of influence Source: External and formal internal Segmentation key
"Actually we make use of a nationwide segmentation key, based on which municipalities must accommodate refugees, allocated by regulating entities" [7 & 8]
Procedural steps
“For immigrants there is a different procedure, the integration program” [1] “Yes, the questionnaires that go with those, um, well, there are just special questionnaires so you just pose the right questions.” [11] “We have to deal with minors as well. This implies that you have to perform the interviews in a different way.” [11]
Inability to segment
Novelty of service
"we are not yet in a stage in which we can approach the problem in such way[segment demand]" [8] Quality: Not of influence Source: Formal internal Equity in service provision
"The need of all refugees is the fact that they want to be heard " [1] "No I consider them all as individual, and do my best to get them maximal results for them" [2] High influx
of demand
"We noticed that the pressure got higher because more room needs to be made, and well, capacity here then also should be utilized" [1] “I think ever man is equal; where you come from, what kind of skin color you have, it does not matter. We all need a safe place for ourselves, we need shelter, we need food, and every addition is nice but not necessary [9] ”
Interpretive code
Descriptive code
Data reduction (first-order indicators) Connected IS mechanism Dominant network members Status in the network
"What entity (x) says holds" [1] "However the status of the information is not always up-to-date" [1]
Quality: Reliability Supporting
services
"As a municipality you say what you want, and they do it" [8] "They ask for the information, because they get it a bit late" [8]
Quality: Timeliness "I receive some files shortly before starting the
eight days; sometimes two weeks before and then one week before they start ehm, the procedure […] which gives me little time.. " [2] Table 7 example of coding additional findings
To ensure data quality throughout the data collection and analysis, different safeguard measures were taken to ensure the reliability and validity following (Yin, 2009) and are explicitly depicted in Table 8. Validity and reliability Taken measures
Construct validity o Semi-structured interviews
o Pilot-testing of the interview protocol
o Having the key informants review the draft transcriptions (follow-up interviews)
o Triangulation by using multiple researchers and sources
Internal validity o Pattern matching techniques were used in order to build explanation from the derived data. Further rival explanations were addressed in case the expatiations from the literature review did not match the findings
External validity o Existing theory formed the basis for the constructs in this research setting
Reliability o A interview protocol was used in order to standardize the semi-structured interviews
4. Findings
The following three sections focus on the findings of this research organized under the two demand management themes of interest; demand forecasting and demand segmentation. Each section pays attention to how demand is managed and the underlying influencing mechanisms induced by information sharing activities (i.e. how, why and with whom information is shared). Additionally, we found several influencing factors that set the stage in terms of information sharing (see section 4.3).
4.1 Demand forecasting
We found that throughout the network forecasting methods aiming at anticipating future demand requirements are not used that often, the reasons of which can be linked to the source and quality of information. In relation to the source of information, the first reason being the dependency on external entities; as the demand flow into the network is controlled by external entities (e.g. the European Union), internal demand management starts once the asylum seeker has entered the network; “Erm.. You do not steer based on the volume which enters the Netherlands, those are measures taken at the European level” [Governmental entity 4]. This dependency makes anticipating demand challenging, as external entities set the actual demand requirements for the network itself. Within the network, in relation to both the source and quality of information, we found that the majority of network members were unable to forecast their demand as it was based on decisions made (and the speed of which these were made) by other network members; “Well in general we cannot do that much, as we rely on (entity x) which handles the application” [Governmental agency 1]. Showing that service requirements for parts of the network are determined by other network members and their services requested when required. The reactive nature of these requests causes the quality of the shared information to fall short in terms of timeliness, increasing demand management difficulties for other network members and urging the need to reactively deal with demand; “There is a bus on route with 36 people, be ready...”[Commercial 9]. From which the second reason follows, that most network members just go with the flow and deal with their demand requirements reactively based on information received from other network members which they regard as valid, as it shared by a formal3 source;“Well I have to say we go with the flow, so there is no day the same as the day before, and euhm... one moment we have a lot of families here and we try to cope with that, the other moment we have a lot of minors here and we try to cope with that, the other day we have a lot of gay people in here and we have to cope with that [...] So we don’t anticipate on that, we just cope with it on a daily basis” [NGO 5]. Lastly we found that forecasting was simply not possible because information was not useful, making it hard to draw inferences based on shared information; “Mental health reimburses
3
very late, sometimes after two years. So those numbers cannot be trusted in terms of cost, nor consumption” [Commercial 3]. Despite the fact that the majority of the network did not use any forecasting methods due to the above mentioned factors, we found that when the quality of shared information is right, forecasting demand is possible. The availability of reliable statistical information i.e., influx prognosis data received from formal network members, enabled NGO 3 to forecast certain demand requirements;“ We receive these numbers every week, enabling us to forecast…” [NGO 3], and share these (forecasts) with other network members; “We make sure we annually meet with our suppliers to check whether stocks are sufficient, and share our forecast information with them..” [NGO 3].
[receiving the information]. They also want to know whether they have to attract additional personnel..” [Governmental entity 8].
Next to solely making use of statistical information or judging available information based on intuition, we identified situations in which a combination was used to foresee different demand scenarios making use of scenario thinking. By making use of statistical data received from internal network members and intuitive expectations regarding ongoing events outside the network (external information), a range of possible scenarios were forecasted; “We don’t know exactly, but when you add everything together; seasonal celebrations, effects of policy, you know what.. between 53000 and 92000..” [Governmental entity 4]. These forecasts, in terms of numbers were found to be shared differently within and outside of the network; information shared external to the network (e.g. with politics) follows a lower number than the information shared within the network, mainly because politics has to take into account the asylum seeker as well as the general public (society). In turn, network members receive and work with the “higher” forecast, the reason being that the network is responsible for fulfilling the initial needs of asylum seekers and must be prepared as such; “As you know when you compose a fire department erm.. you give them shiny cars and hope they are not used, one could say then; can’t we get rid of these cars then because they are standing still and doing nothing, but then you would have a huge challenge in case of a fire outbreak” [Governmental entity 4]. What must be noted is that although this demand scenario information is available, it is only shared with the core network members. This lack of equal access to information is further discussed in section 4.3.
Information sharing Demand forecasting
Source of information Quality of information Inability to forecast Availability of reliable statistical information Using judgement and intuition Scenario thinking
Internal Influence Yes Yes Yes Yes
Reliability +/- + + N.A.
Timeliness +/- N.A. - N.A.
Usefulness +/- N.A. N.A. +/-
External Influence Yes No Yes No
Reliability N.A. N.A. +/- N.A.
Timeliness N.A. N.A. N.A. N.A.
Usefulness - N.A. N.A. N.A.
4.2 Demand segmentation
We found that segmenting asylum seekers based on customer needs i.e. group or service specific service requirements was not always possible. One reason being the high demand fluctuations; “Due to the high demand […] sometimes necessity breaks protocol, and it is simply not possible to keep demand separated” [Governmental agency 1]. To better deal with the impact of these high fluctuations, new services where brought to life. These new services were initiated by shared information from formal network members, initially focusing on fulfilling the minimum needs of the asylum seekers, rather than segmenting them into groups; “we are not yet in a stage in which we can approach the problem in such way[segment demand]” [Governmental entity 8]. Further we found that there seem to remain needs and necessities which cannot be segmented based on service nor group requirements because they concern the basic necessities of all asylum seekers; “The basic needs are bed, bath and bread […] That is the starting point” [Governmental agency 1], and “Well look, everyone that comes to the Netherlands, whether from Somalia, Eritrea or Syria.. stability, safety and being able to build on their future is the same for everyone ” [Commercial 4]. Which results in the unwillingness to segment demand because every refugee has the right to the same basic necessities, so due to the equity aim, they do not want to treat some groups different from others (noting that information sharing was found not to be of influence in services were equity is the norm). Once the equity needs are met, demand segmentation does take place in order to either fulfill particular needs/demands or to work in a more efficient manner to be able to process all the refugees. In each of the found forms of segmentation, information characteristics apparently play a slightly different role.
demand fluctuations. Which urged the need for regulating entities to establish “crisis” policies/procedures, each with its own service offering contingent on group characteristics to enhance network efficiency; “Well when you see that someone comes in from a safe country, then you can make a decision more quickly [concerning asylum] and prevent him from staying extensively long in one of the shelters, since capacity is scare..” [Governmental entity 4]. In turn reducing the need for other services; “they want to make new tracks for the Syrians eh, for example that they even did not get legal aid, if eh, it was expected that they would receive a residence permit but that track is not ehm, implemented yet” [Commercial 2]. Likewise segmentation based on policy takes place during distribution of asylum seekers across municipalities within the Netherlands; “Actually we make use of a nationwide segmentation key, based on which municipalities must accommodate refugees, allocated by regulating entities” [Political party 6], making information from formal sources guiding in policy based segmentation. Thereby noting that this type of segmentation is based on asylum seeker characteristics, rather than their personal needs. Nevertheless, despite the fact that information from formal network members in some cases enables demand segmentation, the quality of this information in terms of timeliness was found not being always accurate; “we are always behind.. we always get the information later” [NGO 5], or was shared too late; “Yeah, sharing or not sharing of information, sharing information to late..” [Governmental entity 8], forcing parts of the network to be rather reactive in dealing with demand, increasing segmentation difficulties.
Information sharing Demand segmentation Source of information Quality of
information Inability to segment demand Dependency on policy Customer needs
Internal Influence Yes Yes Yes
Reliability N.A. + +
Timeliness N.A. - +/-
Usefulness N.A. N.A. N.A.
External Influence No Yes No
Reliability N.A. N.A. N.A.
Timeliness N.A. N.A. N.A.
Usefulness N.A. + N.A.
Table 10 demand segmentation dependencies and linkage to information sharing. + enhances, +/- mixed, - increases
difficulties
4.3 Additional findings
information (scenario forecasts) are only shared with a distinct group of network members; “we share this information weekly with the [what they call ] the small chain” [Governmental entity 4], resulting in either limited or no access to other members. Next to this, we found certain information to only be shared confidentially or on an aggregated level, meaning that not all information could either be shared with other members, or was not reliable enough to share; “We have the whole list every week, but are reluctant to share the off the record information” [Commercial 3].
The third influencing factor found concerns the task within the network, which showed to be of influence in the way demand was managed and information was shared. Although all network members share the same central problem; providing and/or enabling service provision in respect to asylum seekers, they provide different services and have different roles and/or priorities; “It is difficult, as X said we all have our interest, and everyone has his own priorities” [NGO 5]. Resulting in conflicting interests within the network, mainly because where one tries to fulfill minimal needs, others only their obligations, some try to help the asylum seeker, whereas others just aim to make a profit; “Sometimes just from the fact that they see money […] marketing activities are also included ” [Governmental entity 8]. These different views concerning the asylum seekers enhance difficulties in collaboration and sense making of shared information; "look the organizational aims are just different between [entity Y and entity X], I am not saying that the one is better or worse, but it is just different […] and this sometimes causes issues" [Governmental agency 1].
5. Discussion
In order to manage demand requirements information is needed, mainly because this information is the basis of demand management practices. From literature it is known that information sharing can lead to better forecasts (Lee & Whang, 2000), enable segmentation based on customer characteristics, and enhance value by better planning and managerial decisions (Sahin & Robinson, 2002). Nevertheless, difficulties in sharing information among network members arise due to several reasons. One of which being the dynamic, unpredictable, and complex nature of the environment in which organizations need to collaborate (Kapucu, 2006).
Our paper contributes valuable empirical insights into the concept of information sharing in a social responsible service network (i.e., a network where there is little input from customers, urging the need to generate demand requirements by information sharing). As we found that the flow of information is contingent on environmental factors that set the stage in terms of the quality (i.e., reliability, timeliness and usability) of shared information. The perception of this quality however hinges on the end for which the information is aimed to be used. Entities who depend on other entities’ information (external as well as internal), as it sets demand requirements for their services, are challenged in anticipating demand as reliable information is only shared once the service request is made. Making the network reliant on judgement and intuition to foresee various demand scenario’s that might occur under different circumstance. Further, we found that specific demand requirements do not have to be known in order to establish procedures/policies for dealing with varying demand requests that can occur. By segmenting groups based on service requirements, in general, efficiency and effectiveness throughout the network can be enhanced. Consequently, the quality of shared information and environmental factors rule how and to which extend demand management practices are applicable.
5.1 The flow of information
demand patterns, causing the information to fall short in terms of the aforementioned quality dimensions. On the contrary information from formal external sources (e.g. European Union) showed to be perceived as more reliable in terms of identifying demand requirements for the network. The reason being that formal sources are constituted either in a legal manner or have a regulating position in relation to the recipient of information (Kaye, 1995), whereas informal sources do not have such relation. Therefore, we propose that:
P1. The quality of information from external sources is contingent on the relation the receiving entity has with the sharing entity.
In line with this proposition, our findings indicate that information from informal external sources is solely of complimentary value due to its lack of reliability. The voice of a refugee, being such an external source of information, was found not to be of complimentary use, as the asylum stage of the network is highly standardized, and guided by policies/procedures based on group characteristics rather than individual needs. Several influencing factors like the design of the network, environmental, legal, and ethical issues impede these policies and procedures, which in turn complicate and/or constrain the information flow within the SRSN (e.g., certain information cannot be shared due to privacy and/or confidentiality reasons). Extending Jung et al.'s (2015) work by showing that similar constraints hold when taking on a network perspective with respect to social responsible service provision. One such constraint being the design of the network which showed to restrict the internal flow of information in terms of equal access, making supporting members highly dependent on information from “dominant” network members. Being a formal source enhances the reliability of the shared information, suggesting that information shared by “dominant” network members has quality in its own right (Lee et al., 2002). However, we argue that in order to be of anticipatory use, information must be shared timely, as uncertainty increases by reason of the demand flow into the network being contingent on external entities, timely sharing of information between network members in order to cope with demand requirements was found to be a difficult aspect within the SRSN. Therefore, we propose that:
P2. The internal flow of information in terms of quality is contingent on formal (dominant) network members.
5.2 Coping with demand requirements
We found that anticipating demand by making use of forecasting methods becomes a difficult task when reliable information is shared last notice. Supporting earlier claims by Caniato et al., (2011), arguing that generating accurate quantitative forecasting is merely impossible when the phenomenon under investigation is unstable and reliable and/or useful statistical information is lacking. While literature (e.g. Fildes et al., 2009) argues that under such conditions qualitative forecasting methods are preferred, we show that drawing inferences solely based on external information is difficult when the shared information lacks reliability. Therefore, we argue that when available information lacks reliability the independent use of the abovementioned methods is insufficient to anticipate demand. Therefore we argue that under these conditions statistical and judgmental methods must be used complementary to anticipate different demand scenario’s, as Donihue, (1993) argues; by judgmentally taking into account environmental changes and special events (e.g. refugee crisis) to enhance the accuracy of statistical forecasts. Nevertheless, when demand requirements for certain services are contingent on other network members, difficulties arise in applying and/or making use of these forecasts in terms of anticipating demand for the entire network. Therefore, we propose that:
P3. Being dependent on other network members’ their decisions (as these set the demand requirements) impedes anticipatory demand management practices (e.g. forecasting).
Although under these conditions we support literature (e.g. Childerhouse et al., 2002; Lambert & Cooper, 2000) which states that demand segmentation smoothens demand and allows for a more efficient flow by segmenting customers based on service requirements. We argue that factors like excessive demand fluctuations, and more context related factors like; equity in service provision (Karwan & Markland, 2006) and being subject to political rather than economic forces (Bozeman & Bretschneider, 1994), lead to situations in which segmentation is not a solution. Consecutively; (i) unsegmenting demand can enhance utilization of scarce resources, (ii) there seem to remain needs and necessities which cannot be segmented based on policies nor group characteristics because they concern the basic necessities of all asylum seekers, and (iii) political forces constrain to which extend segmentation is acceptable (e.g. on sexual orientation). Therefore, we propose that:
6. Conclusion
This study focused on different factors with respect to information sharing (source and quality) and investigated their influence on demand management practices (forecasting and segmentation), as outlined in Figure 4.
Figure 3 the influence of information sharing on demand management within and outside a SRSN
the first to analyze social responsible service provision beyond a single company perspective, we are able to contribute valuable new insights on a network level. This enables us to determine that the application of demand management activities is not only ruled by the quality and source of shared information, but that the quality of shared information is contingent on the networks’ environment (i.e. influencing factors).
7. Managerial implications
The findings of this research next to theoretical implications, also provide relevant managerial contributions. Although the positive effects of inter-organizational information sharing are known in theory and practice, we provide new insights in demonstrating that influencing factors (e.g. politics, access to information, and dominant entities) set the stage in terms of the quality (timeliness, reliability and usefulness) of shared information, which in turn determines whether anticipatory or rather more reactive demand management practices are applicable. As the design of the network restricts the flow of information in terms of equal access, and the quality of information from external sources is contingent on the relation the receiving entity has with the sharing entity. We found that quality of shared information, equity in service provision and politics are of paramount importance to which, and to what extend others are enabled to make use of segmentation and/or forecasting practices. As operating in a social responsible context excludes getting rid of these factors. We advise managers to align the flow of information throughout the network, such that various demand scenarios can be generated from which each of the network members can derive their own specific demand requirements. This will enable better capacity and labour planning as the network as a whole can prepare for various scenarios.
8. Limitations & future research
While this study renders new insights, it also has limitations. The data collection, shaped mainly by interviews, was in some cases limited by the position the interviewee held within the organization (i.e. a position without a form of authority). In particular when seeking for patterns in the data, a few linkages were unclear between the findings as some of the data was more based on assumptions and opinions of the interviewee rather than on existing facts. This aspect was mitigated during data analysis by scanning for confirming statements made by other interviewees. Further, although single case studies allow to develop theory when the primary objective to be investigated is difficult to quantify, not well understood and needs to be studied in-depth within its natural setting (Yin, 1994), due to the explorative characteristics of such research, usually some limitations occur in the early stage of theory generation. Especially the conclusions from the chosen context, a single in-depth case study and the networks’ specific characteristics (i.e. asylum stage) might not be completely generalizable towards the entire network; to that end, we suggest that for future research other social responsible service networks (e.g. the integration stage of this network or the asylum network in a different countries) are explored and through qualitative research achieve either literal or theoretical replication (Yin, 2009). Furthermore, we would propose to investigate other demand management practices (e.g. controlling demand, influencing demand), not only on an organizational, but also on a network level. Additionally, although we found multiple influencing factors, that set the stage in terms of information sharing, future research should extend these findings by exploring how environmental factors influence information sharing under similar conditions.
Nevertheless, taking these limitations into account, this paper can be considered as a first step in the exploration of the complexities which hold in social responsible service networks.
Acknowledgements
Reference list
Arnold, F., Katona, C., Cohen, J., Jones, L., & McCoy, D. (2015). Responding to the needs of refugees. British Medical Journal, 351 (6731), 1-2.
Baltacioglu, T., Ada, E., Kaplan, M. D., Yurt And, O., & Cem Kaplan, Y. (2007). A New Framework for Service Supply Chains. The Service Industries Journal, 27(2), 105–124.
Barros, A., & Oberle, D. (Eds.). (2012). Handbook of Service Description. Boston, MA: Springer US. Bharosa, N., Lee, J., & Janssen, M. (2010). Challenges and obstacles in sharing and coordinating
information during multi-agency disaster response: Propositions from field exercises. Information Systems Frontiers, 12(1), 49–65.
Berman, E. M. (1998). Productivity in public and non-profit organizations: strategies and techniques. Sage.
Bozeman, B., & Bretschneider, S. (1994). The “publicness puzzle” in organization theory: A test of alternative explanations of differences between public and private organizations. Journal of Public Administration Research and Theory, 4(2), 197–223.
Blumberg, B. (2011). Business research methods.
Cachon, G. P., & Fisher, M. (2000). Supply Chain Inventory Management and the Value of Shared Information. Management Science, 46(8), 1032–1048.
Caniato, F., Kalchschmidt, M., & Ronchi, S. (2011). Integrating quantitative and qualitative forecasting approaches: organizational learning in an action research case. Journal of the Operational Research Society, 62(3), 413–424.
Childerhouse, P., Aitken, J., & Towill, D. R. (2002). Analysis and design of focused demand chains. Journal of Operations Management, 20(6), 675–689.
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 & Industrial Engineering, 62(3), 801–818.
Clarkson, G., Jacobsen, T. E., & Batcheller, A. L. (2007). Information asymmetry and information sharing. Government Information Quarterly, 24(4), 827–839.
Crandall, R., & Markland, R. (1996). Demand Management-Today’s challenge for service industries. Production and Operations Management, 5(2).
Croxton, K. L., Lambert, D. M., Garcia-Dastugue, S. J., & Rogers, D. S. (2002). The demand management process. International Journal of Logistics Management, 13(2), 51–66.
Denize, S., & Young, L. (2007). Concerning trust and information. Industrial Marketing Management, 36(7 SPEC. ISS.), 968–982.
Denzin, N. K. (1978). The research act: A theoretical introduction to sociological methods. Igarss 2014. Donihue, M. R. (1993). Evaluating the Role Judgment Plays in Forecast Accuracy. Journal of
Forecasting, 12, 81–92.
Eisenhardt, K. M. (1989). Building Theories from Case Study Research. Academy of Management Review, 14(4), 532–550.
Eisenhardt, K. M., & Graebner, M. E. (2007). Theory building from cases: Opportunities and challenges. Academy of Management Journal, 50(1), 25–32.
Ellram, L. M., Tate, W. L., & Billington, C. (2007). Services Supply Management: The Next Frontier for Improved Organizational Performance. California Management Review, 49(4), 44–66.
Ellram, L. M., Tate, W. L., Billington, C., & Ellram, L. M., Tate, W. L., & Billington, C. (2004). Understanding and Managing the Services Supply Chain. Journal of Supply Chain Management, 40(4), 17–32.
Fildes, R., Goodwin, P., Lawrence, M., & Nikolopoulos, K. (2009). Effective forecasting and judgmental adjustments: an empirical evaluation and strategies for improvement in supply-chain planning. International Journal of Forecasting, 25(1), 3–23.
Fitzsimmons, J., & Fitzsimmons, M. (2013). Service management: Operations, strategy, information technology. McGraw-Hill Higher Education.
Jung, H., Lee, C., & White III, C. C. (2015). Socially responsible service operations management: an overview. Annals of Operations Research, 230(1), 1–16.
Kapucu, N. (2006). Interagency Communication Networks During Emergencies: Boundary Spanners in Multiagency Coordination. The American Review of Public Administration, 36(2), 207–225. Karwan, K. R., & Markland, R. E. (2006). Integrating service design principles and information
technology to improve delivery and productivity in public sector operations: The case of the South Carolina DMV. Journal of Operations Management, 24(4 SPEC. ISS.), 347–362.
Kaye, D. (1995). Sources of information, formal and informal. Management Decision, 33(5), 13–15. Kembro, J., & Näslund, D. (2014). Information sharing in supply chains, myth or reality? A critical
analysis of empirical literature. International Journal of Physical Distribution & Logistics Management, 44(3), 179–200.
Lambert, D. M., & Cooper, M. C. (2000). Issues in Supply Chain Management. Industrial Marketing Management, 29(1), 65–83.
Lee, H. L. (2002). Aligning Supply Chain Strategies with Product Uncertainties. California Management Review, 44(3), 105–119.
Lee, H. L., & Whang, S. (2000). Information Sharing in a Supply Chain. International Journal of Technology Management, 1(1), 79–93.
Lee, Y. W., Strong, D. M., Kahn, B. K., & Wang, R. Y. (2002). AIMQ: A methodology for information quality assessment. Information and Management, 40(2), 133–146.
Mentzer, J.T., 2006. A telling fortune. Supply chain demand management is where forecasting meets lean methods. Industrial Engineering, 38, 42–47.
Miles, M. b, Huberman, M. a, & Saldana, J. (2014). Qualitative Data Analysis. European Journal of Science Education (Vol. 1).
Northcraft, G. B., & Chase, R. B. (1985). Managing Service Demand at the Point of Delivery. Academy of Management Review, 10(1), 66–75.
Osborne, S. P., Radnor, Z., & Nasi, G. (2013). A New Theory for Public Service Management? Toward a (Public) Service-Dominant Approach. The American Review of Public Administration, 43(2), 135– 158.
Premus, R., & Sanders, N. R. (2008). Information Sharing in Global Supply Chain Alliances. Journal of Asia-Pacific Business, 9(2), 174–192. http://doi.org/10.1080/10599230801981928
Oloruntoba, R., & Kovács, G.B. (2015). A commentary on agility in humanitarian aid supply chains. Supply Chain Management, 20(6), 708–716.
Rexhausen, D., Pibernik, R., & Kaiser, G. (2012). Customer-facing supply chain practices - The impact of demand and distribution management on supply chain success. Journal of Operations Management, 30(4), 269–281.
Sahin, F., & Robinson, E. P. (2002). Flow coordination and information sharing in supply chains: review, implications, and directions for future research. Decision Sciences, Vol. 33, Iss., 4pp(4), 505–536. Sheu, C., & Wacker, J. G. (1994). A Planning and Control Framework for Non‐profit Humanitarian
Organizations. International Journal of Operations & Production Management, 14(4), 64–78. Siggelkow, N. (2007). Persuasion with case studies. Academy of Management Journal, 50(1), 20–24. Tella, E., & Virolainen, V.-M. (2005). Motives behind purchasing consortia. International Journal of
Production Economics, 93-94, 161–168.
Voss, C., Tsikriktsis, N., & Frohlich, M. (2002). Case research in operations management. International Journal of Operations & Production Management, 22(2), 195–219.
Wadhwa, S., & Saxena, A. (2007). Decision knowledge sharing: flexible supply chains in KM context. Production Planning & Control, 18(5), 436–452.
Wu, I. L., Chuang, C. H., & Hsu, C. H. (2014). Information sharing and collaborative behaviors in enabling supply chain performance: A social exchange perspective. International Journal of Production Economics, 148, 122–132.
Yallof, J., & Morgan, C. (2003). Beyond Performance Standards: How to get the most from your outsourcing relationship. Benefits Quarterly, 19(3), 17.
Yang, T. M., & Maxwell, T. A. (2011). Information-sharing in public organizations: A literature review of interpersonal, intra-organizational and inter-organizational success factors. Government Information Quarterly, 28(2), 164–175.
Yin, R. K. (1994). Case Study Research: Design and Methods. Applied Social Research Methods Series, 5, 219.
Zeng, A. Z., & Pathak, B. K. (2003). Achieving information integration in supply chain management through B2B e-hubs: Concepts and analyses. Industrial Management and Data Systems, 103(8-9), 657–665.
Appendix I (Interview guide)
About the researchers
In light of the “Master Thesis Project” two supply chain management students at the University of Groningen, Robbin-Jan Haar and Szymon Idzikowski make part of a encompassing research project concerning the topic: The European refugee crisis from a service supply chain perspective. Because of the restrictions which come with a “Master thesis project” the time line of the project limited, and due the end of July.
Introduction and goals of the research
The largest exodus of refugees since the Second World War has brought major challenges for refugee service organizations across Europe. One of the main challenges faced concerns managing the wide range of needs of the refugees. At the same time it is difficult to establish what exactly the needs of these asylum seekers are, as their backgrounds vary in terms of nationality, age, gender, etcetera, which in the end implies different (service) needs.
We believe that challenges concerning management of asylum seeker needs can be overcome through (i) flexibility, by enabling organizations to deal with high levels of uncertainty, and (ii) interorganizational information sharing, which has shown to enhance supply chain performance. Through our research, we intend to explore how these challenges can be dealt with. The goal of this interview is to gather a wide range of knowledge for our research project on what we call “the refugee chain”; the European refugee crisis as seen from a service supply chain management perspective. With the insights gathered, we hope to contribute to managing the crisis, and serving the best interest of the refugee.
Procedure of the interview
Before the interview starts a form of confidentiality will be filled in, in order to maintain common interests and values. The interview will try to restrict itself to approximately 60 minutes. Questions about the focal firm, the internal business, and finally concerning external business will be asked. After the interview, there is room for feedback and further questions on the procedure. The information gathered will be transcribed and used to be coded into condensed information for the research. The transcriptions of the interview will be sent to the interviewee within one week after the interview to have a look at it and check if the interviewer interpreted everything correctly. If the interviewee wishes not to answer a question, or its subsequent questions, of course that is all right. The interviewee may withdraw at any time. If there are any questions before, during or after the interview, please ask.
Further notes
