A Decision Support Model for Barriers and Optimal Strategy Design in Sustainable Humanitarian Supply Chain Management

HSC involves the supply, distribution, logistics, and delivery activities during natural disasters or emergencies in the affected area; it is also known as the ‘relief supply chain’ (Seifert et al., 2018). The relief supply chain should be prepared, effective, and responsive to save human lives and distribute relief to the affected people. To ensure the smoothness of operations of HSC, different entities (e.g., suppliers, government bodies, and other stakeholders) need to communicate closely to deliver the activities quickly and efficiently (Gossler et al., 2020). Figure 1 illustrates the supply chain network for an HSC. The concept of SHSC derives from the integration of sustainability and HSC. The overarching philosophy for sustainable development is integrated organically into disaster risk reduction. It includes preventing and mitigating disasters, preparing for them, and distributing relief helps reduce suffering, save lives, and contribute to development, as outlined by the United Nations (2015). Similarly, ensuring the sustainability of HSC requires proper planning, preparation, assurance that specific resources are available, and a certain degree of independence, as Abbas et al. (2021) suggested. Furthermore, Cao et al. (2021) stated that disasters resulting in enormous casualties, property loss, and environmental disturbance harm sustainability. It is reported that 80% of operations are related to logistics in HSC (Li et al., 2019). Thus, to ensure the sustainability of HSC operations, investigating an SHSC requires urgent attention, which is also supported by Abbas et al. (2021), Boostani et al. (2021), Dubey and Gunasekaran (2016), and Karuppiah et al. (2021). However, many barriers impede SHSC operations, while attempts to prioritize and strategies to mitigate these barriers have not been explored extensively. The prioritization of SHSC challenges and strategies is essential because organizations can select appropriate mitigation strategies based on their capabilities.

In humanitarian logistics, the essence of balancing economic, environmental, and social aspects has been discussed in several studies (Abbas et al., 2021; Cao et al., 2018; Kunz & Gold, 2017). Chen et al. (2020) considered sustainable humanitarian operations as a mechanism to decrease vulnerability by addressing immediate needs and developing more resilient and long-term prospects at a minimum cost economically, socially, and environmentally. Cao et al. (2018) emphasized that the inequitable distribution and shortages of supplies contribute to social disturbance, negatively impacting social stability and sustainability from an operational viewpoint. Kunz and Gold (2017) shed light on quick recovery methods from crises while describing SHSC. Haavisto and Kovács (2014) posited that SHSCs should also focus on environmental factors along with economic and social sustainability. Considering these and other examples, the literature has conceptualized SHSC from multiple perspectives. Cao et al., (2017, 2018) defined an SHSC as the integration of sustainable development and HSC. Boostani et al. (2021) considered humanitarian relief supply chains should minimize total costs and maximize social welfare with the least possible environmental impact as their key objectives. Agility and adaptability are key concepts that deserve attention in the SHSC setting (Dubey & Gunasekaran, 2016). Drawing on the literature (e.g., Boostani et al., 2021; Cao et al., 2017, 2018), in this paper, SHSC is defined as designing HSC operations to deliver the desired emergency relief at the right time and in the correct quantity while optimizing the social, environmental, and economic aspects.

Rapid and unpredictable occurrences of demand and the prompt delivery of essential relief supplies are exceptional complex features of HSC (Beamon & Balcik, 2008). Oloruntoba (2005) discussed the challenges various humanitarian organizations face during international relief and reconstruction efforts. Such challenges include damage assessments, needs-based assessments, the devastation’s scale, funding and donations, logistics, and coordination problems. Furthermore, a significant challenge to implementing humanitarian supply chain management (HSCM) is related to information sharing (Dubey & Gunasekaran, 2016). Lack of proper communication and a manual supply chain approach make the HSC process challenging (Kunz & Gold, 2017). Moreover, the lack of coordination hinders HSCM, resulting in delays in delivering relief materials at the last minute and ineffective response to disaster relief activities (Sahay et al., 2016), which also impedes sustainable humanitarian supply chain management (SHSCM). In addition, HSCM often encounters challenges in controlling its inventory due to the rapid fluctuations in demand (Sentia et al., 2023). Further, inefficient distribution network design—including customer dissatisfaction, longer lead/delivery time, and over or under-utilization of distribution centers—negatively affects supply chain performance (Kunz & Gold, 2017; Sassanelli & Terzi, 2022).

Challenges relating to social and environmental issues substantially impact effective and efficient HSCM processes. Cao et al. (2017) identified carbon emissions, emergency costs, and weighted completion times as significant issues in HSC. Dubey and Gunasekaran (2016) and Karl and Karl (2022) identified some barriers, such as a lack of training and lack of health and safety awareness, as considerable barriers to social sustainability in HSC operations. However, those studies did not prioritize the challenges to guide the decision-makers about the intensity of the barriers and their impact on social, economic, and environmental welfare. Hong and Guo (2019) argue that a lack of environmental awareness hampers the smooth operation of HSC activities. Moreover, failing to engage the local people during relief distribution remains a substantial challenge in humanitarian operations (Abbas et al., 2021). To ensure sustainability, all the stakeholders must play their roles effectively, especially the local people of disaster-ridden areas, whose role remains crucial. However, the literature on humanitarian operations management has been deafeningly silent on optimizing challenges (Salem et al., 2019). As relief products are distributed in remote locations, it is challenging to ensure proper tracking (Fernandez & Suthikarnnarunai, 2017). Transparency in SHSC activities requires agreement from various parties involving government, private, and local stakeholders; otherwise, the lack of transparency impedes the SHSCM process (Dubey & Gunasekaran, 2016). Karuppiah et al. (2021) also identified that the lack of control and monitoring is essential barriers to SHSC. Furthermore, a lack of sense, flexibility, responsiveness, integration, and partnership significantly affects a quick response to the disaster, which impedes the SHSCM process. Moreover, Dubey and Gunasekaran (2016) identified the lack of adaptability to the culture, the lack of collaboration, and the limitations of supply chain partners as substantial inhibitors to the SHSCM process. Despite the numerous challenges to SHSC, the literature does not provide an applied decision support model to confront the barriers to managing operations more seamlessly. Table 1 summarizes the challenges to SHSC.

For efficient and effective HSCM implementation, overcoming the challenges that inhibit successful humanitarian disaster response operations is necessary (Agarwal et al., 2019). Dubey et al. (2014) argue that mitigating HSC challenges is crucial to quickly regaining and restoring normalcy in disaster-prone regions. Considering the severity of the problem related to SHSC, priority-based, widely accepted, and appropriate strategies is imperative for improving humanitarian activities. Practitioners and academicians suggest several strategies instead of any single approach to overcoming HSC barriers (Kabra & Ramesh, 2015b), which is also true for SHSC.

Literature indicates specific strategies for overcoming challenges but lacks strategies for addressing them at different stages of the SHSC. For instance, multiple suppliers and flexible sourcing strategies are critical areas of the procurement process to respond quickly to challenges (Chen et al., 2020). Several researchers have mentioned that flexible transportation strategies, such as an operational mix of transportation and transport availability, effectively distribute relief in the SHSC (Jermsittiparsert & Kampoomprasert, 2019). Similarly, maintaining strategic emergency stocks and encouraging collaboration (cooperation) among supply chain members helps to increase supply chain flexibility, resilience, and agility to mitigate SHSC challenges (Salvadó et al., 2017). Furthermore, to mitigate inventory and transportation-related challenges, several strategies, such as coordination among humanitarian organizations, assisted in increasing efficiencies within disaster relief operations (Dubey & Gunasekaran, 2016). Disaster relief operations are more effective when humanitarian aid organizations coordinate more effectively to deliver relief materials on time (Singh et al., 2018). The prepositioning of relief items efficiently fosters a productive and resilient HSCM (Li et al., 2019). Ineffective inventory management (Li et al., 2019), the lack of information sharing (Dubey & Gunasekaran, 2016), and the lack of resource planning and assurance (Abbas et al., 2021) can be addressed by adopting several strategies such as managing inventory policies, information integration and sharing, and planning emergency activities (e.g., shelter, rehabilitation, and reconstruction) (Bag et al., 2022; Li et al., 2019).

The literature also asserts the importance of outsourcing logistics services to minimize improper transportation planning and delays in distribution (Laguna-Salvadó et al., 2019). Resource sharing has been shown to have a major impact on operational outcomes (Kovács & Tatham, 2009), and decentralized warehouses (Chen et al., 2020) in relief operations and improvements to the relief chain’s effectiveness (Patil et al., 2021) could address relief distribution problems. In addition, Kunz and Gold (2017) focus on contingency planning to minimize the HSC crisis. To manage the barriers of SHSC, a learning and innovative culture, particularly through adoption, is essential (Dubey & Gunasekaran, 2016). A summary of strategies for mitigating barriers in SHSC design is presented in Table 2. Though numerous studies focus on strategies to mitigate the barriers of SHSC, an overarching framework to determine optimal strategies to mitigate different barriers is not present.

In phase 1, we applied a qualitative approach to identify barriers to SHSCM in the literature review and the strategies that would overcome those barriers. Then, we verified those barriers and strategies with the findings from the interviews. Using a semi-structured interview protocol, context-specific data were collected from nine decision-makers involved in sourcing, storing, and distributing humanitarian aid at different operational levels (e.g., local, zonal, and national levels). The participants were asked to describe the barriers to SHSCM (WHATs) and the strategies (HOWs) to overcome the barriers. To collect data, we adhered to the principle of ‘[continuation] until ideas are saturated’. As we noticed no new barriers or strategies emerged from the last two interviews, we stopped interviewing after collecting data from the ninth respondent. The duration of the interview was approximately 30–45 min. Table 3 presents the demographic profile of the respondents.

Interviews were noted, recorded, and transcribed. The content analysis technique was applied to the scripts to analyze the data collected from the interviews. A comparison was conducted between the barriers and strategies outlined in the interviews and those found in the literature, and changes were made as necessary to ensure the content validity of the findings.

In phase 2, we deployed a quantitative approach. Along this line, the importance weights of WHATs were determined (\({W}_{i}\) in Fig. 2) using the analytical hierarchy process (AHP) (Saaty & Vargas, 1980). Then, we determined the significance of the strategies using QFD, a tool widely used to translate organizational problems into efficient and effective strategy design (Akao, 1990). In this phase, to determine the importance weights of barriers, we asked the interviewees to compare each SHSC management barrier (WHATs) using a 1–9 point scale (Saaty & Vargas, 1980). The weights of the WHATs thus derived were then used as inputs for QFD (Akao, 1990) to determine the importance of the strategies (HOWs) to mitigate the barriers (see A.I. in Fig. 2). The systematic steps of QFD are illustrated below.

Stage 1: SHSC barriers were identified.

Stage 2: Relative. importance of barriers (\(={ W}_{i}\)) were calculated.

Stage 3: Strategies (HOWs = \({{\text{DR}}}_{j}\)) to mitigate the barriers were determined.

Stage 4: Relationships between barriers and strategies (WHAT-HOW relationship) were calculated.

Stage 5: Determining the ranking of strategies based on the barriers and strategies’. relationship. score, weights (A.I. and R.I.) were assigned to the HOWs.

Stage 6: Determining the time and cost savings from the simultaneous implementation of the strategies, relationships between the HOWs were determined.

To. determine. the. relationship between WHAT and HOW, the interviewees were asked to determine the impact of various mitigation strategies (HOWs) to mitigate barriers (WHATs) using the scale. ‘0 = no’, ‘1 = little’, ‘moderate = 3’, or ‘9 = strong’ (Park & Kim, 1998). The weights represent the relative importance of the relationship. As a result, the following equation determined the importance of each strategy:where \({{\text{AI}}}_{j}\) absolute importance of design requirement (DR_j)/strategies, which are also referred to as strategies S_j; \({W}_{i}\) weight of the ith barriers which are derived from AHP; \({R}_{ij}\) relationship value between the ith barrier and jth strategy (9, 3, 1, or 0); \(n\) number of strategies; \(m\) number of barriers.

Similarly, we can determine the relative importance of strategy. j using the following equation:where \({{\text{AI}}}_{j}\) absolute importance of S_j strategy. \(\sum_{j=1}^{n}{{\text{AI}}}_{j}\) summation of the absolute importance of all strategies.

In phase 3, we developed a DS model to identify the.most.efficient.strategies.to. mitigate the SHSC barriers by using the optimization technique. Our interviewees were asked to discuss the cost and time necessary to implement each. strategy, and the budget required to execute the most important strategies. Additionally, the interviewees were asked whether the simultaneous implementation of strategies would result in cost or time savings.

The optimal strategies were determined by optimizing the absolute importance (AI) values of the strategies within budget and time constraints. For optimization, we utilized nonlinear binary integer programming. The roof matrix shown in Fig. 2 was used to identify the relationships among the strategies to assess the cost and time savings associated with simultaneous implementation. Different symbols indicate the degree of interrelationship between strategies in the roof matrix.

Below is a formula for the optimization problem:

Based on the decision-makers’ opinions relating to the challenges that impact SHSC management in Bangladesh, we identified a study-specific list of barriers and revised the list identified from the literature review (presented in Table 1) to ensure the context and content validity of our findings. From the interview, we have identified several new barriers, such as the lack of social and environmental awareness, the prevalence of corruption and political interference, the lack of contingency planning, the lack of pre-disaster fund accumulation, the lack of logistical support, and the lack of technological support. The final list of barriers, as obtained from the literature review and interview process, is presented in Table 4.

In this phase, we also identified a list of strategies to mitigate the barriers and revised the list of barriers identified in the literature review (presented in Table 2). From the interviews, we have identified several new strategies, including an increased awareness of health, sanitation, safety, and waste reduction; helping livelihood recovery and alternative income-generating activities; and ensuring the actual needs-based and efficient distribution of relief to the right people. The final list of strategies obtained from the literature review and interviews is presented in Table 5.

In phase 2, we determined the weights of the barriers to SHSC (WHATs) and the importance of strategies (HOWs) to mitigate the barriers. The weights of the barriers to SHSC (WHATs) are shown as Wi in column two of Fig. 3. Then, using the systematic steps of QFD, the absolute importance (AI) value of strategies was derived from the WHAT-HOW relationship matrix (see Fig. 3) to assign weights to strategies and determine the importance of each. We found that strategy S4 (building relationships with network members) had the highest AI value, while strategy S11 (supporting post-disaster livelihood) had the lowest AI value. Additionally, we identified the interrelationships among the strategies/HOWs in this phase, as shown in the roof of Fig. 3. For example, S1 [improving information systems (digitalization)] and S11 (supporting post-disaster livelihood) demonstrated a very strong relationship. Thus, if these strategies are implemented, substantial cost and time savings can be gained. For brevity, we have presented only the relationship strengths in the roof matrix of Fig. 3, while cost and time savings from interrelated strategies are presented in Tables 7 and 8, respectively.

In phase 3, we determined the optimal strategies (HOWs) to mitigate the barriers to SHSC. The objective function refers to maximizing the AI score of each strategy. Through the solution of the optimization problem, we attempted to determine an optimal strategy based on the collected relevant data. Following is the formulation of the optimization problem:where \(x_{j} \ge 0\) and x ∈ {0, 1}, AI absolute importance scores of the strategies S₁… Sj,\({c}_{j}{x}_{j}\) the cost of implementing strategies, \({{\text{cs}}}_{ij}\) cost savings from simultaneous. implementation of strategies, and \({{\text{ts}}}_{ij}\) time savings from simultaneous implementation of strategies. The values of AI_j, S_ij, and C_ij are obtained from Fig. 3. In addition, it should be noted that the required cost data C_ij and savings data S_ij are obtained through interviews with the QFD team.

We solved the model using the generalized reduced gradient (GRG) nonlinear solving technique to determine the optimal strategies to mitigate barriers. We used the AI data obtained from the QFD analysis to represent a sample case. The data on implementation time and cost were collected from expert opinions. Table 6 presents the input data.

There will be some cost and time savings if more than one strategy is implemented. The cost and time savings data are presented in Tables 7 and 8, respectively.

Using the input data, we solved the model to determine the best strategies subject to budgetary and time constraints. In the sample experiment, we considered the maximum implementation budget and time to be US$40 million and 45 months, respectively. Table 9 presents the optimal strategies. We have observed that it would be optimal to implement St2–St10, St12–St14 to gain the maximum benefits of cost and time savings. In this case, the cost and time savings are US$10 million and 21.5 months, respectively.

Further, we have conducted a scenario-based analysis under different implementation budgetary and time constraints. The scenarios are presented in Table 10.

Under different scenarios, we obtained different sets of optimal strategies, as presented in Table 11.

From the scenario analysis presented in Table 11, we observed that the number of strategies in optimal decision-making increased concomitant to an increase in the implementation of budget and time, thus increasing the value of total absolute importance (objective function).

The results from the qualitative study in Phase 1 confirmed all the barriers and strategies outlined in the literature (see Tables 4 and 5). The qualitative study also explored several new barriers (e.g., lack of social and environmental awareness) and strategies (e.g., reducing waste; helping livelihood recovery and alternative income-generating activities, and ensuring the actual needs-based and efficient distribution of relief to the right people) (for details see Tables 4 and 5).

In Phase 2, we found that the lack of contingency planning, the prevalence of corruption and political interference, and lack of social and environmental awareness are the most important barriers, while logistics outsourcing, SC performance management, and SC flexibility are the most important strategies. In this phase, Fig. 3 also illustrates the association between the strategies depicted in the roof matrix. For example, S1 [improving information systems (digitalization)] and S11 (supporting post-disaster livelihood) were found to have a very strong relationship. This approach (finding strong correlations among strategies and joint implementation) helps to reduce costs and time substantially. Response and recovery time are crucial elements to a humanitarian crisis, and time-saving is a very important factor along with cost savings in humanitarian SC operations.

In Phase 3, the results of the optimization model showed that all except S1 [improving information systems (digitalization)], S11 (supporting post-disaster livelihood), and S12 (SC flexibility) could be implemented within the budget (US$40 million) and time (45 months) to achieve the highest benefits of cost (US$7.9 million) and time (19 months) savings. Moreover, mitigation strategies have the flexibility to tackle different crisis situations, such as the absence of logistical support that requires quick resolution and a higher budget, which may be mitigated by establishing relationships with network members, outsourcing logistics, or developing contingency plans, which is in line with Baharmand et al. (2019). Conversely, improving information systems could mitigate the lack of technological support, communication, and information sharing in an SC network and concur with Jermsittiparsert and Kampoomprasert (2019). In other words, a mitigation strategy can also be applied to several crises, and several strategies can address a problem. In this way, we have performed a scenario-based analysis under various implementations of budgetary and time constraints to determine the most suitable optimal strategies from different sets (see Tables 10 and 11). The analysis addresses dynamic changes in strategies based on changes in situations and offers flexibility to the managers to fit with the context. It also highlights the necessity to investigate key barriers within SHSC and their priorities to enable decision-makers to design effective strategies to achieve the desired outcomes.

Our study provides several theoretical implications. First, using a systematic approach, this study has identified the major barriers to HSC within an emerging economy while designing and implementing a sustainable humanitarian supply chain. It has also systematically prioritized the strategies needed to overcome those barriers. From the theoretical perspective, the study has developed a systematic decision support framework to identify SC barriers and develop strategies using a semi-structured interview, an AHP, a QFD, and a nonlinear binary integer programming technique, which can be explored, discussed, and expanded for research purposes. According to the study, the most important barriers to achieving SHSC systems and practices include the lack of contingency planning, corruption, political interference, and lack of awareness of social and environmental issues. The study revealed that logistics outsourcing, SC performance management, and SC flexibility are critical to increasing SC capability for efficient SHSC.

Third, our research generates and examines various SC barriers and their corresponding strategies that contribute to optimal strategies for the supply chain in disaster management. In general, in view of supply chain research, and in particular, supply chain disaster management, our method is unique and addresses the void that there is a lack of research in the literature on humanitarian supply chain management to manage extreme disruptions (Thompson and Anderson, 2021). The results of our study advance the literature on supply chain in disaster management related to the SHSC by empirically investigating how to identify and prioritize the barriers, and determine the optimal strategies for mitigating them in order to achieve the desired cost and time performance, which is new in SHSC literature.

Finally, using a systematic approach, our study has developed an innovative DS model to expound the SHSC nexus, which opens up a new discourse in the supply chain management domain. As a result of this move, we are responding to the call for research to determine optimal strategies for the SHSC to be efficient under the DS model (Dubey et al., 2022; Zanon et al., 2021). Conceptualizing SHSC, along with the application of the QFD-integrated optimization technique, could produce an appropriate optimal strategy; there is limited empirical attention given to this subject in supply chain management to date (e.g., Kaivo-oja et al., 2014).

Our study also offers several managerial implications. First, the lack of DS modeling among many countries raises questions about the planning and achievement of sustainable performance goals in humanitarian operations (i.e., food, housing, rescue attempts, health, and medical) (Laguna-Salvadó et al., 2019). Identifying and prioritizing barriers and selecting an optimal strategy via our proposed DS model would enable managers to sustainably improve the planning and implementation of HSC processes. Thus, decision-makers can use this information to focus on implementing specific strategies that address critical challenges, leading to more effective and targeted humanitarian supply chain management.

Second, this model can be used by managers from humanitarian organizations to identify HSC barriers and prioritize those barriers in situations where resources are scarce—specifically, organizations that are struggling with cost and time management. Our DS model suggests how to ensure an efficient and sustainable design for humanitarian operations, considering time and cost savings.

Finally, concerning HSC, most stakeholders (i.e., donors) are interested in whether the operations they support in terms of cost and time have consequential and positive impacts on the affected people (Medina-Borja and Triantis, 2014). Understanding the donor behavior in humanitarian aid operations and educating, assuring, and preparing donors in advance is vital. This is because donors need to know how their contributions will be used effectively and efficiently (Ülkü et al., 2015). Therefore, we argue that our DS model, which aims to identify and prioritize barriers and determine an optimal strategy to implement SHSC processes, also provides valuable information that can be used to attract and motivate potential donors. This would enable designers of HSC management to develop a sustainable approach as a critical success factor.