1 Introduction

In this article, we aim to systematize selected contemporary literature on humanitarian logistics and humanitarian supply chain management (HLSCM), which has attracted a considerable amount of attention from scholars, practitioners and policy makers alike (Kovacs and Spens 2010). Rising interest in this field has been justified by a myriad of humanitarian challenges that society has faced over the past few years (Dubey and Gunasekaran 2015), examples of which include natural disasters and armed conflicts among others. As the subject of humanitarian logistics and supply chain gains more relevance the literature surrounding it has significantly increased, meaning a systematization of this literature now seems appropriate, as does identifying research gaps for future studies.

Consequently, the contribution of this article is to provide a systematization of selected contemporary works in the relevant literature on humanitarian logistics and supply chain that have been published by journals indexed in Scopus and Web of Science. Inspired by procedures adopted by highly cited literature reviews (e.g. Lage Junior and Godinho Filho 2010), this paper delivers:

  • An identification of the main articles on the field of humanitarian logistics and supply chain indexed in Scopus and Web of Science;

  • A classification of the relevant literature above based on a variety of characteristics; and

  • An original research agenda for future studies, based on gaps found in the current state-of-the-art body of knowledge.

This article is organized as follows. After our introduction (Sect. 1), we present the research methods we used (Sect. 2) to frame the classification and coding system we used to scrutinize the relevant literature (Sect. 3). We then briefly detail our conceptual background in Sect. 4, while Sect. 5 sheds light on our results and subsequent discussion. We conclude in Sect. 6 by presenting an original, new research agenda forward in this field.

2 Research methods

A literature review has as its main objective to show the central structures of a subject or topic, with the aim of identifying research progress that has been made, as well as literature gaps that remain (Hart 1999; Baker 2000). In this context, we use and apply the methodology and steps proposed by Lage Junior and Godinho Filho (2010) and subsequently tested by Jabbour (2013) and Mariano et al. (2015) in our literature review. As such, we observe the following steps:

  • First: Identifying the main articles available on the subject in academic databases and considering the principal keywords related to the topic;

  • Second: Screening the articles found in the first step in order to eliminate articles outside the subject area;

  • Third: Developing and applying a classification system to identify central structures of the subject or topic considered;

  • Fourth: Providing a literature review using the classification system elaborated in the third step above; and

  • Fifth: Identifying gaps, opportunities and challenges regarding future research studies in this area.

For our first step, the main articles about humanitarian logistics we identified during August and September 2016 contained the keywords of “Logistics”, “Supply Chain Management” and “Humanitarian” in the academic databases of Scopus and Web of Science. We chose the two databases because they both compile data about abstracts and citations of scientific journals, books and conference proceedings from fourteen of the largest publishers in the world. Here, we used different combinations of keywords to increase the scope and reach of our search. After this first step, we performed screening with the objective of identifying all articles outside of the scope of our identified topics. Consequently, our final database comprised 87 articles that, in turn, were classified based on the coding presented in Table 1, and described in the following section. Finally, taking into account our proposed coding, a descriptive statistic was used to identify the main gaps remaining from the literature, as shown in Sect. 5.

Table 1 Classifying and coding
Full size table

3 Classification and coding

Considering the method proposed by Lage Junior and Godinho Filho (2010) and Jabbour (2013), we defined a set of classifications to organize the identified articles into specifics groups. This classification set included eight categories numbered from 1 to 8, and for each of them a group of coding was defined and used via letters from A to K. For example, a code of 2B means that this article is classified in section B of category 2, and such articles could be classified in one or more our categories above. Taking such points into account, our eight classifications are briefly defined below as:

  • Classification 1—Economic context: The degree of economic maturity of the countries in which the study occurred, coded from Ato D;

  • Classification 2—Focus: The main theme considered in any study, coded from A to C;

  • Classification 3—Method: The method used in a study, classified and coded from A to K;

  • Classification 4—Type of disaster: The different forms and durations of disasters considered in the study analysed, coded from A to E;

  • Classification 5—Phase of the disaster relief: The more important phase of disaster relief addressed in a study, coded from A to D;

  • Classification 6—Type of humanitarian organization: The types of humanitarian organization addressed by the author of the study analysed, coded from A to E;

  • Classification 7—Region of authorship: The region of authorship, coded from A to E. This category also considers whether there is a significant volume of academic output for a specific region by the authors; and

  • Classification 8—Region of disaster: The region where the disaster occurred, coded from A to F.

All the descriptions of our classifications and codes are shown in Table 1, and the research articles considered in our literature review are detailed in Table 2.

4 A brief conceptual foundation of humanitarian logistics and supply chains

Humanitarian supply chain management (HLSCM) is intimately tied to the broader context of disaster management which itself is a subject of much contemporary popularity. For example, recent works such as Yang et al. (2014), have used Data Envelopment Analysis to build an emergency response network for earthquakes, Anparasan and Lejeune (2017) have proposed a model of emergency responses to epidemics that can be used in countries that have limited resources, and Sushil (2017) has proposed the use of a qualitative and interpretative framework called SAP–LAP in the context of disaster management.

Two subjects in HLSCM can be considered very important as they have been widely studied, namely: humanitarian supply chains (HSC) and humanitarian logistics (HL). During the past decade, HSC has received greater attention among academics and practitioners (Kovacs and Spens 2010), and many HSC works are trying to better explore this subject. Examples include: coordination of HSC (Balcik et al. 2010; Akhtar et al. 2012), specifically studying the drivers and barriers of the coordination of HSCs (Kabra and Ramesh 2015b; Kabra et al. 2015), and developing frameworks to improve HSC implementation (John et al. 2012).

However, it has been observed from the literature, that most HSCs are unstable, unpredictable, and slow to respond to the needs of affected people (Yadav and Barve 2015), especially when related to those disasters. Such disasters not only disturb the normal functioning of society, but can also leave huge and negative impacts on the people directly or indirectly impacted by them. It is not possible to predict natural disasters, but actions can be taken to deal with such complex crises and reduce the impact of natural disasters on people and society (Kovács and Spens 2007; Kabra and Ramesh 2015b; Kabra et al. 2015).

Wassenhove (2006) defines a disaster as a “disruption that physically affects a system as a whole and threatens its priorities and goals”, and considers HSC a central point for at least three reasons, as HSC: (i) serves as a bridge between disaster preparedness and response, between procurement and distribution; (ii) is crucial to the effectiveness and speed of response for major humanitarian programs, such as health, food, shelter, water and sanitation; and, (iii) can be one of the most expensive parts of relief efforts and operations, and thus deserves special attention.

Table 2 Coding of the studies considered (Part 1)
Full size table

In this sense, a very important concept for HSC is resilience, which is the ability of a supply chain to absorb the impacts of any rupture caused by a disaster and to recover from it. Here, DuHadway et al. (2017) developed a framework that allows us to understand the risks caused by intentional or unintentional breaks in HSCs, and Kaur and Singh (2016), in turn, have studied relationships between such resilient supply chains and sustainability outcomes.

In a disaster situation, logistics can be considered a critical activity that differentiates between a successful and a failed relief operation (Cozzolino et al. 2012). Indeed, the impact of a disaster is mostly seen on human mortality and their livelihood perspective, but a huge loss to economies is also associated with such disasters too (see Yadav and Barve 2015). From this perspective, disaster management and relief aid require complex logistical activities, as the resources they need are rarely available at the location of the disaster. These logistical activities are generally referred to as HL (Kunz et al. 2014).

An important point raised by Oloruntoba et al. (2016) is the fact that HLSCM still lacks theoretical development, and they suggest using the theories of behavioral and organizational economic internationalization to further progress it. Coles et al. (2017) emphasize the gap between theory and practice in the area of disaster management, and to remedy such failure have conducted a review of the International Federation of Red Cross and Red Crescent Societies (IFRC) to do so.

5 Results and discussion

The purpose of this section is to understand the research results obtained from our classification of the articles described in Table 2. Our results are presented through use of the defined categories detailed in Table 1 in the following subsections of: economic context, focus, methods, disaster type, relief phase, organizational type, author region, article purpose and interesting research gaps. We begin with economic context, and then detail each of the subsections above in turn.

5.1 Economic context

Based on the articles that were classified in categories 1A, 1B, 1C and 1D shown in Fig. 1, most such studies did not analyse a specific region (52%), and those which did investigate a region primarily analysed non-mature economies (33%). According to The United Nations Office for Disaster Risk Reduction (UNISDR 2016), the main countries affected by disasters in terms of people killed between 1992 and 2012 were Haiti, Indonesia and Myanmar. Considering that a natural disaster, in particular, would occur again due to geographic conditions, a research gap in this topic is:

\({ GAP}_{1 }\) Which lessons would be learnt from non-mature economies in order to foresee, and be prepared for natural disasters?

Fig. 1
figure 1

Results obtained for Category 1

Full size image

5.2 Focus

Our second category identifies the main focus addressed in the previously considered articles in our literature review on HL and HSC. Here, relevant articles are classified in category 2A—SCM, 2B—Logistics, and in category, 2C—Both. As shown in Fig. 2, results indicate that most of these studies (54.02%) focus mainly on logistics rather than considering the area of Supply Chain Management, or the interface between these two knowledge areas. One probable explanation for this result is that after a disaster, the transportation of injured people and moving supplies to devastated areas is one of the main tasks in this situation (Habib et al. 2016, p. 1). Implications arising are that researchers have studied more immediate responses than preparation and/or prevention events, and that preparation and prevention activities, which include a supply chain perspective, have been neglected by the relevant authorities due to the fact that there are few studies which analyse preventive actions to deal with such disasters. Supply chain perspective in a context of humanitarian operations means planning supply and demand issues considering volume and location of inventories in order to be available for an immediate response, alternative routes for transportation of goods and negotiation with suppliers to propose emergency plans (Holgun-Veras et al. 2012b). As such, we propose the following:

\({ GAP}_{2 }\) How are public and private sectors supply chains involved and organized to support the preparation and prevention of situations like natural and man-made disasters?

Fig. 2
figure 2

Results obtained for Category 2

Full size image

5.3 Methods

In this category, we identify the main research methods used in articles that address HLSCM. From our results presented in Table 2, most of the relevant studies are conceptual (26.44%), as shown in Fig. 3.

However, when we analyse each category in isolation, in category 3A articles dealing with the topic using qualitative methods are the majority (19.05%), whereas articles that approach the topic quantitatively (9.52%), or those that use mixed qualitative and quantitative methods together (3.17%), are very few. Consequently, it seems plausible to argue that there is a need for researchers in the HLSCM area to use more quantitative methods or mixed methodologies in their work. In order to do so, it is necessary that data on different aspects of disasters are collected and made available on an open access basis, such as The International Disaster Database. This is because the more data or information regarding calamitous events are available, the greater the chances of developing a strategy of prevention or assistance to victims. Additionally, opening up databases on natural and man-made disasters could enable us to develop simulations of potential impacts, and to forecast potential disasters in order to develop efficient action plans regarding them. of actions. From the points above, we propose:

\({ GAP}_{3 }\) Which barriers exist to make quantitative studies feasible in the field of HLSCM? How is big data being used in the context of HLSCM?

Fig. 3
figure 3

Results obtained for Category 3

Full size image

5.4 Type of disasters

In order to deal with situations of disaster in a preventive way or by helping its victims, it is necessary to be aware of the two main types of disasters: natural or man-made, and also the speed of these events, i.e., if occasions are of a slow start (slow-onset) or a sudden start (sudden-onset). In this sense, as shown in Fig. 4, not even half of the studies (43.16%)Footnote 1 consider such type and speed aspects, and when they do, they are mostly directed at natural disasters of a sudden occurrence which, in turn, are explained by the need to operate efficiently to assist victims. This finding appears logical given the fact that the majority of the articles analysed are theoretical, so the distinction between different kinds of disasters is not always discussed in them. However, different kinds of disasters and their respective pace would require specific resources and capabilities in the operations of humanitarian logistics. Therefore, a research gap emerging is:

\({ GAP}_{4 }\) Which resources and capabilities could be developed by organizations in order to deal with the different kinds, and pace of, disasters?

Fig. 4
figure 4

Results obtained for Category 4

Full size image

5.5 Phases of the disaster relief

In order to complement the finding of the Type of Disaster category, we identified how humanitarian logistics studies consider the Phases of Disaster Relief. Consequently, as shown in Fig. 5, most such works (41.38%) did not observe or discuss these phases of disaster which, in turn, could be easily understood, since most of the articles on this subject are conceptual, as already explained in Sect. 5.3. However, regarding the articles that do consider this phase element, such works concentrate on the perspective of how the difficulties of immediate response (5B—20.69%) can be overcome. This finding highlights that the main focus of such phase research is in how to deal with disasters rather than how to foresee or to be prepared for them. This result aligns to item 5.2, which may mean that either researchers have studied more emergency situations than preparation situations, or planning and preparation activities have been neglected by the relevant authorities. Therefore, it could be interesting to identify:

\({ GAP}_{5 }\) Which initiatives or plans of prevention to natural and man-made disasters are developed in countries devastated by these disasters? Which kind of approach has been adopted after facing such disasters?

Fig. 5
figure 5

Results obtained for Category 5

Full size image

5.6 Type of humanitarian organization

Category 6 identifies which organizations, national or global, are observed in the work of humanitarian logistics, since these organizations in various situations play a fundamental role in helping disaster victims. Based on results in Table 2 and Fig. 6, the relevant articles here mostly do not comment on such organizations (6E 65.52%) or, when they do, present brief comments on supranational aid agencies (6A—10.34%) and governmental organizations (6B—9.20%). Looking at these notes, we ask:

\({ GAP}_{6 }\) How can humanitarian organizations coordinate with each other in order to support the preparation/prevention, immediate response, and reconstruction phases of disaster relief?

Fig. 6
figure 6

Results obtained for Category 6

Full size image

5.7 Region of authorship and disasters

Categories 7 and 8 sought to geographically locate the main areas or regions of authors on the theme of HLSCM in their studies and, consequently, the main continents considered in such studies. Most authors are from the United States (18.39%) or from Europe, and specifically, from Finland (13.79%). An explanation why most European work originates in Finland stems from the significant works of Prof. Gyöngyi Kovács and Prof. Karen M. Spens from the Hanken School of Economics in Finland. Regarding region of disasters, the majority of articles (55.17%) do not allow this identification, since they are conceptual works, as indicated in Sect. 5.3. However, by observing the articles that explain this information, it was possible to identify that there is a great deal of interest from researchers in the Asian and American continents. One possible explanation for this finding lies in the fact that in recent years’ serious disasters have occurred in these localities.

5.8 Purposes of articles analysed

The articles selected were also analysed in terms of their objectives in order to synthesize the main streams of the HLSCM field, which are presented in Table 2 using content analysis, and a conceptual map was developed as shown in Fig. 7.

Fig. 7
figure 7

Conceptual map

Full size image

Five streams of research interest were identified in the field of HLSCM namely: logistical coordination, framework, traditional logistics and SCM versus HLSCM, performance measurement, and model, based on the number of topics found.

The logistics and SCM coordination stream is composed of understanding the lack of coordination between aid members, the necessity of developing relationships between the players involved in humanitarian operations and identification of challenges to promote coordination within humanitarian logistics activities. The framework suggests researchers investigating the topic of HLSCM through particular theoretical lenses, and consequently, proposing new future avenues of research interest. The traditional logistics and SCM vs HLSCM stream addresses which similarities exist between the activities and decisions of traditional ones, while the performance measurement stream discusses developing indicators in the HLSCM field. Lastly here, the model stream proposes mathematical models to plan routes of localisation of inventories, and to support decision making on resource allocation and recovery after disasters.

5.9 Main gaps and fields of interest

Based on our analysis, we identified six research gaps as well as the main fields of interest and trends for the HLSCM area. To summarize, the six research gaps identified during our literature review process are detailed below.

  • \({ GAP}_{1}\) Which lessons would be learnt from non-mature economies in order to foresee, and be prepared for natural disasters?

  • \({ GAP}_{2}\) How are public and private sectors supply chains involved and organized to support the preparation and prevention of situations like natural and man-made disasters?

  • \({ GAP}_{3}\) Which barriers exist to make quantitative studies feasible in the field? How is big data being used in the context?

  • \({ GAP}_{4}\) Which resources and capabilities could be developed by organizations in order to deal with the different kinds, and pace of, disasters?

  • \({ GAP}_{5}\) Which initiatives or plans of prevention to natural and man-made disasters are developed in countries devastated by these disasters? Which kind of approach has been adopted after facing such disasters?

  • \({ GAP}_{6}\) How can humanitarian organizations coordinate with each other in order to support the preparation/prevention, immediate response, and reconstruction phases of disaster relief?

The main streams of research interest in the field are: coordination, frameworks, traditional versus HLSCM, performance measurement, and model.

Combining the information from our results and research gaps and streams, it could be argued that that more practitioner-focussed research is needed in the field of HLSCM, that preparation and prevention should be addressed either by academics and/or relevant authorities, and that supply chain context needs analysing in order to discuss coordination between aid members. Additionally, understanding the resources and capabilities of the players and agents involved of humanitarian operations seems pivotal to comprehending our proposed research gaps above.

6 Conclusions

This article synthesizes the research literature on HLSCM in order to organize it under a conceptual map and integrate existing ideas to create new ways of thinking and understanding this theme, using articles identified in Scopus and Web of Science using the terms “Humanitarian Supply Chain” or “Humanitarian Logistics”. As a result of our initial search, 155 articles were refined using the filters “Article”, “Review”, “Article in Press”, “Source Type Journal” and “English Language”, which produced 87 articles available for full analysis and review.

The main results of our analysis are that the majority of our reviewed articles were theoretical, and as a consequence, few of them discussed issues related to localisation of disaster, type of disaster, phase of disaster relief, and type of humanitarian organization. Their focus was mainly on logistics.

This article contributes to the literature of the HLSCM field by proving a synthesis of this theme and highlighting new perspectives on how it has been addressed, along with potential development areas to further guide future research. The limitations of this article relate to the cognitive process of analysing the identified articles herein, and the filters selected to choose the articles reviewed, which we discussed.

Nonetheless, based on our findings we have proposed six new research gaps and developed an original conceptual map which charts five streams on how to further integrate the humanitarian logistics and supply chain management field, which both represent a new way to further understand this theme.

Further research is specifically needed to apply the concepts of HLSCM in different contexts. Here, wider geographical perspectives could empirically test the global validity of theories used in HLSCM research and understand context dependency in HLSCM. A requirement for further empirical and theoretical work exists regarding international humanitarian operations, as well as not-for-profit organizations.