Faculty of Economics and Business
Master of Supply Chain Management
RESEARCH MASTER THESIS
Logistics coordination in disaster relief through clusters.
Benefits and constraints
1
Acknowledgements
Abstract
In the field of humanitarian relief, a cluster approach has been introduced as a means to coordinate relief actors and mediate a more efficient and effective response to the affected people. Logistics accounts for a great proportion of the relief response and its importance is increasingly recognized, as it makes possible the acquirement, flow and storage of goods and materials necessary for the relief intervention. In this study, I review both the benefits and challenges of coordination in the Logistics Cluster relief chain, in terms of transportation and procurement operations. The contribution of this research is based on information emerging from one in-depth case study. The paper provides reflections on supply chain coordination through clusters, in the field of humanitarian logistics, thus contributing to the body of relevant literature and practice. The study finally puts forward several propositions with regards to the benefits and constraints of the Logistics Cluster coordination tied to the relationships between logistics functions and coordination mechanisms. This study concluded by proposing several future research suggestions.
2 Table of content
Abstract ... 1
1. Introduction ... 3
2. Theoretical background ... 5
2.1. Supply Chain coordination... 5
2.2. Clusters as a means of coordination ... 7
2.3. Cluster coordination in humanitarian logistics ... 8
2.3.1. Humanitarian logistics ... 8
2.3.2. Humanitarian Cluster Approach ... 9
2.4. Conceptual Framework ... 11
3. Methodology ... 11
3.1. Introducing the case ... 12
3.2. Data collection ... 14
3.3. Data analysis ... 15
4. Results ... 16
4.1. Information sharing ... 16
4.2. Information technology ... 19
4.3. Joint decision making ... 19
4.4. Joint working ... 20
5. Discussion... 22
6. Conclusions, limitations and further research ... 26
6.1. Implications for research and practice ... 26
6.2. Limitations and further research ... 27
Bibliography ... 29
Appendix A: Glossary of acronyms ... 35
Appendix B: Estimated damage caused by reported natural disasters 1975-2012 ... 36
Appendix C: Natural disasters summary 1900-2011 ... 37
Appendix D: Number of natural disaster by country 1976-2005 ... 38
Appendix E: List of analyzed documents ... 39
3 1. Introduction
The increasing number and complexity of natural disasters, that is phenomena that cause disruptions in the functioning of a community involving widespread human, material, economic or environmental losses (UNISDR, 2009), and the growing number of organizations engaged in relief operations have made specialization of tasks among actors and coordination both important and challenging (van Wassenhove, 2006; Jahre and Jensen, 2010). Recent statistics (The International Disaster Database (EMDAT), 2012) show that in 2011 the number of natural disasters increased dramatically, to nearly 400 from 100 in 1975, which affected approximately 200 million people. Concurrently, the estimated damages caused by reported natural disasters registered a value of 35 million dollars in the same year (EMDAT, 2012)1. Humanitarian relief chains employ many actors and each of them have different interests, mandates, capacity, and logistics expertise, but no sole actor has sufficient resources to respond effectively to a major disaster (Balcik et al., 2010). Events following the Indian Ocean tsunami in 2004 and the Darfur crisis in 2004/2005 drew attention to the problems of coordination, more precisely, the relationships and interactions among the involved organizations (Jahre and Jensen, 2010). While some affected populations received increased attention, others were not at all well served (Adinolfi et al., 2005; OCHA, 2007). The relief management of natural disasters particularly implies very critical decision-making related to the well-being of affected people and since eighty-percent of disaster relief is logistics, which focuses on the efficient management of flow of goods, information and services, to respond to urgent needs of the affected populations (Sheu, 2007), then the response to the logistical problems in such difficult emergency situations should be quick and efficient (Barbarogosġlu et al., 2002; van Wassenhove, 2006). Despite this and the growing requirements for agencies to coordinate, theory offers little information on how precisely logistics coordination is beneficial or whether there are obstacles preventing its realization. Coordination in the commercial area has been greatly studied, however, in humanitarian relief chains it is still incipient (Balcik et al., 2010). Research so far has revealed that apart from the high number of stakeholders (van Wassenhove, 2006) and the fierce competition for donor funding (Thomas and Kopczak, 2005), coordination in disaster relief is also limited by cultural barriers, and technological incompatibility (Schulz and Blecken, 2010). A coordination initiative in humanitarian relief chains is the cluster system, basically “a template for how coordination should be carried out in a number of areas”, as defined by Jahre and Jensen (2010; p. 658). Altay and Labonte (2011) found that not all humanitarian actors considered the establishing of the cluster approach (CA), including the cluster for logistics operations (the Logistic Cluster, henceforth LC), to be valuable, firstly because it implied a top-down
1
4
United Nations coordination and secondly because it carries additional workload for agencies and no supplementary resources to perform it (e.g. unnecessary meetings) (IRIN, 2008). It is only recently that LC‟s evolution and performance has been formally evaluated through reports and since its implementation in 2005 literature was not able to capture the advantages and impediments of the LC with regards to coordination. In the light of this incipient discussion in literature, I aim to answer the following research question: what are the benefits and constraints of coordination in the Logistics Cluster in sudden onset natural disasters?, where benefits are here seen as positive outcomes and constraints (or disadvantages) as negative effects that limit the desired outcome of coordination. The emphasis is put on circumstances of immediate response in sudden onset natural disasters, due to their unpredictability, critical consequences and complex mobilization of resources (Chakravarty, 2014; Jahre, Jensen and Listou, 2009; Murray, 2005), which make coordination efforts much more difficult. Due to their unpredictability, sudden onset natural disasters such as the Haiti earthquake leave room for little preparation and therefore coordination, making the preparedness stage to be outside the scope of this research.
General logistics literature has indicated the necessity and benefits of coordination, while practice has revealed that there are matters that prevent coordination from being beneficial (Balcik et al., 2010; Stephenson, 2005). This situation gives a twofold opportunity for research. First, it is useful to revise general theory on coordination and clusters and see how the humanitarian cluster system can benefit from coordination and what salient particularities make the coordination effort disadvantageous. Since there is a need to explore this duality in the field of humanitarian logistics, this research contributes to the growing stream in literature attempting to better understand the different aspects of relief chain coordination through clusters and identify the implications carried by the establishment of the LC in particular. Second, studying the LC system and analyzing its benefits and constraints makes it possible to contribute to the general coordination knowledge of sudden onset disasters. The aimed contribution of this research to practice is to reveal the coordination benefits and constraints encountered by the LC in particular logistics activities, and determine whether they can be overcome. Also, by challenging coordination theories this research sheds new light on humanitarian cluster coordination, which can provide new insights for humanitarians in terms of trade-offs for HL cluster coordination. In practice, great effort has been made to establish and manage clusters, with the greater purpose of alleviating suffering of affected people. Thus, drawing essential knowledge from agency reports should be valuable for this purpose.
5
methodology, which will cover one disaster relief case representing an example of the LC coordination, followed by a discussion of the research findings. Through an analysis of the Haiti earthquake I explore both the positive and negative sides of the coordination mechanisms employed by the LC, which are afterwards translated in a set of propositions. The final section elaborates on the implications for theory and management practices, presents limitations and identifies suggestions for future research.
2. Theoretical background 2.1. Supply Chain coordination
A traditional supply chain (SC) consists of suppliers, manufacturers, distributors, and retailers with the objective of providing value to the end consumer and leveraging profitability gain for each partner (Sahin and Robinson, 2002). Supply chain management (SCM) refers to the management of a set of activities starting from the ordering and receiving of raw material, through manufacturing, distribution and to the final delivery (Sirias and Mehra, 2005). Although the definition of SCM is not unique and its origin is not well documented in literature (Chan and Chan, 2010), many papers imply a logistical perspective, meaning that activities cover functions such as inventory, procurement, production planning or transportation (Chan and Chan, 2010; Malone and Crowston, 1994; Simatupang, Wright and Sridharan, 2002). A profitable and well-integrated SC requires the coordination of “numerous activities spread over multiple functions and organizations” (Arshinder et al., 2008; p. 317; Jayaram et al., 2011; Sirias and Mehra, 2005; Xu and Meng, 2014). The motivation behind SC coordination lies in the ability of SC members to share the risks and benefits, and also helps in improving the performance measures of the SC (Cruijssen et al., 2007; Arshinder et al., 2008). Additionally, researchers state that a lack of coordination may negatively affect forecasts, capacity utilization, inventory levels, customer focus and satisfaction, inventory costs, time to market, order fulfillment response or quality (Ramdas and Spekman, 2000). However, coordination efforts of organizations conscientious of its advantages face also various constraints and require developing strategies for sharing the benefits, costs, and investments among partners (Fugate, Sahin and Mentzer, 2006). A frequently mentioned impediment is the competition between members and their conflicting objectives, as each member is looking to optimize its own individual functions (Sahin and Robinson, 2002). The constantly evolving dynamic structure of a SC is also a factor in the coordination equation (Sahin and Robinson, 2002).
6
Consequently, in this paper, I will use the term „coordination‟ to describe the management of interdependent processes and activities between two or more organizations that work closely to execute SC operations so that they reach mutually defined goals.
A frequent case of coordination, involves members or organizations found at the same level in the SC or that deliver similar services -horizontal coordination- (Jahre, Jensen and Listou, 2009). In this respect, Cruijssen et al. (2007) and Kumar and van Dissel (1996) mention in their literature reviews that benefits of horizontal coordination include cost reduction, economies of scale, specialization, a more efficient use of labor force, knowledge and learning opportunities.
While the benefits of horizontal coordination are promising, it is difficult to pinpoint effective coordination mechanisms, i.e. methods or tools used to manage the interdependence between organizations (Xu and Beamon, 2006). According to the review of Arshinder et al. (2008) coordination mechanisms recurrent in literature are: information sharing, information technology (IT), joint decision making and joint working.
Information sharing between SC members yields better decision making and helps alleviate the effects of SC dynamics (e.g. ordering, capacity allocation and production planning) (Cruijssen et al., 2007). Transportation orders, inventory levels, demand, production schedules or performance metrics are examples of information types (Arshinder et al., 2008; Cruijssen et al., 2007) and can be shared through e-mail or meetings (Jayaram et al., 2011; Malone and Crowston, 1994; Simatupang et al., 2002).
IT is by itself a coordination mechanism. It allows activities of planning, tracking or estimating lead times and enables the exchange of goods among partners (Arshinder et al., 2008). EDI (Electronic Data Interchange) or ERP (Enterprise Resource Planning) systems are some examples. Additionally, information sharing coordination can be mediated by means of IT (e.g. e-mail, fax, Internet) (Arshinder et al., 2008; Kumar and van Dissel, 1996).
Joint decision making or centralized decision making is a way through which members participate to make jointly agreed decisions (e.g. joint consideration of costs, replenishment quantities and frequency, product development) (Arshinder et al., 2008; Heydari, 2014). Similarly to information sharing, joint decision making can be reached by scheduling frequent meetings and using technology (Arshinder et al., 2008; Malone and Crowston, 1994).
7
development of product). I will also consider it a follow-up of the decisions jointly made among the partners and as a separate mechanism in itself.
Researchers argue that the consideration of coordination mechanisms has a positive effect on the SC performance (Arshinder et al., 2008; Fugate, Sahin and Mentzer, 2006). Although not a focal point of previous investigations, impediments that hinder the outcome of the coordination mechanisms are often present. Examples found in literature refer to the overuse, uneven use or opportunistic use of pooled information and resources by partners (Kumar and van Dissel, 1996). Other impeding reasons found by Cruijssen et al. (2007) in literature include: the additional coordinating and controlling costs, and the difficulty to establish a fair allocation of the shared workload and corresponding benefits.
2.2. Clusters as a means of coordination
General literature mentions that coordination initiatives are often hampered by the difficulty to find partners with whom to coordinate (Cruijssen et al., 2007). However the cluster concept suggests that companies should look for partners found in close proximity. A popular approach to clusters is the one of Porter (1998), who states that from a competitive advantage point of view, the immediate business environment outside companies is of strategic importance. Companies ought to create connections with buyers, suppliers, and other institutions (Porter, 1998). In this respect, Porter introduced the concept of clusters and defines them as “geographic concentrations of interconnected companies and institutions in a particular field” (p. 78); the geographic span of a cluster can range from a region, a state, a single city or neighboring countries (Porter, 2000). According to Martin and Sunley (2003), Porter‟s cluster model is seen by policy-makers as a tool for promoting geographical competitiveness, innovation and growth. Other analysts have brought the definition of clusters closer to that of a supply chain. For example, van den Berg, Braun, and van Winden (2001) characterize them as localized networks of organizations, with closely linked processes through the exchange of goods, services, information and/or knowledge.
8
contacts and exchanges between members of the cluster, thus mediate a propitious setting for coordination.
In an attempt to deconstruct the cluster concept, Martin and Sunley (2003) state that, although it was argued that clusters raise the productivity, innovativeness, competitiveness, profitability and employment in the corresponding geographical area of clusters, there is no clear evidence for these claims. Moreover, they state that the boundaries set by clusters may in fact damage coordination between policies and groups and that clusters carry potential disadvantages (e.g. cost increases, over-specialization, institutional and industrial lock-in, over-reliance on localization for performance) (Martin and Sunley, 2003). Coupled with the previously mentioned horizontal coordination difficulties, the concept of cluster coordination becomes problematic. Also, other settings might add to this complexity. Take for example the HL context, which until recently received little attention; therefore there are many aspects that need enlightening.
2.3. Cluster coordination in humanitarian logistics
Relief management in sudden onset natural disasters represents the perfect context to study SC coordination, due to the ever increasing number and complexity of disasters, and the growing number of organizations engaged in relief operations, which made coordination of actors both important and challenging (Jahre and Jensen, 2010; van Wassenhove, 2006). In this context, HL was recognized as having an essential role (Kunz and Reiner, 2012), since a staggering proportion (80 per cent) of the relief management accounts for the logistics operations (van Wassenhove, 2006).
2.3.1. Humanitarian logistics
HL in particular has attracted increasing interest in the recent years, especially after the events of the Asian tsunami in 2004, when the logistic response was regarded as being inefficient. Thomas and Kopczak (2005) stated that HL faces several challenges, one of which is the limited coordination among major logistics organizations. There is a general belief in HL literature that coordination between humanitarian organizations is beneficial because it helps increase the impact or efficiency of the overall operation (Schulz and Blecken, 2010). Also, the lack of, or sub-optimal coordination leads to a wasteful use of resources and time, particularly important in a relief operation (Schulz and Blecken, 2010).
9
(Kaynak and Tuğer, 2014). In the context of HL, Balcik et al. (2010) found that operations following immediately after a disaster focus largely on procurement and transportation, therefore I will investigate coordination within these functions in particular.
Transportation is a major component of the disaster relief operation (Balcik et al., 2010). It can be extremely challenging especially on the „last-mile‟ distribution, the final stage of a humanitarian relief chain that refers to the delivery of relief supplies from local distribution centers to beneficiaries, due to damaged infrastructure, limited transportation resources and high amounts of supplies to be transported (Balcik et al., 2008). Commodities such as food, clothing, medicine, medical supplies, machinery and also personnel need to be transported from different points of origin to different destinations of the affected area quickly and in an efficient manner (Haghani and Oh, 1996).
Also, in the immediate response phase of natural disasters there is an abundance of aid agencies (Long and Wood, 1995). Therefore, it is often unknown which resources are available, and the level of involvement of each participant (Tomasini and van Wassenhove, 2004). Supply procurement in the post-disaster environment meets challenges when considering the time-consuming processes involved (Balcik et al., 2010).
A common denominator of all activities in disaster relief is information (Bui, Cho, Sankaran and Sovereign, 2000). Especially in the response phase, accurate and timely information related to the needs, response requirements and gaps in national capacity to handle the crisis are imperative (Harrald, Cohn and Wallace, 1992, in: Bui et al., 2000). Considering that relief organizations make decisions based on the available information, and that these decisions affect the development of affected areas (Bui et al., 2000), it is natural to think that information sharing is imperative.
2.3.2. Humanitarian Cluster Approach
10
examine it (Bui et al., 2000). Also, cultural (language) and technical incompatibilities limited the information exchange, causing a faulty coordination (Bui et al., 2000; Schulz and Blecken, 2010).
The CA was set up to resolve these issues and was defined as “a system of coordination in which a lead organization, designated for priority areas of response, is responsible for organizing coordination at global and country level, strengthening global preparedness, developing global guidance and acting as provider of last resort” (IASC, 2010; p. 17).
The CA was introduced to improve coordination between organizations involved in the same area, hence a horizontal coordination, as defined earlier (Jahre and Jensen, 2010). It is based on a top-down (centralized) coordination system, in which a central authority coordinates the network (Sahin and Robinson, 2002). In general, the leadership aspect of coordination was considered to be a crucial aspect of the CA and beneficial, considering that the designated lead agency has the responsibility of its own performance and the entire cluster as well (Stoddard et al., 2007). Also, in relief operations, situations tend to worsen when participants are not sure about their role and act independently, without consulting or coordinating with the others (Bui et al., 2000), thus a central authority is needed. Also, in a centralized system the central authority can be delegated with certain decision rights (Alonso, Dessein and Matouschek, 2008).
According to Jensen (2010) the goals of the humanitarian cluster system refer to coordination mechanisms identified in general literature: to facilitate the coordination between members, encourage joint working, collate and share information, identify gaps in the response and be the provider of last resort. Coordination and joint working are facilitated through the establishment of global cells and the attendance at on-site meetings (Jensen, 2010). Additionally, these mechanisms give the opportunity for members to share and disseminate information, to discuss relevant issues and make joint decisions. While the cluster leads have the possibility to make sectoral decisions, joint working or joint decision making is preferable for the smooth functioning of clusters (Jensen, 2010).
11 2.4. Conceptual Framework
There is a growing interest in literature to better understand the systematic effort of humanitarian logistics coordination (physical flows and corresponding information). Literature research so far fell short on achieving a greater understanding of the humanitarian cluster setting. It is in these circumstances that I corroborate the efforts in literature so far and analysis of the growing experience of the LC embedded in organizational reports, with the aim of understanding the corresponding benefits and constraints.
From the literature review, I found that in the case of sudden onset disasters, a significant part of the logistics coordination efforts is concentrated in the immediate response stage of the disaster relief. The boundaries of this research are logistics activities covering the areas of transportation and procurement, as suggested by Balcik et al. (2010). The analysis of coordination operations will be centered on specific mechanisms, identified earlier in the theoretical section, namely information sharing, IT, joint decision making and joint working.
This study makes no a priori hypothesis with regards to the type of mechanisms used for the coordination of each function or to the lower-level means used to apply these mechanisms. Nonetheless, literature gives the advantage of already mentioning several sub-types (e-mail, Internet, meetings, decision delegation), which I will use as a starting point, but will not limit the research. It is in the aim of this research to reveal the relationship between the logistics functions and the coordination mechanisms and see whether specific mechanisms work or not for the coordination of a certain function.
Consequently, the above mentioned logistics functions, cluster coordination mechanisms and outcomes are the variables of the conceptual framework (Figure 1) which will be used to answer the proposed research question. This theoretical framework is confronted with the reality of a case in order to identify the implication of coordination in the humanitarian cluster system.
3. Methodology
12
2002), one of the objectives of this paper. Single in-depth case study is recommended for gaining greater depth for explorative purposes (Voss et al., 2002). Thus, it helps gain a deeper understanding of HL‟s coordination operations, mechanisms, and outcomes in terms of corresponding benefits and constraints. The study analyses the Logistics Cluster in a natural disaster instance, more precisely the Haiti earthquake from 2010, and its disaster management coordination mechanisms during the response phase. This disaster was chosen on the condition that it made possible the investigation of coordination between relief agencies in the LC (horizontal coordination). Also, due to the high media coverage, the involvement of many actors (900 organizations were active in Port-au-Prince at the end of March) (R#13) and the subsequent attention from practitioners and researchers, it was possible to access a wide range of sources of evidence (e.g. reports, media news and articles) that can provide increased validity to the research and prevent the results from being biased (Edmondson and Mcmanus, 2007; Voss et al., 2002).
Figure 1. Conceptual framework showing the potential relationships between variables to be explored
3.1. Introducing the case
The LC is particularly appropriate for this research, since it was founded to strengthen humanitarian coordination at the logistics level. WFP was designated lead agency, partly to address the concerns about the coordination and capacity arising from the critical findings of the Humanitarian Response Review (WFP, 2012). LC‟s services and support come to the aid of not only the broader humanitarian community
Logistics
functions
Transportation
Procurement
Cluster
coordination
mechanisms
Information sharing
IT
Joint decision making
Joint working
Outcome
Benefits
13
by filling identified logistics gaps that affect the supply chain (provider of last resort), but also facilitate coordination between its members (Logcluster, 2012). The services of the LC vary in nature and scale depending on the needs of each operation. They include: coordination, information management services (e.g. infrastructure assessments, port and airport coordination, information about transporters and rates, customs and suppliers of equipment), common means of transport (air, ocean and surface), carrying out procurement operations on members‟ behalf (Schulz and Blecken, 2010), storage (Logcluster, 2012) and planning and language translation (Stoddard et al., 2007). Consequently, the LC offers the appropriate setting for the examination of logistics coordination. The in-depth study is based on the LC‟s involvement in the Haiti earthquake disaster (2010).
Haiti is one of the poorest nations in the Americas (Yates and Paquette, 2011) and ranks 161 out of 187 countries in the 2012 United Nations Human Development Index (WFP, 2014). It is characterized by high levels of poverty, weak governance structures, organized crime, sporadic outbreaks of violence and an extraordinary level of environmental depletion (ReliefWeb, 2010). Throughout the years it suffered a number of disasters, including four hurricanes and storms in 2008 (BBC, 2010). The disaster on January 12, 2010 highlighted Haiti‟s vulnerability to natural disasters. The devastating earthquake hit Haiti with a magnitude of 7.0 on the Richter scale (ReliefWeb, 2010). More than 222,000 people were killed and 3.7 million affected, according to EMDAT.
The earthquake caused widespread damage. Initial reports indicated a high number of casualties and heavy damage, including looting to Port-au-Prince, the capital. The country‟s infrastructure was severely damaged and many buildings were destroyed (see R#15 in Appe, R#8), thus relief efforts were severely hampered. Initially, since Haiti's main airport was out of action, aid agencies had to use the Dominican Republic's capital Santo Domingo as their hub to bring in aid (BBC, 2010). The major port in the city was also severely degraded and no shipping could safely enter the port due to collapsed piers and debris (Yates and Paquette, 2011). The LC was present at the scene immediately, within 24 hours personnel was deployed to set up the LC in Haiti and by 15th January a concept of operations had been arranged (Whiting, 2010).
14 3.2. Data collection
Data for the case study is secondary data and was gathered through desk-based research and exploratory content analysis (Edmondson and Mcmanus, 2007) of agency reports (25 documents), meeting minutes (25 documents) and other online documents made available on several official websites of organizations (LC, IASC, WFP), disaster relief related websites supported by humanitarian organizations (ReliefWeb), general media and articles in the media (BBC), and other online documents. In case study research, it is particularly important to pay attention to reliability and validity (Karlsson, 2009). Validity will be insured through the use of multiple sources of evidence, as previously mentioned (Karlsson, 2009), and reliability of data will be certified by means of using reliable sources, and in this case the organizations provided various insights and trustworthy information, due to their direct implication in the disaster relief operations. The LC‟s official website provided timely and regular updates of the situation, through meeting minutes, situation reports, operational overviews and infrastructure layout maps.
In addition to organizations‟ reports and updates, one disaster and crisis news website was consulted, ReliefWeb. The reliability of this source stems from the fact that it is a digital service of OCHA. It targets humanitarians who need constant and reliable information in order to better understand disasters and make informed decisions. The information provided includes reports, maps and data collected from multiple global sources, making it relevant for the present study. Data collection took the form of a series of keyword searches in the above mentioned sources. The following keywords were used for the Haiti case: „Haiti‟, „earthquake‟, „Logistics Cluster‟.
Due to the great global impact of the Haiti earthquake, in the following years several retrospective reports were released and some of them were found relevant to the present study. Two of them are official reports from UN that evaluate the CA; they contain a detailed overview of the progress of clusters in general and in particular operations.
15
starting with April the earthquake response phases into recovery. Appendix E lists the names and types of documents included in the research.
3.3. Data analysis
This research aims at answering the research question through a qualitative content analysis of data, with the support of the coding software Atlas.ti. The case analysis followed the process of coding presented by Miles and Huberman (1994). The first step consisted of applying a coding approach in order to filter out data relevant in terms of the logistics functions of inter-cluster coordination (i.e. transportation and procurement) included in the framework (see quotations in Table 1 for representative examples of first-order coding). For instance, statements referring to purchasing, buying, sourcing of goods or supplies were coded under the category „Procurement‟ (second-order coding). Then, I examined first-order coding in relation to the coordination mechanisms. “A priori coding” was applied at this stage, meaning that the categories were defined before the analysis, as per the Theoretical Background and are included in the Theoretical Framework (Weber, 1990). To give an example, the first-order code “The Logistics Cluster is planning to transport 12,000 gallons of fuel from Santo Domingo into Port-au-Prince every second day” was assigned to information sharing, while the phrase “130 - 140 smaller 1.5 mt to 2.5 mt vehicles have also been received and will be registered and insured by WFP for use by organizations with mid- to long-term projects in Haiti” was assigned to joint working. This allowed highlighting the analogy between the logistics and corresponding coordination mechanisms.
In short, the coding scheme consisted of the following:
Two logistics functions corresponding to immediate response disaster management stage: transportation and procurement;
Four coordination mechanisms identified in the theoretical background: information sharing, IT, joint decision making and joint working.
Table 1. Coding scheme
First-order codes: representative data Second-order categories:
logistics functions “The Logistics Cluster is planning to transport 12,000 gallons of fuel from Santo
Domingo into Port-au-Prince every second day. This can be augmented if required. Organizations in need can request drums which they will get at cost during this initial period of the emergency.” (R#23)
Procurement
“Depending on demand, two aircraft are available to provide transport up to a maximum of 86 passengers each way. Schedule available at:
http://www.logcluster.org/ops/hti10a/unhas-flight-schedule-sdq-pap-23-january-2010”
(R#25)
16 4. Results
The following sections will present the findings evidenced by the data analysis. The findings are presented to show the outcome of LC‟s coordination mechanisms in the two pivotal logistics functions for the relief response.
The results are presented by showing how organizations within the LC coordinate their logistics activities. Table 2 highlights the congruence between the logistics functions and the high-level coordination mechanisms (information sharing, IT, joint decision making and joint working), while showing the means through which these mechanisms are carried out, that were not identified „a priori‟. The following sections will shed more light on each coordination mechanism.
Table 2. Summary of findings
Mechanism Coordination means
Logistics function Transportation Procurement Information Sharing Meetings E-mail Internet IT
Joint decision making Meetings
Joint working
Meetings
E-mail
Internet
Notes: () relationship present
4.1. Information sharing
“Without information sharing there can be no coordination. If we are not talking to each other and sharing information then we go back 30 years” – Ramiro Galvez, UNDAC (R#1; p. 10).
17
In light of this situation, Dennis King, US Department of State Humanitarian Information Unit (R#1, p. 16) mentioned: “During the first 12-24 hours, there is a fog of information. The situation is changing rapidly and the numbers are changing rapidly. You might as well watch CNN for the changing situation, because any product that we did was quickly out of date and overtaken by events”. In the aftermath of the event, OCHA (R#1) found that information sharing within the cluster was mainly affected by the structural issues of the information management design, the lack of necessary resources (staff, funds and technology) and delays due to translation, collation and analysis. Also, the country suffered severe destruction to critical infrastructure, such as electricity and telephone services. Electricity and communications continued to be difficult quite long after the disaster (R#15) and Internet was a viable but limited communication channel (BBC). Table 3 gives an overview of the findings with respect to information sharing coordination put in practice by the LC in response to the Haiti earthquake.
Documents revealed that much of the information sharing was done through the use of IT (e-mail and Internet) (MM#1-MM#25). Much of the information was related to transportation (e.g. resources, capabilities and needs), but little information was reported with regards to procurement. Mentions were usually related to the limited fuel supply, issue which was becoming serious. The price for fuel in Haiti has increased greatly and organizations were advised to procure their own fuel (R#5).
18
Table 3. Overview of findings for information sharing coordination
Mechanism Details Benefits Constraints
Inf or m a ti on shar ing Me et ing s
- 150 organizations attended at least one LC meeting in Port-au-Prince until the 28th of March (R#3)
- 13 meetings were held in Santo Domingo until 31st of March (R#16) - 24 meetings were held in Port-au-Prince until 25th of March (R#4)
- information sharing - discuss and clarify issues -receive immediate feedback - time consuming (R#12, R#18)
- important groups excluded because of language constraints
- location was not of easy access for the participants (R#12, R#18)
- assumes incredible memory on the part of the participants (R#1)
E
-m
a
il
- a Haiti LC mailing list was set up to ensure flow of information within the
humanitarian community: Haiti.logs@logcluster.org (R#4) - information centralization Int er n et
- a dedicated webpage with daily information on the Haiti response was available on the LC site: http://www.logcluster.org/ops/hti10a (R#4)
- the official sites of ReliefWeb, OCHA and WFP were in the top 10 websites referring to the LC (R#7)
- the highest single day of website traffic was 1,901 visitors on January 19th (R#7)
- the majority of the LC‟s website use is attributed to the main administrative offices of organizations, not by the staff in the field (R#7)
- in the first three months, 29 situation reports, 4 bulletins, several in-depth snapshots on operations, procedures and infrastructure were published on LC‟s website (R#4)
- WFP made available maps of affected areas (R#17) - LC site accessible to everyone -timely provision of important information - updates on security issues - provided unstructured documents, which could not be used to pull structured data or turn them into actionable pieces of information (R#1)
- field staff could not afford to download large files; people started using paper and data formats that could work offline
- poor access to Internet (R#1)
- fragmented information, forcing the staff to put the pieces together; few had the available time and bandwidth to do so (R#1)
19
In the field, most information sharing took place in cluster meetings. For people who could not attend the meetings, the method of exchanging key assessment metrics was through e-mails or reports shared on the LC website. Throughout the operation there were many meetings organized at the two LC primary locations (Port-au-Prince - Haiti and Santo Domingo - Dominican Republic). In the first weeks of the operations, there were daily meetings of LC and starting with 22nd of January the meetings were held every two days. Meetings make it possible for organizations to communicate the gaps they encountered on the ground (Jensen, 2012). Major decisions were made during the meetings, which were held in French (one of the official languages of Haiti). There was no translation for the other participants who spoke only English. Also, most of the information and documents created were in English. This created a conflictive environment and as a consequence some agencies did not return to the meetings (R#11, R#12).
4.2. Information technology
In Haiti, the LC had dedicated information management capacity, such as GIS (Geographic Information Systems) mapping services for the coordination of logistics operations and the decision-making support. More specifically, it leveraged the consolidation of logistics information from organizations and local authorities, the identification of logistics gaps and bottlenecks, the facilitation of LC meetings and publishing of records of decisions taken, the provision of updated maps to facilitate decision-making for transportation, coordination of the use of common logistics services (i.e. cargo movement requests, storage requests and cargo tracking services), provision of a platform for the sharing and dissemination of logistics information, operational data (e.g. customs procedures, air operations and prioritization of cargo) and the publication of situation reports, bulletins, snapshots, flash news and briefings.
4.3. Joint decision making
Many participants in the global-level cluster have expressed concerns that WFP is not sufficiently consultative and lacks a participatory approach to joint strategic decision-making. The consequence of this situation was that the accountability and value of the cluster in the eyes of participants was undermined (Stoddard, 2007).
20
150 attend at least one meeting (R#3). Also, it was reported that the LC lacked the involvement of appropriate national actors (R#7).
4.4. Joint working
Joint working was declared as desirable and essential for the proper functioning of the cluster (Jensen, 2012). Initiatives for joint working in transportation, such as coordinated vehicle routing, implied that organizations delegated the LC to realize many of the transportation activities. This made it possible for agencies to use full truck-loads, which implied less overall costs. The organization that conducted these services was Handicap-Atlas International and was appointed by the LC. It was present in all transportation nodes (airports, ports, road crossing) from Dominican Republic and Haiti and was running a large fleet containing terrain trucks, to provide free interagency transport (R#9). The LC‟s presence was regarded by some stakeholders as being very dynamic and effective (R#10). Others considered the distribution to being problematic and slow-paced (R#10). Some of the delays were attributed to security issues. For the coordination of security requests, the LC delegated The Joint Operations Tasking Centre (JOTC). These restricted the movement of trucks. The Cluster‟s trucks ran from the border to the suburbs of the capital in escorted convoys (R#10). Sometimes, transportation had to be done with alternative means. For procurement, the LC made arrangements to contract a fuel supply for common logistics services. It was also involved in tracking the availability and prices of local supplies and equipment, especially the cost of hire cars and houses, which were highly demanded (R#10). The contribution of LC in this regard was considered to be very useful (R#10).
For procurement and transportation coordination, an example of joint working between organizations is represented by trainings. This was due to the lack of staff training and appropriate skill sets, which were issues mentioned in several evaluations (R#11). During the immediate response, it became evident that with the pre-existing low level of education in Haiti and the large number of deaths among the workforce, there was a huge lack of suitable local staff especially in the logistics field. Although the importance of trainings was recognized, it encountered a lack of funding from donors (Heraty, 2010).
21
mentions, joint working led to reduced operational duplication and redundancy, which are important as they bring cost savings (Stank, Keller, Daugherty, 2001). Table 3 gives an overview of the findings with respect to joint working coordination.
Table 4. Overview of findings for joint working coordination Details D el ega ti on D el ega ti on D el ega ti on
- inside Haiti, transportation by trucks was done through Handicap-Atlas International, which had a fleet of over 56 trucks, which could be used by all NGOs (R#10)
- 71 different organizations have utilized the LC facilitated delivery system in Port-au-Prince - until 15th of February, 42 several agencies delegated the LC to carry out trucking transportation from Dominican Republic to Haiti, a total of 450 truckloads (R#10)
- transportation coordination is also mediated by personnel deployed at borders to coordinate and facilitate the crossing (e.g. at the Jimani border in Dominican Republic); this was done to resolve logistics bottlenecks and improve the efficiency for humanitarian cargo entering Haiti (R#12)
- air passenger services to and from Santo Domingo were established by the UN Humanitarian Air Service (UNHAS), at an early stage; by mid-February it had carried nearly 2,500 people from more than 200 different organizations (R#10, R#12)
- 85% of all relief and reconstruction materials were transported by road (R#12)
- in February, a big procurement program for transitional shelters began and the LC offered to co-ordinate; surprisingly, the operation was done by the Shelter Cluster (R#10)
T
rai
nin
g
- trainings related to warehouse keeping, driving, vehicle maintenance, distribution and local procurement, and, to a lesser extent, logistics supervision and management at the logistics assistant level were targeted (WFP, 2012)
- trainings for logistics cluster coordinators were primary gaps identified (R#2)
- numerous references to a lack of French and Creole skills included in evaluations (R#11) - ensured a coordinated and comprehensive response to the emergency
- training courses were widely appreciated for their high quality and contribution for inter-agency coordination (R#2)
- had moderate to high effects on enhancing participants‟ knowledge, skills and ability to effectively coordinate within the cluster (R#2, R#7)
C oordi na ti on m eans Int er n
et - requests for delegating transport activities were made by filing a special form (cargo movement requests - CMRs) available on the LC website:
http://www.logcluster.org/ops/hti10a (R#9)
- guidelines for how to fill the form were available on the LC website (MM#5)
E
-m
a
il - requests for joint cargo movement were also done through requests sent via e-mail
(MM#25)
Me
et
ing
22 5. Discussion
This research first builds on the literature review to identify coordination mechanisms which are then tested through an in-depth case study research to develop a framework for coordination within the humanitarian logistics cluster in the setting of a sudden onset disaster. The main goal of this analysis was to reveal coordination benefits and constraints by exploring the relationships between the logistics functions and coordination mechanisms and to identify the specific means through which these mechanisms are put in practice in order to deepen the understanding of the corresponding benefits and constraints. Based on the case study evidence, a revised framework is proposed (Figure 2). Three remarks accompany this framework:
(1) Not all the possible relationships within the theoretical framework were found after the data analysis. While this research was directed towards the investigation of transportation and procurement coordination, the analysis revealed that procurement coordination was rather disregarded, whereas transportation received increased attention. The main procurement information was related to fuel supply. This was mainly because the LC does not serve as a procurement service provider for partners, but is only involved in producing consolidated supplier and vendor information (R#2). However, the WFP 2012 report supports the idea that a primary gap identified by partners was precisely procurement.
(2) Some findings and relationships between variables (relationships are depicted in the framework through arrows) have confirmed theories in general literature, while others are new empirically grounded findings that emerged from data analysis. The latter category of relationships will receive additional attention, through the formulation of propositions, unfolded in the following sections.
23
Figure 2. Proposed framework showing the relationships between according to the analysis
5.1. Information sharing
Often disaster response is characterized by lack of reliable information (Kovacs and Spens, 2007). Therefore, frequent updates and changes are necessary. Information sharing provides the ability to quickly adapt to changes, discover disruptions in response and make better decisions (Christopher & Peck, 2004), while growth of IT and communication capabilities such as the Internet enhance the ability to integrate SC members (Stank, Keller and Daugherty, 2001).
The analysis reveals that the means put in place for information sharing coordination are meetings, e-mail and Internet, making the findings in line with literature (Malone and Crowston, 1994; Simatupang et al.,
Logistics
functions
Transportation ProcurementCluster coordination
mechanisms
Meetings Meetings Meetings Information sharing Internet Internet E-mail E-mail ITJoint decision making
Joint working
Outcome
Benefits:
- security information - trainings increase skills
Constraints:
- limitation of joint decision making
24
2002). However, information sharing coordination after the Haiti earthquake did not take place in propitious circumstances. As a result of the earthquake, Haiti suffered severe destruction to critical infrastructure, such as electricity and telephone services. Electricity and communications continued to be difficult quite long after the disaster (R#15) and Internet was a viable but limited communication channel (BBC). Nonetheless, it was important that mechanisms easily accessible for all participants would be in place, including the field staff, local community and people affected. For Haiti, it turned out that this was not the case, since a great part of information sharing was done through the use of IT (e-mail and Internet). Consequently, I conclude the first proposition.
Proposition 1. When information sharing is done through means that do not take into consideration the existing local conditions and resources, it becomes a constraint for an effective response.
Due to costs restrictions, field staff had trouble downloading needed documents made available online. Therefore, they started using paper and other data formats that could work offline. A frequent issue was that data analysis and tracking were very difficult (e.g. maps were in an inadequate format and could not be used by other information systems). Also, the scattered information required them to put the piece together, which involved time and costs (R#1). Hence, the second proposition emerges.
Proposition 2. When information is not in an appropriate format, it becomes a constraint for coordination effort.
Several participants in the operations considered distribution to being problematic and slow-paced. Some of the delays were attributed to security issues, which restricted the movement of trucks (R#10). And sometimes, transportation had to be done with alternative means. Security information is of utmost importance for distribution, considering that humanitarian organizations generally operate in an insecure environment where ransacks can frequently happen (Liberatore, Ortuno, Tirado, Vitoriano and Scaparra, 2014). This leads to a third proposition.
Proposition 3. Sharing information on security issues is beneficial for the unfolding of operations.
25
should have been restricted and the meetings‟ topics should have been limited to a clear purpose, considering that alternative means of communication were present.
Proposition 4. If too many meetings take place as a means of information sharing, that becomes a constraint.
5.2. IT
Literature mentions that IT can be by itself a coordination mechanism which allows activities such as planning and tracking (Arshinder et al., 2008; Kumar and van Dissel, 1996). Data analysis uncovered that an IT capability evident in the response phase of Haiti was represented by the GIS information management capacity, used for the coordination of logistics operations and the corresponding decision-making support. However, these capabilities have been regarded as limited and restrictive. Also, initiatives in the form of EDI and ERP systems, especially helpful for transportation and procurement coordination, were not evident in the response phase of Haiti. Consequently, the lack of important IT resources was a cause for a constricted coverage of coordination.
Proposition 5. A lack of IT capabilities can support a flawed coordination.
5.3. Joint decision making
Literature mentions that language incompatibilities can limit the information exchange, causing a faulty coordination (Bui et al., 2000; Schulz and Blecken, 2010). This analysis revealed that language can be a barrier for joint decision making. Due to the fact that not all organizations‟ members could communicate in the same language, some organizations did not get involved in coordination efforts with the other humanitarians and did not attend meetings. This led to fragmented humanitarian effort. Thus, the following proposition can be reasoned.
Proposition 6. Language can constitute a huge constraint for an effective and efficient coordination; if communication and information sharing is not done through a common language for all or at least a great majority of humanitarian organizations, thus leading to a faulty relief intervention.
Meetings have been identified as a means for logistics organizations to make joint decisions. The unfolding of events showed that meetings focused more on information sharing. It becomes obvious then that the opportunity to focus more on action and active decision making was wasted.
26 5.4. Joint working
The novelty of this research consists of identifying an approach, namely trainings, that the humanitarian community can use in order to improve coordination and overcome a factor than limits successful coordination and joint working, which is the lack of skills of staff involved in relief operations. In the case of Haiti the LC faced with important gaps in what concerns staff‟s logistics and language skills. Reports revealed the necessity for more dedicated, trained and experienced coordinators (R#11).
Proposition 8. In the logistics field trainings can fill the gap of lack of skills and experience and help coordination efforts.
6. Conclusions, limitations and further research 6.1. Implications for research and practice
With the increased frequency of sudden onset disasters and the serious outcomes in the form of widespread human, material, economic and environmental losses, the humanitarian community recognized the importance of logistics coordination in the relief response and have become actively engaged in building a tighter and more prepared community in the form of clusters. While clusters in commercial chains received great interest, scholars only recently turned their attention to humanitarian clusters. Many aspects related to the specificities of coordination in the HL cluster need enlightening. This paper concentrated its attention over the particular coordination mechanisms applied by the LC and their corresponding benefits and constraints. Consequently, several theoretical and managerial contributions stand out in this research.
Firstly, the main research contribution of this paper is an empirically derived framework for cluster coordination in the HL cluster. It shows the coordination mechanisms, the means through which they are carried out and the outcomes of their applicability.
Secondly, in the middle of the struggle of literature to grasp more information on what issues limit coordination efforts in reaching a beneficial outcome, this research shows that the coordination mechanisms put in place can have a negative impact. For example, information sharing can act as a constraint when it is done excessively, without a clear purpose and without taking into consideration the resources and conditions of people working on the field.
27
disadvantageous, information that literature failed to provide so far. Meetings for instance, when too many and unstructured, can affect joint decision making.
Insights obtained from this research reveal implications for practice as well. There is much room for improvement with the increased requirements from the larger community in general and donors in particular. The deduced framework depicting the relationships between coordination mechanisms, coordination means and corresponding outcomes could be of use for practitioners who continually struggle to make cluster coordination beneficial for the entire community. The relief management of natural disasters particularly implies very critical decision-making. Practitioners can extend the knowledge derived in this study and further investigate how it can help to strengthen the functioning of clusters in HL.
6.2. Limitations and further research
Unquestionably, there is much more research necessary to test and scrutinize the findings of this paper, which presents several limitations.
28
29 Bibliography
1. Adinolfi, C., Bassiouni, D.S., Lauritzsen, H.F. and Williams, H.R. 2005. Humanitarian Response Review. Office for the coordination of Humanitarian Affairs
2. Alonso, R., Dessein, W. and Matouschek, N. 2008. When Does Coordination Require Centralization?. American Economic Review, 98 (1): 145–179.
3. Altay, N. and Green, W.G. 2006. OR/MS research in disaster operations management, European
Journal of Operational Research, 175 (1): 475– 493.
4. Altay, N. and Labonte, M. 2011. Humanitarian Logistics and the Cluster Approach: Global Shifts and the US Perspective. In Christopher, M.G. and Tatham, P.H. (eds.), Humanitarian Logistics:
Meeting the Challenge of Preparing for and Responding to Disasters: 85-101. London: Kogan
Page.
5. Arshinder, K., Kanda, A. and Deshmukh, S.G. 2008. Supply chain coordination: perspectives, empirical studies and research directions. International Journal of Production Economics, 115 (2): 316–335.
6. Balcik, B. and Beamon, B.M. 2008. Facility location in humanitarian relief. International
Journal of Logistics: Research and Applications, 11 (2): 101–121.
7. Balcik, B., Beamon, B.B., Krejci, C.C., Miramatsu, K.M. and Ramirez, M. 2010. Coordination in humanitarian relief chains: Practices, challenges and opportunities. International Journal of
Production Economics, 126 (1): 22-34.
8. Balcik, B., Beamon, B.M. and Smilowitz, K. 2008. Last mile distribution in humanitarian relief.
Journal of Intelligent Transportation Systems, 12 (2): 51–63.
9. Barbarogosġlu, G, Ozdamara, L. and Cevik, A. 2002. An interactive approach for hierarchical analysis of helicopter logistics in disaster operations. European Journal of Operations Research, 140 (1):118-133.
10. Bui, T., Cho, S., Sankaran, S. and Sovereign, M. 2000. A Framework for Designing a Global Information Network for Multinational Humanitarian Assistance/Disaster Relief. Information
Systems Frontiers, 1 (4): 427-442.
11. Cao, E., Wan, C. and Lai, M. 2013. Coordination of a supply chain with one manufacturer and multiple competing retailers under simultaneous demand and cost disruptions. International
Journal of Production Economics, 141: 425-433.
30
13. Chan, H.K. and Chan, F. 2010. A review of coordination studies in the context of supply chain dynamics. International Journal of Production Research, 48 (10): 2793-2819.
14. Christopher, M. and Peck, H. 2004. Building the resilient supply chain. International Journal of
Logistics Management, 15 (2): 1–13.
15. Cruijssen, F., Cools, M. and Dullaert, W. 2007. Horizontal cooperation in logistics: opportunities and impediments. Transportation Research Part E: Logistics & Transportation Review, 43 (2): 129-42.
16. Edmondson, A.C. and Mcmanus, S.E. 2007. Methodological fit in management field research.
Academy of Management Review, 32 (4): 1155-1179.
17. EM-DAT, The International Disaster Database (Centre for Research on the Epidemiology of Disaster – CRED, available at http://www.emdat.be/ (accessed at 17.02.2013)
18. Eisenhardt, K.M. 1989. Building theories from case study research. Academy of Management
Review, 14: 532-550.
19. Eisenhardt, K.M. and Graebner, M.E. 2007. Theory building from cases: opportunities and challenges. Academy of Management Journal, 50 (1): 25-32.
20. Logistics Cluster - Service Portfolio, published at 12.04.2012, available at: http://logcluster.org/document/service-portfolio (accessed at 15.02.2014)
21. Fugate, B., Sahin, F. and Mentzer, J.T. 2006. Supply chain management coordination mechanisms. Journal of Business Logistics, 27 (2): 129-161.
22. Haghani, A. and Oh, S.C. 1996. Formulation and solution of a multi-commodity, multi-modal network flow model for disaster relief operations. Transportation Research Part A: Policy and
Practice, 30 (3): 231–250.
23. Harrald, J.R., Cohn, R.. and Wallace, W.A. 1992. We were always re-organizing...: some crisis management implications of the Exxon Valdez oil spill. Industrial Crisis Quarterly, 6: 197-217. 24. Heydari, J. 2014. Coordinating supplier's reorder point: A coordination mechanism for supply
chains with long supplier lead time. Computers & Operations Research, 48:89–101.
25. Humanitarian Response.info (OCHA), available at:
31
26. IRIN. 2008. Chad: mixed verdicts on coordination of massive relief effort. IRIN Humanitarian
News and Analysis; available at:
http://www.irinnews.org/report/76386/chad-mixed-verdicts-on-coordination-of-massive-relief-effort [accessed 23 March 2014]
27. Jahre, M. and Jensen, L-M. (2010). Coordination in humanitarian logistics through clusters.
International Journal of Physical Distribution and Logistics Management, 40 (8/9): 657-674.
28. Jahre, M., Jensen, L-M and Listou, T. 2009. Theory development in humanitarian logistics: a framework and three cases. Management Research News, 32 (11): pp.1008 – 1023.
29. Jayaram, J., Xu, K. and Nicolae, M. 2011. The direct and contingency effects of supplier coordination and customer coordination on quality and flexibility performance. International
Journal of Production Research, 49 (1): 59-85
30. Jensen, L-M. 2010. Humanitarian Cluster Leads as Fourth-Party Logistics Providers. Journal of
Humanitarian Logistics and Supply Chain Management, 2 (2): 148-160.
31. Karlsson, K. 2009. Researching Operations Management. Routledge.
32. Kaynak, R. and Tuğer, A.T. 2014. Coordination and collaboration functions of disaster coordination centers for humanitarian logistics. Procedia - Social and Behavioral Sciences, 109: 432 – 437.
33. Kovacs, G. and Spens, K.M. 2007. Humanitarian Logistics in Disaster Relief Operations.
International Journal of Physical Distribution and Logistics Management, 36 (2): 99-114.
34. Kumar, K. and van Dissel, H. 1996. Sustainable collaboration: Managing conflict and cooperation in interorganizational systems. MIS Quarterly, 20 (3): 279-296.
35. Kunz, N. and Reiner, G. 2012. A meta-analysis of Humanitarian Logistics Research. Journal of
Humanitarian Logistics and Supply Chain Management, 2 (2): 116-147.
36. Liberatore, F., Ortuno, M.T., Tirado, G., Vitoriano, B. and Scaparra, M.P. 2014. A hierarchical compromise model for the joint optimization of recovery operations and distribution of emergency goods in Humanitarian Logistics. Computers & Operations Research, 42: 3–13. 37. Long, D.C. and Wood, D.F. 1995. The logistics of famine relief. Journal of Business Logistics,
16 (1): 213-229.
38. Malone, T. and Crowston, K. 1994. The interdisciplinary study of coordination. ACM Computing
32
39. Martin, R. and Sunley, P. 2003. Deconstructing clusters: chaotic concept or policy panacea?,
Journal of Economic Geography, 3 (1): 5-35.
40. Miles, M.B. and Huberman, A. 1994. Qualitative Data Analysis: An Expanded Sourcebook (2nd Ed.). Thousand Oaks, CA: SAGE Publications.
41. Murray, S. 2005. How to deliver on the promises: supply chain logistics: humanitarian
agencies are learning lessons from business in bringing essential supplies to regions hit by the tsunami. London: Financial Times.
42. Porter, M. 1988. Clusters and the new economics of competition. Harvard Business Review, 76 (6): 77-90.
43. Porter, M. 2000. Location, Competition, and Economic Development: Local Clusters in a Global Economy. Economic Development Quarterly 2000, 14 (1): 15-34.
44. Ramdas, K. and Spekman, R.E. 2000. Chain or shackles: understanding what drives supply– Chain performance. Interfaces, 30 (4): 3-21.
45. Sahin, F. and Robinson, E.P. 2002. Flow Coordination and Information Sharing in Supply Chains: Review, Implications, and Directions for Future Research. Decision Sciences, 33 (4): 505-536.
46. Schulz, S.F. and Blecken, A. 2010. Horizontal cooperation in disaster relief logistics: benefits and impediments. International Journal of Physical Distribution & Logistics Management, 40 (8/9): 636-656
47. Sheu, J. B. 2007. Challenges of emergency logistics management. Transportation Research part
E, 43 (6): 655-659.
48. Simatupang, T.M., Wright, A.C. and Sridharan, R. 2002. The knowledge of coordination for supply chain integration. Business Process Management Journal, 8 (3): 289–308.
49. Sirias, D. and Mehra, S. 2005. Quantity discount versus lead time-dependent discount in an inter-organizational supply chain. International Journal of Production Research, 43 (16): 3481-3496. 50. Stank, T.P., Keller, S.B. and Daugherty, P.J. 2001. Supply chain collaboration and logistical
service performance. Journal of Business Logistics, 22 (1): 29-48.
