1 Introduction
The COVID-19 pandemic has had devastating and far-reaching global impacts, affecting public health as well as normalcy and productivity (
Ning et al., 2023). In the early stages of the pandemic, governments prioritized containment of the virus transmission. However, as the virus continued to spread and the contagion persisted, several nations, such as Australia and Malaysia, implemented stricter measures, including city lockdowns, to curb its spread (
Nadig and Krishna, 2020). These government-imposed city lockdown restrictions effectively limited the spread of the virus by restricting movement of individuals and partially curbing production activities of companies. However, these measures also had significant effects on industries and supply chains, particularly impacting the food supply chain and disrupting daily lives of people (
Singh et al., 2021;
Chin, 2020). When facing these challenges, government interventions in the food supply chain have proven to be effective in ensuring food safety and public health (
Cui et al., 2023). Government-led initiatives in the food supply chain greatly contribute to mitigating disruptions in people’s daily lives during pandemic control efforts (
Coluccia et al., 2021). Specifically, government involvement in the food supply chain played a crucial role during the pandemic, effectively replacing disrupted food supply channels and ensuring residents’ access to food. During health emergencies like the COVID-19 pandemic, government intervention can significantly contribute to meeting the dietary needs of the population. Referring to the definition of relief supply chain during emergencies (
Yao et al., 2023), a government-led food supply chain can be defined as one where the government coordinates and organizes the food supply chain, including suppliers, warehouses, and transportation, in a unified manner to replace disrupted food supply chains and ensure sufficient food supply for residents. However, the effectiveness of this supply chain may vary depending on the extent of government intervention and restriction strategies.
During periods of reduced restrictions in countries such as the United States, Singapore, and Indonesia, the issue of food shortages and government-led food supply problems remains prevalent. For instance, the food supply chain in the United States has been significantly affected by the ongoing pandemic. To address these challenges, the Trump administration launched the Farmers to Families Food Box initiative, which aimed to purchase fresh food directly from farmers and distribute it to those in need. Similarly, Singapore heavily relies on food imports, and as a result, its food supply chain has faced continuous disruptions due to the pandemic. Despite the active involvement of government authorities in securing the food supply by sourcing from various countries and incentivizing designated production, food supply challenges persist.
In Indonesia, there were initial disruptions in the supply of imported food, leading to restrictions on staple food retailers to manage demand surges and maintain supply chain stability. Other countries, like Australia, which initially implemented strict measures, encountered significant labor shortages in their fruit and vegetable supply chains due to COVID-19, resulting in disruptions in the market’s food supply. In response to these labor and supply chain shortages, the Australian government imported workers from the Pacific region. However, despite these efforts, a substantial amount of fruits and vegetables were still damaged due to delayed harvesting.
At the beginning of the pandemic, Malaysia implemented a 14-day lockdown, triggering panic buying and food shortages among its residents. In these scenarios, despite continuous efforts to manage the food supply chain during the COVID-19 pandemic, widespread food scarcity still persisted. Although the government-led food supply chain partly replaced the conventional system to ensure residents’ access to food, it suffered from insufficient supply capacity due to labor shortages, limited production capacity, and a surge in public demand caused by strict restrictions. Therefore, this study defines the issue as a shortage in the government-led food supply chain due to inadequate supply capacity during the pandemic. For the sake of clarity and conciseness, this concept will be referred to herein as “food shortages.”
In response to the issue of food shortages during the pandemic, extensive research has been conducted to investigate the relationship between disruptions and food shortages (
Falkendal et al., 2021;
McKay et al., 2021;
Ali et al., 2021a;
Alabi and Ngwenyama, 2023;
Burgos and Ivanov, 2021). Some studies have also explored recovery strategies to alleviate food shortages and disruptions (
Moosavi and Hosseini, 2021;
Rahman et al., 2021;
Rahman et al., 2024). However, the existing literature primarily focuses on non-government-led food supply chains during the pandemic, with limited attention given to the government-led supply chain and its challenges, such as food shortages, disruptions, and recovery strategies. Additionally, existing studies have overlooked the dynamic nature of the evolving pandemic and lack detailed scenario simulations and policy deductions specifically relating to the government-led food supply chain. It is essential to note that the COVID-19 pandemic has highlighted the critical role of government intervention in the food supply chain. Unfortunately, shortages or disruptions in the government-led food supply chain are common occurrences, regardless of the adopted pandemic control strategy, and the underlying mechanisms remain unclear, particularly in the context of the pandemic. Furthermore, while several recovery strategies implemented by governments have demonstrated mixed effects, there is limited understanding of the mechanisms behind these strategies and how to optimize them further. Addressing these challenges requires a complex systems perspective to analyze and solve problems related to the government-led food supply chain during the COVID-19 pandemic. Therefore, studying shortages, disruptions, and recovery strategies in this context bears significant social and economic value. Based on these motivations, the primary objective of this study is to address the following three key questions:
(1) Which links (time and space) in the government-led food supply chain experience shortages during the COVID-19 pandemic? What are the causes of these shortages?
(2) What are the effects of different recovery strategies implemented by the government on alleviating food shortages? Do these effects differ as the pandemic progresses through different phases?
(3) How resilient and stable are the recovery strategies in the event of severe disruptions in the government-led food supply chain?
To address the aforementioned questions, this study conducted a simulation of the government-led food supply chain during the COVID-19 pandemic. The simulation was built using the AnyLogistix software and involved a three-level supply chain model. The study took into account the phased changes in demand that occurred as the pandemic progressed. The aim was to explore the recovery strategies of the government-led food supply chain.
The findings of the study revealed that food shortages were primarily caused by inadequate transportation capacity, the uneven distribution density of district warehouses, and the mismatch between production capacity and demand growth. The study suggests that enhancing transportation capacity and establishing new district warehouses can effectively alleviate food shortages while maintaining stability and robustness in the supply chain. By increasing production capacity, the demand for food supplies during the growth phase can be met. The study also discovered that the establishment of municipal emergency warehouses had a surprisingly positive effect in overcoming potential disruptions. However, it is important to note that food assistance from other provinces and increasing the frequency of inventory checks may have unintended negative consequences. Additionally, food reallocation between district warehouses was found to have a detrimental impact on the supply chain.
This study contributes to the field of food supply chain management in three significant ways. First, it adds to the existing literature by providing insights into the shortages and recovery strategies of government-led food supply chains during major public health emergencies through simulation studies. By examining the underlying patterns of food shortage formation and evolution, as well as developing recovery strategies tailored to different phases of an emergency, this study contributes valuable conclusions and management insights. Secondly, this study provides innovative simulation analysis and discussion on two specific recovery strategies: food assistance from other provinces and food reallocation between district warehouses. The stability and robustness of these strategies were validated through simulating supply chain disruptions. Finally, the findings of this study offer a scientific foundation for government decision-makers to enhance their understanding of the operational mechanisms of government-led food supply chains, as well as the effectiveness of recovery strategies. This knowledge will enable decision-makers to effectively address shortages and disruptions in the food supply chain during major public health emergencies in the future.
The remaining sections of the paper are organized as follows: Section 2 provides a literature review. Section 3 presents the case background, describes the simulation model, and conducts experimental analysis. Section 4 outlines the design of seven recovery strategies and analyzes the simulation results. In Section 5, two disruption scenarios are extended to simulate and further validate the stability and robustness of the relevant recovery strategies. Section 6 discusses the results of the study and offers corresponding management insights. Finally, Section 7 summarizes the conclusions.
2 Literature review
2.1 Food shortages and supply chain disruptions during the pandemic
Several studies have examined food shortages or supply chain disruptions during the COVID-19 pandemic, and numerous recent reports have highlighted the severity of these shortages (
Bukari et al., 2022;
Bidisha et al., 2021). Falkendal et al. (
2021) have noted that trade restrictions and precautionary purchases can lead to severe food shortages. In addition, transportation disruptions, increased demand, panic buying, and other factors can affect food shortages (
McKay et al., 2021;
Ali et al., 2021a;
Alabi and Ngwenyama, 2023). Among these factors, panic food stockpiling during the pandemic has attracted widespread attention (
Wang and Gao, 2021;
Ritzel et al., 2022;
Ben Hassen et al. 2021;
Coleman et al. 2022) due to its significant role in exacerbating supply chain challenges. Min et al. (
2020) have reported that pandemic control measures have varying impacts on the food supply chain, with negative effects related to food categories, procurement channels, and patient numbers. Furthermore, Mahajan and Tomar (
2021) have found that remote food supply chains are particularly affected by sanitary emergencies. At the community level, disruptions in transportation, distribution, and delivery chains severely affect food supply (
Farcas et al., 2020). At the enterprise level, companies face both short-term and long-term impacts, such as working capital shortages and reduced returns (
Chowdhury et al., 2022). Furthermore, Narayanan and Saha (
2021) have found that retailers in the food supply chain face challenges stemming from transportation and labor shortages. Similarly, Hobbs (
2020) have highlighted the potential for labor shortages to significantly impact labor-intensive industries. In contrast, Liang et al. (
2021) have used a computable general equilibrium model to study the impact of labor shortages on food safety, concluding that the impact of labor shortages on the Chinese food supply during the COVID-19 pandemic was relatively limited.
2.2 Recovery strategies for supply chain resilience during the pandemic
Several studies have been conducted to investigate the effectiveness of recovery strategies in the context of the COVID-19 pandemic. Benker (
2021) has observed that while stockpiling behavior can lead to shortages in the food supply chain, it can also serve as an effective household recovery mechanism through modest additional purchases. However, Bender et al. (
2022) have cautioned that while increasing storage capacity can enhance household resilience, it may also inadvertently trigger negative effects on the food supply chain system. Sid et al. (
2021) have proposed recovery strategies for disruptions in the agri-food supply chain; however, their study lacked comprehensive evaluation indicators such as time and cost (
Ali et al., 2021b). To effectively address the impact of the pandemic, Gholami-Zanjani et al. (
2021) have put forth a resilient food supply chain model that accounts for demand uncertainty and pandemic disruptions. Chitrakar et al. (
2021) have demonstrated that novel food processing technologies have played a significant role in safeguarding the normal functioning of the food supply chain during the COVID-19 pandemic. Meanwhile, Shen and Sun (
2023) have revealed that an integrated supply chain structure and comprehensive intelligent platforms are conducive to alleviating abnormal demand and logistics disruptions caused by the pandemic. Bag et al. (
2023) have demonstrated that big data analytics can be utilized as a recovery strategy for manufacturing companies to restore and enhance supply chain resilience. Ekren et al. (
2023) have shown that additive manufacturing contributes to the resilience of the supply chain, while Ge et al. (
2022) have emphasized that strategic facility location can promote supply chain efficiency and improve its resilience accordingly. Furthermore, through an examination of a survey involving 698 respondents, Kent et al. (
2022) have discovered that consumer-driven strategies can help improve the resilience and sustainability of food systems. Liao et al. (
2022) have employed game theory to study government-led recovery strategies, affirming the necessity of anti-epidemic measures and subsidies, and providing valuable insights for government decision-making.
2.3 Simulations of food supply chains during the pandemic
The literature reviewed above examines various methodologies used to investigate food supply chains during the COVID-19 pandemic, with a primary focus on non-simulation approaches. However, these methods often fail to replicate real-world scenarios adequately and lack a diverse range of evaluation indicators, thereby limiting their ability to provide nuanced guidance. In contrast, simulation methods, which integrate multiple evaluation indicators and realistic scenarios, have gained widespread acceptance in recent years within supply chain research (
Ivanov, 2020;
Ivanov, 2019;
Zheng et al., 2022;
Ding et al., 2022;
Achmad et al., 2021;
Rahman et al., 2021). These methods offer realistic responses and effective problem assessment capabilities. They are particularly advantageous in studying food supply chains, enabling the analysis of disruptions and recovery strategies within complex systems (
Tundys and Wisniewski, 2020;
Galal and El-Kilany, 2016;
Utomo et al., 2018). For instance, Huang et al. (
2021) utilized simulations to predict disruptions in the lobster supply chain during the pandemic, foreseeing product backlogs and shortened delivery lead times. Wang et al. (
2022) demonstrated the potential of blockchain technology to enhance the qualification rate of agricultural products and distribution efficiency. Rahman et al. (
2024) showed that increasing order frequency to multiple suppliers and collaborating with third-party transporters can help alleviate the impact of panic buying. In terms of policy studies, Baležentis et al. (
2021) employed Monte Carlo simulations to assess agricultural policy measures, revealing the effectiveness of measures aimed at reducing financial burdens. Zhu and Krikke (
2020) utilized system dynamics simulations to identify optimal policy measures for managing food supply chains during the pandemic. Moreover, Burgos and Ivanov (
2021) employed simulation techniques to simulate the impact of various disruptions on a real food supply chain problem in Germany, providing recommendations for managing disruptions. However, simulation studies have their limitations. For example, Singh et al. (
2021) focused solely on integrating warehouses into the public distribution system without considering different supply chain levels. Despite these limitations, simulation methods still offer valuable insights and solutions for enhancing the resilience and efficiency of food supply chains. Furthermore, in other types of supply chains, Lohmer et al. (
2020) highlighted the advantages of blockchain technology in improving supply chain resilience through agent-based simulation. Similarly, Katsoras and Georgiadis (
2022) indicated that maintaining basic inventories is the optimal strategy for ensuring the stability of closed-loop supply chains during a disaster.
2.4 Summary
In response to the issue of food shortages during the pandemic, numerous studies have examined the relationship between disruptions and food shortages. Some studies have also explored recovery strategies to mitigate these shortages and disruptions. However, most of these studies have predominantly focused on non-government-led food supply chains during the pandemic, with very few investigating government-led supply chains and their associated challenges such as food shortages, disruptions, and recovery strategies. Additionally, while simulation is a comprehensive method capable of incorporating multiple evaluation indicators and realistic scenarios for effective problem assessment, existing studies have largely overlooked the dynamic stages of the pandemic and lack detailed scenario simulations and policy deductions specifically related to government-led food supply chains. In response to the practical needs and deficiencies of current research, our study employs simulation modeling and recovery strategy analysis of the government-led food supply chain during the COVID-19 pandemic.
3 Modeling and simulation
3.1 Simulation modeling of Scenario 0
In contrast to the lack of discussion on real scenarios in previous simulation studies of supply chains (
Mustafee et al. 2021), our study conducted simulation modeling based on a government-led food supply chain during the COVID-19 pandemic in a virtual city. We divided the study into four phases: Phase I: Lasts 19 days, during which a few new COVID-19 cases were discovered. Phase II: The city-wide silent management is implemented for 28 days. Phase III: Spanning 19 days, during which the express industry starts to resume work and production. Phase IV: Lasting 16 days, during which business and market activities resume, and normal life and production is fully restored at the end of Phase IV. Food demand is a critical factor that influences the functioning of the government-led food supply chain and varies depending on the stages of the pandemic. Building on the research conducted by Tsao (
2016), this study assumes that properly preserved food does not deteriorate instantly. Initially, our study establishes a model for the average daily food demand of residents during non-pandemic conditions. We assumed that there are 24 million people in the city. Adhering to the dietary guidelines, assuming a daily consumption of 300 g of cereals, 300 g of vegetables, 350 g of fruits, 200 g of meat, 300 g of milk, 35 g of soybeans and nuts, and 35 g of oil and salt per person, we have estimated the daily food demand per person to be 1.52 kg. For the sake of simplicity in our modeling, we have not differentiated between men, women, and children, resulting in an estimated total daily food demand for the entire city of around 36500 tons. Additionally, assuming there are 6000 communities in the city, the average daily food demand for each community is approximately 6.1 tons, representing the normal demand.
Taking into account the impact of pandemic control measures on food production and transportation, the daily purchase, distribution, and consumption of food have been significantly affected, leading to shortages. In response to this, the government has intervened by providing supplementary food, which we refer to as government-led demand. Therefore, based on the phased development of the pandemic, this study defines government-led demand as situations in which the government supplements a certain percentage of the normal demand (75%, 125%, 50%, and 25% respectively). The four mentioned proportion parameters represent the four phases of the emergency response. In Phase I, when the government implements lockdown measures, food supply channels from other provinces are temporarily interrupted, and the government must provide this missing quantity to residents. According to relevant statistics, we assumed that 85% of grain, 75% of meat and vegetables, and 65% of eggs and milk are supplied by other provinces. Therefore, in Phase I, this study takes the average of the above three values as the demand level (i.e., 75%).
In Phase II, as residents run out of food stocks and panic intensifies further, government-led demand would exceed that of the normal demand. For instance, the government provides approximately 800 tons of vegetables and meat daily to a population of about 1.25 million. Based on the theoretical food demand per person (200 g of meat and 300 g of vegetables per person), the food demand is estimated at 625 tons. Therefore, government-led demand is estimated to be 125% of normal demand.
In Phase III, as the express delivery industry resumes operations, residents access food through community group buying and other channels. For example, at the beginning of Phase III, more than 70% of the vegetable industry resumed operations and their production levels exceeded 70% of its normal output. The multiplication of these two ratios equals approximately 50%. Therefore, government-led demand is assumed to be 50% of the normal demand during this period.
In Phase IV, as businesses and markets begin to resume their normal activities, the residents are able to access increasingly greater amounts of food through various channels. For instance, in the agricultural sector represented by vegetable production, the production resumption rate at the start of Phase IV exceeded 90%, with enterprises approaching an 80% resumption rate. The multiplication of these two ratios amounts to nearly 75%. Hence, government-led demand is estimated to be 25% of the normal demand during this period. It is also worth noting that the demand ratio in this study was rounded to multiples of 25% to ensure that the data were proportional and facilitate comparative analyses.
Based on the characteristics of government-led food supply chain, a three-level food supply chain was established, comprising designated suppliers, district-level warehouses, and communities. The order and transportation flow of this supply chain are depicted in Fig.1. Orders from the communities are directed to the closest district warehouse, which then dispatches vehicles for distribution. Designated suppliers handle the replenishment demand generated by district warehouses, delivering food to the respective district warehouses.
In the context of COVID-19 pandemic, the government has established a district-based oversight of the government-led food supply chain, with a particular emphasis on warehouses. Therefore, this study incorporates 16 district warehouses into the model. Additionally, in response to the pandemic, we assumed that 1000 suppliers are designated to ensure a consistent supply. As the number of suppliers exceeds the number of district warehouses, a rescaling ratio of 50:1 has been implemented. This has resulted in the inclusion of 20 suppliers and 120 communities in the model, while there are 1000 designated suppliers with 6000 communities. This adjustment aims to accurately reflect the real-life scenario without altering the network structure.
The parameters for Scenario 0 are detailed in Tab.1. In the simulation model, the scaled community generates an average daily demand of 305 tons. To better represent the food demand characteristics of the residents, the order generation interval was set to one day, and the expected lead time (ELT) was also one day. During the pandemic, vehicles played a crucial role in food distribution, with each district warehouse equipped with two type I vehicles having a capacity of 305 tons, ensuring that one type of transportation could fulfill a community order. According to speed limits, these vehicles were set to operate at a speed of 50 km/h, with loading and unloading processing times of 0.5 h each. Due to the limited number of designated suppliers and the significant spatial distance between them and district warehouses, each designated supplier is assigned five type II vehicles, each capable of carrying 610 tons. These vehicles also operate at a speed of 50 km/h, with loading and unloading processing times of 1 h each. In the AnyLogistix software, an (s, S) inventory control policy was implemented for both district warehouses and designated suppliers, with an inventory check period of one day. Considering the perishable nature of fresh vegetables such as tomatoes, the (s, S) inventory control policy parameters for district warehouses are set at 1200 and 12000 tons, ensuring that the available inventory can sustain a steady supply for 4–5 days after replenishment. Given the pre-existing food stocks in district storage departments before the pandemic, the initial stock of district warehouses in this study was established at 6,000 tons. To prevent designated suppliers from experiencing inventory shortages when supplying to district warehouses, the inventory-related parameters for the designated suppliers in the simulation model were configured to be twice that of the district warehouses. Specifically, the values of s were set at 2,400 tons and S at 24,000 tons, with an initial inventory of 12,000 tons. Lastly, to ensure that the average daily capacity of the designated supplier aligns with the total normal demand, the production time for each ton of product for each designated supplier was set to 40 s.
3.2 Simulation analysis
The simulation results for Scenario 0 are shown in Fig.2. Except for Fig.2(b), the x-axis of the other images indicates the duration of the pandemic. Specifically, 0–19 days corresponds to Phase I, 20–47 days to Phase II, 48–66 days to Phase III, and 67–82 days to Phase IV.
Fig.2(a) depicts the statistics of the ELT service level by orders, which represents the proportion of daily on-time orders to total demand orders. Our findings indicate a significant decline in the ELT service level by orders during Phase II, indicating a clear shortage of food during that period. However, there is a swift improvement in the ELT service level by orders during Phase III.
Fig.2(b) illustrates the statistics of lead time, which represents the time taken to complete all orders issued by the community. The x-axis denotes the completion time. The results show that the majority of orders in Scenario 0 have a lead time of 0–2 days. However, a small number of orders still have a lead time of 2 to 12 days, exceeding the expected lead times.
Fig.2(d) displays the statistics of late orders, indicating the number of orders that are delivered late per day. It is observed that late orders are most frequent during the first three phases, with Phase II recording the highest number of late orders.
Fig.2(e) presents the statistics of order backlog, representing the number of orders that do not have the required quantity of food products. Our findings reveal that the order backlog is most pronounced during the first three phases, with Phase II having the largest backlog and a significant increase in the number of backlog orders.
Fig.2(c) depicts the statistics of the daily available inventory at the district warehouses, which represents the sum of the daily available inventory at all district warehouses. By analyzing the data in Fig.2(d) and Fig.2(e) in comparison with the distribution of district warehouses and communities, it can be inferred that during Phase I, when the daily available inventory at district warehouses fluctuates significantly, backlog orders are generated, and the number of late orders is similar to that of backlog orders. This suggests a slight shortage of available inventory in some district warehouses. In Phase II, the daily available inventory at district warehouses is relatively low but not close to zero. At the same time, the number of backlog orders increases significantly, while the quantity of late orders remains high, indicating a serious shortage of daily available inventory at some district warehouses.
Fig.2(f) presents the statistics of the daily available inventory of the designated suppliers, which represents the sum of the daily available inventory at all designated suppliers. The results demonstrate substantial fluctuations in the daily available inventory of designated suppliers during Phases I and II. Notably, in Phase II, the daily available inventory of designated suppliers tends to approach zero, indicating a significant shortage in their daily available inventory.
Fig.2(h) displays the statistics pertaining to the shipped vehicles, delineating the daily transportation times for different vehicle types. The red curve corresponds to vehicle type I, while the blue curve represents vehicle type II. Notably, during Phase II, there is a significant surge in shipped vehicles of type I, subsequently maintaining a high level. This indicates a substantial transportation pressure.
Fig.2(g) depicts the statistics on vehicle capacity utilization, illustrating the daily capacity utilization for different vehicle types. The red curve pertains to vehicle type I, whereas the blue curve represents vehicle type II. The capacity utilization of vehicle type I aligns with the changes in demand occurring during Phases I, III, and IV. Conversely, in Phase II, as the demand from the community exceeds the capacity of vehicle type I, the capacity utilization of vehicle type I experiences a slight decrease compared to that of Phase I, while the shipped vehicles of type I witness a significant increase. This indicates an issue of inadequate transportation capacity, particularly during Phase II when there is a surplus of late orders.
By combining the insights obtained from Fig.2 and the spatial distribution of warehouses and communities, it becomes evident that the emergence of backlog orders during Phase I is directly correlated with the decline in the daily available inventory fluctuation in district warehouses. This suggests that certain district warehouses facing heightened demand pressure encounter shortages in their daily available inventory, ultimately leading to food shortages. In Phase II, the production capacity of designated suppliers fails to adequately meet the surge in demand, leading to near-depletion of their daily available inventory. Consequently, district warehouses can only maintain a relatively low level of daily available inventory. Furthermore, the continuous increase in daily backlog orders and the slight daily available inventory margin in district warehouses during Phase II indicate that some warehouses experience substantial demand pressure and severe shortages in their daily available inventory, thereby exacerbating food shortages. Additionally, the correlation observed between increased late orders and heightened transportation times for type I vehicles, coupled with lower capacity utilization, highlights the problem of inadequate transportation capacity, which further contributes to food shortages in Phase II. Moreover, although Phase III experiences a rapid alleviation of food shortages, the supply chain is still affected by residual effects, particularly the lingering pressure from backlog orders during Phase II, resulting in a ripple effect that impacts the entire supply chain.
In summary, the simulation results have shed light on the issue of food shortages faced by the residents in the initial three phases. During Phase I, the shortages were a result of insufficient daily inventory in district warehouses, caused by a surge in demand. Phase II saw severe shortages due to the nearly exhausted daily inventory among designated suppliers. This was further aggravated by limitations in production capacity, mismatched demand growth, deficits in district warehouse inventory, and inadequate transportation facilities. In contrast, Phase III shortages were mainly influenced by ripple effects originating in Phase II.
4 Recovery strategies
In Section 3, we discuss the simulations conducted in this study, which were based on the government-led food supply chain. Specifically, this city faced challenges during the pandemic, such as changing consumer demands and disruptions like labor shortages and limited transportation capacity. Therefore, it is important to explore recovery strategies based on Scenario 0 in Section 4.
4.1 Recovery strategy 1 (RS1)—Opening new district warehouses
This subsection examines the effect of establishing additional district warehouses on the resilience of the government-led food supply chain. The establishment of new district warehouses can effectively address the unequal distribution of warehouses across districts, ease the demand pressure in certain areas, and improve transportation resources in those regions.
4.1.1 Simulation modeling of Scenario 1
Fig.3 illustrates the changes in the government-led food supply chain after the establishment of the new district warehouse. Since the new district warehouse has no initial inventory, the simulation model for this scenario sets the initial inventory of the new district warehouses to zero tons and assigns two type I vehicles to transport food to the communities.
During the COVID-19 pandemic, this recovery strategy was implemented through the establishment of a new district warehouse. Accordingly, the simulation model in this subsection also includes the new district warehouses and conducts simulation experiments.
4.1.2 Comparison results and simulation analysis
Tab.2 provides a comparison between the Scenario 1 and Scenario 0 data after 10 independent experimental replications.
The comparison of data indicators used for analysis in Scenarios 0 and 1 is divided into two types: average ELT service level by orders in each phase and average lead time in each phase. The results show that the implementation of new district warehouses slightly improves the average ELT service level of orders in Phases I and III, while significantly improving the average ELT service level in Phase II. Additionally, there is a significant improvement in average lead times across all phases, suggesting that a well-distributed district warehouse system can enhance the resilience of the government-led food supply chain.
The simulation results for Scenario 1 are depicted in Fig.4. Based on these results, the implementation of new district warehouses significantly enhances the resilience of the government-led food supply chain. This recovery strategy effectively addresses the issue of excessive lead times for certain orders and substantially improves the ELT service level during the initial three phases. However, as shown in Fig.4(c) and Fig.4(e), there is no significant improvement in the available inventory in district warehouses or reduction in backlog orders due to unimproved production capacity among designated suppliers. On the other hand, the decrease in late orders indicates that the implementation of new district warehouses promotes expedited order delivery. Therefore, mitigating food shortages and enhancing resilience within this scenario primarily rely on increased transportation resources and shared demand pressure resulting from the opening of new district warehouses.
In summary, while the implementation of new district warehouses does not fully resolve the issue of generating a large number of order backlogs in Phase II due to constraints in daily available inventory and capacity of designated suppliers, it does enhance the average lead time across all four phases and strengthens the resilience of the government-led food supply chain.
4.2 Recovery strategy 2 (RS2)—Food reallocation between district warehouses
Redistributing food among warehouses is a critical approach in alleviating daily inventory shortages during emergencies. Governments often adopt this strategy to effectively mitigate food shortages. In this section, we thoroughly investigate the impact of implementing this recovery strategy on the resilience of government-led food supply chains. Furthermore, we modify the original model from Scenario 0 by integrating food sharing and redistribution between district warehouses.
4.2.1 Simulation modeling of Scenario 2
Fig.5 illustrates the operational dynamics of the government-led food supply chain during the reallocation of food between district warehouses. Under this recovery strategy, when a warehouse experiences a shortage of food supplies, other district warehouses step in to redistribute food and alleviate the deficit. In contrast to Scenario 0, in this scenario, the replenishment demand of the district warehouse is generated and shared between designated suppliers and district warehouses, who collaborate to transport food to the designated warehouse. Each district warehouse is equipped with a type III vehicle for food transportation, sharing the same parameters as the type I vehicle.
4.2.2 Comparison of results and simulation analysis
Tab.3 provides a comparison of the results between Scenario 2 and Scenario 0 after conducting 10 independent replications of the experiment. Our findings indicate that food reallocation between district warehouses has a negative impact on all phases except Phase IV. The average ELT service level by orders slightly decreases, and the average lead time slightly increases when this strategy is adopted in Phase I and Phase III. Moreover, the negative impact is most severe when this recovery strategy is adopted in Phase II, with a significant decrease in the average ELT service level by orders and a significant increase in the average lead time.
Fig.6 presents the simulation results of Scenario 2. Based on these results, food reallocation between district warehouses severely undermines the resilience of the government-led food supply chain. It can be observed that food reallocation between district warehouses leads to a significant reduction in the daily available inventory at the district warehouses, resulting in a significant increase in the number of backlog orders. Consequently, the number of late orders increases, leading to a significant reduction in the ELT service level by orders. Furthermore, analysis of vehicle type I statistics reveals that food inventory being transported between district warehouses often fails to convert into effective available inventory in time, further compromising the performance of the government-led food supply chain. Additionally, in Phases III and IV, the lower level of ELT service level by orders suggests that the government-led food supply chain is further affected by a negative ripple effect. Particularly, backlog orders must be addressed in Phases III and IV, resulting in a significant increase in late orders and ultimately impacting the ELT service level by orders and the resilience of the government-led food supply chain.
In summary, the reallocation of food between district warehouses has a negative impact on the resilience of the government-led food supply chain. The implementation of this strategy results in a significant amount of food inventory being in transit, which delays its availability. Consequently, there is a backlog of orders due to a scarcity of food products, leading to an increase in late orders and exacerbating food shortages. Additionally, the ripple effect caused by this strategy significantly affects the performance of the supply chain in Phases III and IV.
4.3 Recovery strategy 3 (RS3)—Opening municipal emergency warehouses
The opening of municipal emergency warehouses is a common measure taken by governments during major public emergencies. Therefore, this subsection aims to explore the impact of enabling municipal emergency warehouses on the resilience of the government-led food supply chain.
4.3.1 Simulation modeling of Scenario 3
Fig.7 illustrates the government-led food supply chain with the municipal emergency warehouse opened. In this scenario, the replenishment demand from district warehouses is evenly divided between municipal emergency warehouses and designated suppliers, who collaborate to distribute food to the corresponding district warehouse.
Due to the confidential nature of the inventory in municipal emergency warehouses, this study assumes that these warehouses have an infinite inventory. Each municipal emergency warehouse is equipped with five type II vehicles, which is consistent with the number of type II vehicles used by the designated suppliers.
4.3.2 Comparison of results and simulation analysis
Tab.4 presents a comparison of the results from Scenarios 3 and 0 based on 10 independent experiment replications. Our findings reveal that opening municipal emergency warehouses during Phase I does not significantly impact the average ELT service level by orders, although it does slightly reduce the average lead time. However, the most noticeable positive impact of this recovery strategy is observed in Phase II, where both the average ELT service level and the average lead time of orders significantly improve.
The simulation results for Scenario 3 are presented in Fig.8. According to these results, the opening of municipal emergency warehouses effectively enhances the resilience of government-led food supply chains. By opening these warehouses, the daily available inventory in district warehouses increases within the range of 37-47 days. This helps reduce the number of backlog orders significantly, resulting in fewer late orders and alleviating food shortages in Phase II. Ultimately, this recovery strategy significantly addresses the issue of excessive lead time for certain orders, while also slightly improving the ELT service level for orders.
In summary, the opening of municipal emergency warehouses contributes to enhancing the resilience of the government-led food supply chain in Phase II. This strategy effectively addresses inventory shortages in district warehouses by providing essential food supplies, resulting in a substantial reduction in backlog and late orders, ultimately alleviating food shortages.
4.4 Recovery strategy 4 (RS4)—Food assistance from other provinces
This subsection seeks to examine the impact of food assistance from other provinces on the resilience of the government-led food supply chain.
4.4.1 Simulation modeling of Scenario 4
The government-led food supply chain, supported by food assistance from other provinces, is depicted in Fig.9. To simplify the model without losing generality, this scenario focuses solely on food assistance from two adjacent provinces. When the inventory level of the district warehouse reaches the re-order point, the replenishment demand for the district warehouses is evenly divided between the designated suppliers and the assistance warehouse. These entities work collaboratively to distribute food to the respective district warehouses. Similarly, the replenishment demand for the assistance warehouse is evenly distributed among the designated suppliers, who then distribute the food products to the assistance warehouse. Given the substantial quantity of food supply assistance provided by other provinces and the speed limits on highways the scenario assigns vehicle type IV to transport the assisted food. This vehicle has a capacity of 1220 tons and operates at a speed of 100 km/h The loading and unloading processing times are both set at 2 h. The assistance warehouse adopts the (s, S) inventory control policy, with s set at 0 tons and S set at 40,000 tons. This policy takes into account the large quantities of food provided by the provinces and their ease of preservation. In addition, five type II vehicles are allocated to transport food from the assistance warehouse to the district warehouse, matching the number of type II vehicles provided by the designated suppliers.
4.4.2 Comparison of results and simulation analysis
Tab.5 provides a summary of the results from Scenario 4 compared to Scenario 0 after 10 independent replications. Our findings show that food assistance from other provinces, as a recovery strategy, improves the food shortage situation in Phase II. It leads to a higher average ELT service level by orders and a shorter average lead time compared to Scenario 0.
The simulation results of Scenario 4 are depicted in Fig.10. In Fig.10(g) and Fig.10(h), the green curve represents type I vehicles, the blue curve represents type II vehicles, and the red curve represents type IV vehicles. According to the simulation results of Scenario 4, food assistance from other provinces has an impact on a portion of Phase II, with its ripple effect extending to Phase III. Compared to the simulation results of Scenario 0, there is a decrease in the ELT service level by orders ranging from 38 to 47 days. This strategy leads to a reduction in the daily available inventory in district warehouses, resulting in an increase in backlog orders and a higher number of late orders. Furthermore, due to the ripple effect, this scenario also exhibits lower ELT service levels by orders compared to Scenario 0 during certain intervals of Phase III, particularly from 48 to 54 days.
In summary, while the average ELT of orders suggests that food assistance from other provinces can generally mitigate food shortages in Phase II, the simulation of this scenario indicates that it has a negative impact on part of Phase II and its ripple effect will extend to Phase III. This negative effect may be attributed to longer food delivery and loading times.
4.5 Recovery strategy 5 (RS5) - Enhancing transportation capacity
When the average daily demand of a community exceeds the capacity limits of a certain vehicle, introducing vehicles with larger capacity can effectively enhance transportation capacity. Therefore, this subsection aims to explore the impact of introducing vehicles with a larger capacity in Phase II to enhance transportation capacity, thereby strengthening the resilience of the government-led food supply chain.
4.5.1 Simulation modeling of Scenario 5
Based on the model of Scenario 0, vehicles with larger capacity are introduced in Phase II to transport goods between the district warehouses and the communities. The goal is to investigate the impact of enhancing transportation capacity on the resilience of the government-led food supply chain. This simulation model corresponds to Scenario 5. In Phase II, each district warehouse replaces its type I vehicles with two type V vehicles. The capacity of the type V vehicles is set at 400 tons to match the increased demand, and their speed is also set at 50 km/h.
4.5.2 Comparison of results and simulation analysis
Tab.6 compares the results of Scenario 5 and Scenario 0 after conducting 10 independent replications of the experiment. The results indicate that utilizing larger capacity vehicles to enhance transportation capacity in Phase II can significantly improve the average ELT service level of orders and considerably reduce lead times, thus effectively mitigating food shortages.
Fig.11 illustrates the simulation results of Scenario 5. In Fig.11(g) and Fig.11(h), the green curve represents type I vehicles, the blue curve represents type II vehicles, and the red curve represents type V vehicles.
According to the simulation results of Scenario 5, increasing transportation capacity has a significant positive impact on the resilience of the food supply chain and helps alleviate food shortages. The statistics of ELT service level and lead time confirm that employing larger capacity vehicles to enhance transport capacity in Phase II has a ripple effect, contributing to improving the resilience of the food supply chain in Phase III. Further analysis of vehicle capacity utilization and shipped vehicles data provides insights into the underlying mechanism of this strategy.
In Scenario 5, the capacity of type V vehicles surpasses that of type I vehicles, resulting in a lower number of shipped vehicles when type V vehicles are used to transport products between district warehouses and communities. This strategy enhances transportation capacity by minimizing the number of late orders through improved efficiency, rather than reducing the number of backlog orders. Additionally, this recovery strategy also leads to a significant decrease in late orders at the beginning of Phase III, further enhancing the resilience of government-led food supply chains.
In summary, employing larger capacity vehicles to enhance transportation capacity during Phase II significantly enhances the resilience of the government-led food supply chain. Furthermore, the ripple effect of this recovery strategy contributes to promoting supply chain resilience during Phase III.
4.6 Recovery strategy 6 (RS6) - Increasing the frequency of inventory checks
This subsection aims to investigate the impact of more frequent inventory checks on the resilience of the government-led food supply chain. By conducting inventory checks more frequently, district warehouses can replenish timely and ensure designated suppliers produce on time, potentially enhancing overall resilience.
4.6.1 Simulation modeling of Scenario 6
In Scenario 0, both district warehouses and designated suppliers conducted daily inventory checks. However, during pandemics, district warehouse managers often opt for more frequent checks, typically 2–3 times a day. Simulation modeling of Scenario 6 explores the effects of increasing the frequency of inventory checks by setting the inventory check cycle to 0.5 days, effectively doubling the frequency of checks.
4.6.2 Comparison of results and simulation analysis
Tab.7 compares the results of Scenario 6 compared to Scenario 0 after 10 independent experimental replications.
Our findings indicate that increasing the frequency of inventory checks is an effective strategy for mitigating food shortages in the initial three phases. The most significant improvements are observed in Phases I and II, with notable enhancements in both the average ELT service level by orders and the average lead time. Notably, in Phase III, this strategy helps maintain the average ELT service level by orders at a consistent value of 1.
The simulation results of Scenario 6, as shown in Fig.12, further emphasize the enhanced resilience of the government-led food supply chain. Compared to Scenario 0, more frequent inventory checks facilitate faster restocking for district warehouses and designated suppliers. Consequently, in Phases I and II, there is a substantial increase in the daily available inventory of district warehouses, leading to a reduction in the number of backlog orders and a more even distribution of late orders. As a result, there is a significant improvement in order service levels during these two phases. However, increasing the frequency of inventory checks adversely affects the transition period from Phase II to Phase III, with the ELT service level by orders recovering at a slower pace until reaching 1. This phenomenon can be attributed to the evenly distributed occurrence of late orders witnessed during Phase II.
In summary, increasing the frequency of inventory checks effectively boosts the daily available inventory of district warehouses, mitigates the number of backlog orders, and evens out the distribution of late orders. This leads to notable enhancements in the ELT service level by orders during the initial two phases, thereby improving overall resilience. However, there are also some negative effects. Specifically, the ELT service level by orders takes longer to recover to 1.
4.7 Recovery strategy 7 (RS7)—Increasing production capacity
This subsection aims to investigate the impact of increasing production capacity on the resilience of the government-led food supply chain, as it can effectively help designated suppliers alleviate the shortage of daily available inventory in Phase II.
4.7.1 Simulation modeling of Scenario 7
In Scenario 0, the designated suppliers have a production time of 40 s per ton of product. In this subsection, the production efficiency is improved by 25%, reducing the production time to 30 s per ton of product. This simulation model corresponds to Scenario 7.
4.7.2 Comparison of results and simulation analysis
Tab.8 compares the results of Scenario 7 and Scenario 0 after 10 independent experimental replications. The results indicate that increasing production capacity in Phase II helps increase the average ELT service level by orders, in addition to reducing the average lead time.
The simulation results for Scenario 7 are depicted in Fig.13. Based on these findings, increasing production capacity proves to be effective in improving resilience and mitigating food shortages. Compared to Scenario 0, the daily available inventory at designated suppliers rises significantly in Phase II as a result of the increased production capacity. Consequently, the daily available inventory at district warehouses also increases during this phase, leading to a significant reduction in backlog orders. Ultimately, the number of late orders decreases in Phase II, albeit to a lesser extent compared to the total number of orders, resulting in a slight improvement in the ELT service level by orders.
In summary, increasing production capacity during Phase II substantially strengthens the resilience of the government-led food supply chain. This strategy enables district warehouses to replenish their inventory promptly by increasing the daily available inventory at designated suppliers. Consequently, the number of backlog and late orders decreases, thereby enhancing the overall resilience of the system.
5 Extension
This section aims to assess the robustness and stability of the aforementioned recovery strategies in the face of severe disruptions in the government-led food supply chain. The model simulates two disruption scenarios across all phases: one involving the closure of designated suppliers and the other involving the closure of district warehouses. The study examines seven recovery strategies: opening new district warehouses (RS1), opening municipal emergency warehouses (RS3), receiving food assistance from other provinces (RS4), enhancing transportation capacity (RS5), increasing the frequency of inventory checks (RS6), and increasing production capacity (RS7). As Section 4 demonstrates that food reallocation between district warehouses (RS2) has detrimental effects, this strategy is not explored in this section.
5.1 Evaluation of recovery strategies with the closure of the designated suppliers
This subsection simulates a scenario where designated suppliers are closed and evaluates the performance of six recovery strategies implemented in response. In this scenario, each designated supplier in each phase has a 50% probability of closure due to the pandemic, reopening after a 14-day closure period.
Tab.9 provides an overview of the results, with Scenario 0 serving as the baseline. Scenario 8.1 represents the closure of designated suppliers in Phase I, whereas subsequent scenarios simulate the implementation of various recovery strategies based on Scenario 8.1. Our findings indicate that the closure of designated suppliers in Phase I does not have an immediate impact on supply chain performance, but significantly affects the average ELT service level by orders and average lead time in Phase II. Compared to the results in Section 4, the opening of municipal emergency warehouses demonstrates a significantly higher recovery effect on the average ELT service level by orders, with a significantly shorter average lead time observed in Phase II, second only to enhancing transportation capacity. Additionally, while both opening new district warehouses and increasing the frequency of inventory checks notably reduce the average lead time, their recovery effect on the average ELT service level by orders is significantly diminished.
In Tab.10, Scenario 0 serves as the baseline scenario, while Scenario 8.2 simulates the closure of designated suppliers in Phase II. The other scenarios represent the implementation of relevant recovery strategies based on Scenario 8.2. Our findings indicate that the closure of designated suppliers in Phase II has a negative impact on performance in Phases II and III. Compared to the results in Section 4, strategies such as enhancing transportation capacity, opening municipal emergency warehouses, food assistance from other provinces, and increasing production capacity exhibit significantly higher effectiveness in recovering the average ELT service level by orders in Phase II. Moreover, the most notable improvements in average lead time are achieved through enhancing transportation capacity, followed by opening municipal emergency warehouses. In Phase III, all six strategies promote some degree of recovery, with the top-performing strategies being the opening of new district warehouses, food assistance from other provinces, and increasing production capacity.
In Tab.11, Scenario 0 serves as the baseline scenario, Scenario 8.3 simulates the closure of designated suppliers in Phase III, and Scenario 8.4 simulates the closure of designated suppliers in Phase IV. The results indicate that there is almost no impact on the government-led food supply chain when the designated supplier closures occur in Phase III and Phase IV. Therefore, the implementation and evaluation of recovery strategies based on these two scenarios were not explored in this subsection.
5.2 Evaluation of recovery strategies with the closure of the district warehouses
In this subsection, we examine a simulated disruption scenario that involves the closure of district warehouses and the subsequent implementation of six recovery strategies. The scenario assumes that each district warehouse in each phase faces a probability of closure due to the pandemic, and operations resume 14 days after closure. We have assumed a closure probability of 0.5, and the experiment was conducted across phases.
Tab.12 presents the simulation results. Scenario 0 serves as the baseline scenario, while Scenario 9.1 simulates the closure of district warehouses in Phase I. Subsequent scenarios implement relevant recovery strategies based on Scenario 9.1. The findings indicate that the closure of district warehouses in Phase I has a negative impact on supply chain performance in both Phases I and II, with Phase II experiencing the most significant adverse effects. Compared to the results in Section 4, strategies such as enhancing transportation capacity, opening new district warehouses, and opening municipal emergency warehouses show improved recovery. However, other strategies show a slight decrease in recovery. Notably, enhancing transportation capacity and opening municipal emergency warehouses notably improve the average ELT service level by orders and average lead time in Phase II.
Tab.13 summarizes the results of our simulations. Scenario 0 is the baseline scenario, while Scenario 9.2 simulates the closure of the district warehouses in Phase II. The other scenarios implement relevant recovery strategies based on Scenario 9.2. Our findings reveal that the closure of district warehouses in Phase II has a negative impact on supply chain performance in Phases II and III, with Phase II experiencing the most significant adverse effects. Compared to the results in Section 4, only opening new district warehouses shows improvement in both the average ELT service level by orders and the average lead time for orders in Phases II and III. Furthermore, enhancing transportation capacity and opening municipal emergency warehouses in Phase II demonstrate enhanced effectiveness, significantly improving the average ELT service level by orders and the average lead time. On the other hand, the performance of food assistance from other provinces and increasing production capacity appears to be unstable, and this strategy may even lead to negative impacts.
Tab.14 presents the simulation results, wherein Scenario 0 serves as the baseline scenario. Scenario 9.3 simulates the closure of district warehouses in Phase III, and subsequent scenarios implement recovery strategies based on Scenario 9.3. Our findings reveal that the closure of district warehouses in Phase III has a negative impact on supply chain performance in Phases III and IV. The top three recovery strategies in Phases III and IV are the opening of new district warehouses, the opening of municipal emergency warehouses, and the improvement of inventory check frequency. These three strategies significantly improve the average ELT service level by orders and reduce the average lead time. In contrast, food assistance from other provinces and increasing production capacity yield less stable results.
In Tab.15, Scenario 0 is again considered as the baseline scenario. Scenario 9.4 simulates the closure of district warehouses in Phase IV, with other scenarios implementing recovery strategies based on Scenario 9.4. Our findings indicate that the closure of district warehouses in Phase IV only adversely affects performance in the current phase. Compared to the results in Section 4, the opening of new district warehouses and the opening of municipal emergency warehouses are the only strategies that demonstrate a notable recovery effect in Phase IV.
In summary, strengthening transportation capacity and opening new district warehouses enhance the stability and resilience of the supply chain in the face of potential disruptions to designated suppliers and district warehouses. Surprisingly, the opening of municipal emergency warehouses has a positive recovery effect. However, strategies such as food assistance from other provinces, increasing inventory check frequency, and increasing production capacity appear less stable and may even yield negative effects.
6 Discussion and implications
6.1 Comparative analysis
The findings of previous studies have been affirmed by the present study. This study establishes that the discrepancy between production capacity and the growth in demand is one of the primary factors leading to food shortages. This is consistent with the conclusions drawn by various studies. For instance, Ali et al. (
2021a) and Singh et al. (
2021) found that the imbalance between supply and demand is a major cause of disruptions in the supply chain. In addition, Burgos and Ivanov (
2021) demonstrated that surges in demand can have an impact on the resilience of the supply chain. Furthermore, Rahman et al. (
2021) emphasized the importance of maximizing production capacity to meet the increase in consumer demand. Similarly, this study arrives at a similar conclusion based on the performance of increasing production capacity in Phase II. Moreover, this study examines the role of new district warehouses as backup facilities, effectively optimizing inventory levels and ultimately contributing to resilience. Thus, the investigation into the implementation of new district warehouses in this study confirms the findings of Zheng et al. (
2022), which suggest that the addition of backup facilities and the optimization of their inventory levels are effective measures for responding to the pandemic.
This study presents contrasting findings and novel insights compared to previous research. First, unlike the study by Burgos and Ivanov (
2021), which suggested a minimal impact of transportation disruptions, our research reveals that inadequate transportation capacity is a crucial factor contributing to food shortages. This can be attributed to various realistic scenarios and supply chain patterns. Secondly, while Cui et al. (
2023) and Rahman et al. (
2021) demonstrated the positive effects of emergency supply policies, our study goes further to show that the establishment of municipal emergency warehouses falls short in terms of resilience impact compared to measures such as opening new district warehouses under normal circumstances. Thirdly, while confirming the importance of incorporating backup facilities and optimizing inventory levels as highlighted by Zheng et al. (
2022), our study also underscores the significance of adequate distribution density among district warehouses in supply chain performance. Furthermore, previous studies have established that disruptions in warehouses and suppliers significantly affect supply chain resilience (
Ding et al., 2022;
Burgos and Ivanov, 2021). Therefore, based on the aforementioned findings, our study introduces two additional disruption scenarios to evaluate the stability and robustness of recovery strategies. It demonstrates that enhancing transportation capacity and opening new district warehouses exhibit strong stability and robustness, while the opening of municipal emergency warehouses surprisingly yields positive recovery effects. Finally, our research explores rarely examined recovery strategies, including food assistance from other provinces, increased frequency of inventory checks, and food reallocation between district warehouses. Notably, our study uncovers the detrimental impact of food reallocation between district warehouses on the government-led food supply chain.
There are also similarities and differences between this study and other research approaches. On the one hand, non-simulation methods studies provide the theoretical and practical context for our research, such as government intervention (
Cui et al., 2023), food shortages (
Bukari et al., 2022;
Bidisha et al., 2021), supply chain disruptions (
Farcas et al., 2020;
Alabi and Ngwenyama, 2023), and recovery strategies (
Sid et al., 2021;
Liao et al., 2022). However, these studies face challenges in integrating food shortages, disruptions, and recovery strategies. Furthermore, they often fall short in replicating real scenarios and lack a variety of evaluation indicators, thereby limiting their ability to offer nuanced guidance. Therefore, this study employs a simulation methodology while introducing various evaluation indicators like lead time and ELT service level by orders to achieve a precise assessment of supply chain resilience.
On the other hand, compared to simulation methods research such as Ivanov (
2020) and Moosavi and Hosseini (
2021), this study refers to their simulation frameworks but creatively constructs the framework and model of government-led food supply chain with consideration of multiple elements such as disruptions and recovery strategies. Moreover, based on the findings of Zheng et al. (
2022) and Rahman et al. (
2021), this study draws further conclusions. For instance, while confirming the importance of adding backup facilities and optimizing their inventory levels as highlighted by Zheng et al. (
2022), this study further underscores the importance of proper distribution. Additionally, based on the findings of Ding et al. (
2022) and Burgos and Ivanov (
2021), this study introduces two additional disruption scenarios, namely district warehouses disruption and designated suppliers disruption, to explore the stability and robustness of the recovery strategies that have not been discussed in these papers. In summary, by integrating food shortages, disruptions, recovery strategies, evaluation indicators, and simulation methods based on the aforementioned studies, this research enhances the model’s rationality and contributes to its value.
6.2 Theoretical implications
To underscore the contributions of this study to the field of government-led food supply chain management during public health emergencies, the theoretical implications derived from the results of each recovery strategy are discussed as follows.
First, when faced with insufficient transportation capacity such as demand exceeding the capacity of vehicles, recovery strategies to enhance transportation capacity should be prioritized. The results of this study suggest that enhancing transportation capacity can significantly improve resilience and alleviate food shortages in Phase II. Moreover, this recovery strategy has good stability and robustness even in the face of disruptions.
Secondly, the opening of new district warehouses is necessary and effective. Our research indicates that by opening new district warehouses, we cannot only alleviate demand pressure on warehouses in specific regions, but also increase transportation resources. This, in turn, enhances the resilience of the supply chain and mitigates food shortages.
Thirdly, in cases where the designated supplier capacity is insufficient to meet demand, diversifying food supply sources and strategies to increase production capacity can help fulfill the needs of the community. Our findings highlight three strategic approaches that can improve the resilience of the government-led food supply chain in Phase II. These include obtaining food assistance from other provinces, establishing municipal emergency warehouses, and enhancing production capacity. Remarkably, the opening of municipal emergency warehouses when faced with possible disruptions has proven to have a positive effect.
Additionally, maintaining a balanced and appropriate frequency of inventory checks is vital in sustaining adequate levels of daily available inventory. Although our findings support the effectiveness of increased inventory check frequencies as a recovery strategy, administrators must carefully adjust the frequency during different phases. We discovered a potential trade-off between Phase II and Phase III, where excessively frequent checks may impede supply chain resilience. Furthermore, the stability of this strategy during disruptions remains uncertain, necessitating cautious implementation and monitoring.
Finally, precision is crucial when executing emergency replenishment by redistributing food between district warehouses. Our findings emphasize the importance of accounting for the decrease in daily available inventory at the district level. Inaccurate reallocation of food between district warehouses can have significant negative impacts on the resilience of the government-led food supply chain.
6.3 Managerial insights
In times of public health emergencies, such as the COVID-19 pandemic, ensuring the resilience of the government-led food supply chain is of utmost importance in securing the daily dietary needs of residents. Therefore, it is essential for government policymakers to prioritize addressing shortages within this supply chain and implementing precise and comprehensive recovery strategies aimed at enhancing its resilience. Likewise, both enterprises (designated suppliers) and consumers (community residents) can contribute to the resilience of the supply chain.
For governments, prioritizing the enhancement of transportation capacity is crucial to ensure the smooth distribution of food supplies during pandemics, particularly considering the rising demand. This necessitates investing in infrastructure and logistics to effectively overcome logistical challenges. Furthermore, increasing the number of district warehouses and ensuring their even distribution density can help alleviate food shortages and strengthen resilience. The strategic placement of warehouses can greatly improve accessibility and distribution efficiency. Additionally, governments should closely monitor and manage production capacity to align with the evolving demand during different phases of the pandemic. Anticipating demand growth and adjusting strategies accordingly can prevent food shortages. Lastly, opening municipal emergency warehouses as a preemptive measure is essential. This recovery strategy cannot only help improve supply chain resilience but also demonstrate good stability and robustness in the event of dangerous supplier and warehouse disruptions.
For enterprises, close collaboration with government authorities is significant to effectively address supply chain challenges. This entails sharing information, resources, and expertise to enhance resilience and responsiveness. Moreover, enterprises should adopt adaptive supply chain strategies that can promptly respond to changing demand patterns and disruptions. Flexibility in production and distribution processes is crucial for maintaining continuity. Most importantly, actively increasing production capacity helps to improve resilience and alleviate food shortages during the demand growth phase.
As for consumers, practicing responsible consumption habits by avoiding panic buying and hoarding during pandemics is essential. Responsible consumption not only helps prevent shortages and ensures equitable access to essential food supplies for all, but also alleviates pressure on governments and enterprises.
By considering these insights from the government, business, and consumer perspectives, stakeholders can collaborate effectively to enhance the resilience and responsiveness of the government-led food supply chain during public health emergencies such as the COVID-19 pandemic.
7 Conclusions
Despite our evolving understanding of the COVID-19 virus, this pathogen will continue to have a lasting impact on industrial and supply chains, particularly those critical to sustaining the daily diet of the population. Whether governments implement lax or stringent measures, their primary goal remains to ensure food supply stability during the pandemic control phases. When confronting future public health crises, policymakers must prioritize measures to safeguard the dietary needs of the population and enhance the resilience of government-led food supply chains.
Therefore, this study was conducted to analyze and simulate food shortages, resilience, and recovery strategies in the government-led food supply chain during the COVID-19 pandemic. The results of our study indicated that food shortages were primarily caused by inadequate transportation capacity, uneven distribution of district warehouses, and mismatched production capacity. To address these issues and improve the resilience of the supply chain, it is recommended to enhance transportation capacity, establish new district warehouses, and increase production capacity. Additionally, opening municipal emergency warehouses in the event of disruptions can have a positive recovery effect. Although food assistance from other provinces and frequent inventory checks generally boosted resilience, they could also have negative consequences. Surprisingly, reallocating food between district warehouses had severe adverse effects on the supply chain.
However, it is important to note that this study has several limitations and further research and development are warranted. First, this study was retrospective, focusing on shortages and recovery strategies during the COVID-19 pandemic. Future studies should focus on predictive exploration based on the findings of this retrospective study. Additionally, while this study utilized simulation experiments to explore shortage patterns and recovery strategies in the government-led food supply chain, an optimization model should be established to identify the most relevant parameters for recovery strategies through accurate optimization studies. Lastly, supply chain reconstruction has emerged as a critical factor in addressing supply chain disruption risks, emphasizing the need for further discussions on recovery strategies that impact the structure of the supply chain.
PDF
(4693KB)
