Mitigating transportation disruptions in the Australian household hydrogen supply chain

Abstract

Despite growing interest in hydrogen as a clean energy source, limited research has explored the long-term operational challenges facing Australia’s household hydrogen supply chain (HHSC), particularly under transportation disruptions. This study investigates transportation disruptions in vehicles and routes within the Australian HHSC planned over the period 2026 to 2090. It focuses on disruptions across three distribution tiers: national distribution centers (NDCs), regional distribution centers (RDCs), and local distribution centers (LDCs). A multi-period network optimisation model is developed using scenario-based analysis to simulate and evaluate the impacts of various disruptive events over time. Mitigation strategies, including rerouting, additional vehicle hiring, and safety stock positioning at RDCs, are assessed for their effectiveness. The results reveal that combined disruptions, affecting both vehicles and routes, have the most severe impact on the HHSC, particularly when multiple routes and vehicles across NDCs, RDCs, and LDCs are simultaneously affected. While individual disruptions, such as those impacting only routes or only vehicles, also influence performance, their effects are comparatively less critical than the impact of combined disruptions. Mitigation strategies targeting routes, vehicles, and combined disruptions lead to higher demand fulfilment and lower penalty costs, resulting in a significant increase in overall profit. These outcomes are achieved despite the added costs associated with rerouting, additional vehicle hiring, and maintaining safety stock. The findings highlight the importance of targeted, disruption-specific planning to improve demand fulfilment and reduce penalty costs and provide practical implications for managing transportation disruptions in the HHSC.

Graphical abstract

Keywords

hydrogen supply chain / transportation disruptions / vehicle disruptions / route disruptions / disruption mitigation

1 Introduction

The establishment of an efficient and resilient household hydrogen supply chain (HHSC) is fundamental to Australia’s transition toward a cleaner and more sustainable energy future. As a low-carbon energy carrier, hydrogen holds considerable promise in addressing the evolving energy demands of residential sectors, offering a viable pathway for decarbonising household energy consumption. Ensuring the success of the HHSC will require the reliable transport of hydrogen across multiple supply chain stages, from national distribution centers (NDCs) to regional distribution centers (RDCs) and, ultimately, to local distribution centers (LDCs). A recent example that demonstrates the importance of supply chain resilience is the 2022 Queensland floods, which caused significant road closures and logistical delays across regional areas. In the context of an operating HHSC, such disruptions could compromise hydrogen delivery to households and undermine confidence in hydrogen-based energy systems. As hydrogen infrastructure evolves, securing a dependable supply at each stage is key to supporting household energy demand and reducing the carbon footprint of residential energy consumption.

Previous research focused on a multi-period network optimisation model to enhance the efficiency of the HHSC by optimising distribution networks under ideal situations (Almaraz et al., 2015). Such an approach aligns with existing academic discussions on supply chain optimisation, which highlight the importance of network design in improving resilience, cost-effectiveness, and demand fulfilment in energy supply chains (Datta, 2007; Xu and Bo, 2024). However, while ideal situations provide a benchmark for assessing performance, real-world supply chains are subject to disruptions that can significantly impact operations.

As the hydrogen supply chain (HSC) becomes increasingly integral to the energy transition, there is a growing interest to broaden analytical frameworks beyond ideal situations. Specifically, challenges such as route disruptions, vehicle disruptions, and combined disruptions can influence network efficiency, cost structures, and demand fulfilment. Researchers highlighted that transport disruptions in HSC significantly reduce service reliability and increase operational costs, particularly in decentralised networks (Esmizadeh and Mellat Parast, 2021). Another study highlighted that multi-layered disruptions, such as route and vehicle disruptions, can trigger cascading effects that exacerbate delivery delays and inventory shortfalls, thereby reinforcing the critical need for robust mitigation strategies (Katsaliaki et al., 2022). Addressing these factors through robust optimisation strategies and contingency planning is essential to ensuring the long-term viability of HHSCs, particularly as they evolve to support Australia’s net-zero targets and broader energy security goals (Ally et al., 2015; Dawood et al., 2020). Disruptions, whether caused by vehicle breakdowns, route closures, or adverse weather conditions, have the potential to cause substantial delays, increase costs, and pose significant safety risks (Visentini et al., 2014). Despite the importance of these issues, existing literature on HSC remains limited in its treatment of transportation disruptions, particularly regarding the combined effects of simultaneous vehicle and route disruptions. In a nutshell, the inherently greater complexity of the HSC compared to other conventional energy sources calls for additional research to address its transportation disruptions. This study addresses these gaps by introducing resilience into the HSC model and investigating the impact of transportation disruptions, whether related to vehicles, routes, or both, on overall supply chain performance (Ahmadi-Javid and Seddighi, 2013; Golan et al., 2020; Zhu et al., 2024). This study specifically investigates the impact of transportation disruptions on the HHSC over the period 2026 to 2090, with the objective of analyzing the effectiveness of mitigation strategies to enhance its resilience. By proactively addressing potential challenges, the research seeks to support the development of a more robust and adaptable supply chain capable of delivering a consistent hydrogen supply to households. The effective management and mitigation of such disruptions are critical to safeguarding the reliability of household energy provision, particularly as hydrogen assumes an increasingly prominent role in the decarbonisation of household energy systems. For the reasons mentioned above, two key research questions (RQs) guide this study, as follows:

RQ1: What are the negative impacts of transportation disruptions, such as vehicle and route disruptions, on HHSC?

RQ2: What are the different mitigation strategies that can be applied and how do those strategies mitigate different transportation disruptions?

The remainder of the study begins with a review of existing studies in Section 2, structured around transportation disruptions caused by routes, vehicles, and combined disruptions. It identifies gaps in the current literature, particularly the limited attention given to disruptions affecting HHSC. The study then introduces in Section 3 the structure of the Australian HHSC and outlines the assumptions, modeling parameters, and scenario design related to routes, vehicles, and combined disruptions. Section 4 discusses the research methodology, explaining how hydrogen demand is estimated, how disruption scenarios are developed, and how the optimisation model is formulated over multiple time periods. Section 5 presents findings for each situation under ideal, disrupted situations, with and without mitigation strategies. This is followed by a discussion of practical implications drawn from each type of disruption and the corresponding mitigation strategies in Section 6. The paper concludes by summarizing the main contributions and suggesting future directions for research focused on transportation disruptions in HSC in Section 7.

2 Literature review

The HSC faces unique challenges due to the complexity of producing, storing, and transporting hydrogen. While disruptions in traditional supply chains have been widely studied, research on disruptions specific to the HHSC remains limited. Vehicle and route disruptions, as well as the combined disruptions, are not comprehensively addressed in the existing hydrogen supply chain literature. This review explores the gaps in research related to these disruptions and identifies the need for targeted studies aimed at developing effective mitigation strategies to enhance the resilience of the HHSC. The literature review on transportation disruptions in HHSC is classified into four sections: vehicle disruptions in HSC, route disruptions in HSC, combined disruptions in HSC, and research gaps.

2.1 Vehicle disruptions in HSC

Vehicle disruptions in the HSC presented a significant challenge due to the unique transportation requirements of hydrogen. Unlike conventional energy supply chains, such as those for other energy sources like LNG and LPG, hydrogen had to be transported in specialized vehicles, such as cryogenic trucks or pressurised cylinders, which were costly and less readily available. Research on vehicle disruptions in the HSC remained limited, with much of the existing literature focusing on broader supply chain efficiency and optimisation, rather than addressing the specific risks and mitigation strategies required to manage vehicle disruptions effectively (Aldrighetti et al., 2021; Azadnia et al., 2023; Emenike and Falcone, 2020; Sabio et al., 2010). Studies in conventional supply chains have explored mitigation strategies for vehicle breakdowns, including the use of fleet management systems, deployment of backup vehicles, and optimised vehicle routing to minimise the impact of disruptions. However, these strategies have not been comprehensively applied to HSC, where vehicle disruptions can lead to significant delays and unmet demand due to the limited availability of suitable vehicles and the stringent safety regulations governing hydrogen transport. Furthermore, while some research addressed the optimisation of supply chain models, these models often assumed perfect conditions and did not fully account for the complexities of vehicle unavailability or disruptions in hydrogen transport. The impact of vehicle disruptions on the HSC is particularly severe due to hydrogen’s low energy density and specific handling requirements, which necessitate specialized transport infrastructure and limit the flexibility to substitute or reroute deliveries (Foorginezhad et al., 2021; Hassan et al., 2024). The disruptions of a single vehicle could cause extensive delays, and without proper mitigation, financial losses and unmet demand could escalate rapidly (Patnala et al., 2024). Despite the theoretical exploration of backup vehicle systems and fleet redundancy in other sectors, these concepts remained under-researched in HSC (Fitt, 2022). There is a clear gap in the literature regarding practical mitigation strategies that are both profitable and reliable for hydrogen transport. Appendix 1 presents a summary of previous studies on vehicle disruptions in HSC.

2.2 Route disruptions in HSC

In HSC, the geographical remoteness of production facilities and distribution centers increased the risk of route disruptions (Okonkwo et al., 2021). These facilities were often located far from urban areas, making the transportation routes more susceptible to infrastructure disruptions or extreme weather conditions. Some studies acknowledged the vulnerability of hydrogen transport to such disruptions but largely focused on static route optimisation models that did not adequately account for dynamic responses to route disruptions (Bayram, 2016; Humagain et al., 2020; Valença et al., 2021). Research in other sectors suggested that predictive analytics and dynamic traffic monitoring could help mitigate route disruptions by enabling more adaptive logistics systems (Chung et al., 2015; Wang et al., 2016). However, these solutions were not fully integrated into HSC models, where the infrastructure was limited, and rerouting options were more constrained due to the specialized nature of hydrogen transport vehicles. Furthermore, mitigation strategies for route disruptions, such as rerouting or using alternative transport modes, were applied in conventional supply chains but were not yet well developed for the HSC (Sgarbossa et al., 2022). The unique requirements for hydrogen transport, including the need for refuelling stations and dedicated transport infrastructure, made it difficult to implement rerouting strategies effectively. Additionally, hydrogen transport was subject to strict safety and regulatory standards, which can further limit flexibility when a disruption occurs (Ball et al., 2007). While some studies touched on these issues, there was little research on how to build resilient infrastructure that could mitigate the impact of route disruptions and provide alternative transport options in the event of disruptions (Cox, 2021; Edwards et al., 2021). Appendix 2 presents a summary of previous studies on route disruptions in HSC.

2.3 Combined disruptions in HSC

Combined disruptions, involving both vehicle and route disruptions, presented a significant threat to the efficiency and resilience of the HSC. While individual disruptions, such as vehicle or route disruptions, were explored in various supply chain sectors, the combined disruptions received limited attention, particularly in the context of HSC (Moudio et al., 2022). Hydrogen transportation, with its specific safety requirements and reliance on specialized vehicles and infrastructure, is particularly vulnerable to such disruptions. When both vehicles and routes were disrupted, the options for rerouting or additional vehicles hiring were constrained due to the scarcity of appropriate infrastructure and transport modes (Nazib and Moh, 2020). Most existing research treated vehicle and route disruptions separately, with limited exploration of how these disruptions interacted to exacerbate supply chain delays and costs (Browning et al., 2023; Golan et al., 2020). In the HSC, the implications of combined disruptions were severe, as both transport infrastructure and vehicle availability were often limited. Researchers acknowledged that route disruptions can significantly delay hydrogen delivery, but few studies considered the added complexity when vehicles were simultaneously unavailable (Aldrighetti et al., 2021; Ivanov et al., 2017). Combined disruptions often complicate compliance with safety regulations, particularly when both routes and vehicles must be quickly adapted to ensure timely delivery. While some studies touched on the regulatory aspects of hydrogen transport, the literature lacked comprehensive strategies for maintaining regulatory compliance in the face of combined disruptions (Katsaliaki et al., 2022). Moreover, dynamic monitoring and predictive analytics, which were increasingly used in conventional supply chains to pre-empt disruptions, remained widely unadopted in HSC. Appendix 3 presents a summary of previous studies on combined disruptions in HSC.

2.4 Research gaps

Transportation disruptions in the HSC are a pretty new topic to explore, while conventional supply chains benefit from fleet management systems and backup vehicles, these solutions are not fully adapted to the HSC (Emenike and Falcone, 2020; Sabio et al., 2010).

Vehicle disruptions pose significant challenges to the resilience and reliability of HSC. The scarcity of available vehicles can severely hinder the timely distribution of hydrogen to end-users, directly impacting operational performance. However, much of the existing research on HSC has primarily concentrated on broader optimisation models, often overlooking the specific effects of vehicle disruptions. This gap creates a limited understanding of how vehicle shortages, breakdowns, or delays affect the continuity and stability of hydrogen delivery. A focused investigation into vehicle disruptions is essential to develop targeted strategies that strengthen the operational resilience of HSC(Acar and Dincer, 2018a; Aldrighetti et al., 2021; Azadnia et al., 2023). There is a pressing need for hydrogen-specific research that investigates practical and economically viable mitigation strategies tailored to the unique challenges of hydrogen transport. Route disruptions pose significant challenges in hydrogen logistics, primarily due to the remote locations of production facilities and the limited availability of supporting transport infrastructure (Kljaić et al., 2023). While predictive analytics and dynamic traffic monitoring are used in other sectors, these solutions aren’t widely integrated into HSC (Bayram, 2016; Humagain et al., 2020). The research at the time relies heavily on static optimisation models, failing to account for the dynamic nature of route disruptions and the need for flexible, dynamic responses (Gorji, 2023; Okonkwo et al., 2021; Valença et al., 2021). Future research is needed to focus on incorporating adaptive logistics systems that could respond effectively to route disruptions in hydrogen transport.

The combined disruptions in the HSC remain largely unexplored. Most studies treat these disruptions separately, without considering how their interaction exacerbates delays (Katsaliaki et al., 2022; Moudio et al., 2022). Hydrogen’s specialized transport requirements and regulatory and safety constraints made these disruptions particularly problematic (Edwards et al., 2021).

There is a clear need for comprehensive research into combined disruptions in the HHSC, particularly addressing the combined effects of vehicle and route disruptions. While existing studies focus largely on isolated disruptions, few have examined their interacting impacts within HSC. This research contributes to bridging that gap by exploring integrated mitigation strategies, including rerouting, additional vehicle hiring, and maintaining safety stocks, offering practical implications to mitigate vehicle, route and combined disruptions in HHSC.

3 Problem description and assumptions

To address the identified research gaps, this study develops a multi-period network optimisation model to systematically evaluate a range of probable disruption conditions over the period 2026 to 2090. A scenario-based multi-period optimisation is proposed, integrating multi-period network modeling with scenario analysis to simulate diverse disruption events affecting the HHSC. The scenario analysis is structured around three overarching situations, each representing a distinct operational context: (1) an ideal situation with no disruptions (baseline), (2) a disruption situation without the application of mitigation strategies, and (3) a disruption situation with mitigation strategies in place. Within each situation, a series of scenarios are constructed to capture variability and uncertainty. These scenarios are based on differing assumptions related to the frequency, severity, and combination of disruptions, enabling a detailed and comparative assessment of supply chain resilience. In this study, situations define the general conditions under which the supply chain operates, while scenarios represent specific, assumed realizations or variations within each situation. This structured approach supports a comprehensive understanding of the HSC’s performance under uncertainty and helps identify effective mitigation strategies.

3.1 Structure of HHSC

The structure of HHSC is embedded within a three-tiered arc system comprising NDCs, RDCs, and LDCs. These three arcs are interconnected through two critical nodes: the first connects NDCs to RDCs, and the second links RDCs to LDCs, as presented in Fig. 1. This hierarchical design facilitates the systematic flow of hydrogen from national production and storage hubs to regional distribution facilities, and finally to local delivery points, ensuring efficient downstream supply and demand fulfilment. Moreover, the nodes within the HHSC are connected via multiple transportation routes and supported by a fleet of specialized vehicles. These routes enable the movement of hydrogen between NDCs, RDCs, and LDCs, providing flexibility and redundancy within the network.

3.2 Route disruption scenarios

This section outlines various route disruptions within the HHSC, specifically focusing on the transportation routes between the NDCs, RDCs, and LDCs. These disruptions are categorised into single and multiple route scenarios with scenario code, impacting different supply chain areas as explained in Table 1.

3.3 Vehicle disruption scenarios

This section describes vehicle disruptions within the HHSC, focusing on transportation routes between the NDCs, RDCs, and LDCs. These disruptions are classified into single and multiple-vehicle scenarios, impacting different supply chain areas explained in Table 2.

3.4 Combined disruptions

This section outlines combined disruptions within the HHSC, where both route and vehicle disruptions occur simultaneously with scenario code, impacting different supply chain areas as explained in Table 3.

3.5 Mitigation strategies

Mitigation strategies present a structured approach to addressing supply chain disruptions by categorising strategies based on the affected area, route, vehicle, or combined. Each strategy is identified by a unique code and assessed on whether it involves rerouting, hiring additional vehicles, or maintaining safety stock at RDCs. Route-based strategies (SR) focus on altering transport routes to navigate disruptions such as road blockages or traffic delays, always incorporating rerouting and additional vehicle hiring, though not all include safety stock as a contingency measure. In contrast, vehicle-based strategies (SV) address disruptions caused by vehicle failures or shortages, prioritising hiring extra vehicles and maintaining safety stock rather than rerouting. The most comprehensive approach is seen in combined strategies (SC), which integrate rerouting, additional vehicle hiring, and safety stock management, offering maximum resilience against disruptions. These strategies ensure a robust and adaptable supply chain, enabling HHSC to maintain operations despite unexpected challenges and external uncertainties. Mitigation strategies are presented in Table 4.

4 Research methodology

The methodology compares disruption scenarios within the HHSC, focusing on financial implications, demand fulfilment, and worst-case outcomes. Multiple situations, including an ideal situation and disruptions involving routes, vehicles, and combined, are tested across different time periods. Scenarios without mitigation strategies are compared to those with strategies like rerouting, additional vehicle hiring, and maintaining safety stock. The analysis aims to identify the most effective mitigation strategies for improving supply chain resilience and ensuring reliable hydrogen delivery to Australian households.

4.1 Estimating the demand of household hydrogen

This section outlines the modeling framework used to estimate future household hydrogen demand in Australia. Natural gas and LPG consumption data, along with infrastructure and population statistics, were sourced from the Australian Bureau of Statistics and national planning reports. Customised projections were used for 2026–2090. Data are available upon request. The steps to estimate the demand are as presented in Fig. 2.

4.1.1 Phase-out modeling of LPG and NG

The forecasted demands for LPG and NG over the planning horizon are modeled using exponential decay functions with variable annual reduction rates. Equations (1) and (2) incorporate the expected decline in fossil fuel usage as national and regional energy policies push for cleaner household energy sources. Specifically, the projected demands in year t, relative to a baseline year t_0, are presented as follows.

The hydrogen demand for city c in year t, denoted as D_{c,t} is given by

(1)D_t^{\mathrm{L}\mathrm{P}\mathrm{G}}=D_{t_0}^{\mathrm{L}\mathrm{P}\mathrm{G}}\cdot {\prod }_{T=t_0+1}^t\left(1-r_T^{\mathrm{L}\mathrm{P}\mathrm{G}}\right),

(2)D_t^{\mathrm{N}\mathrm{G}}=D_{t_0}^{\mathrm{N}\mathrm{G}}\cdot {\prod }_{T=t_0+1}^t\left(1-r_T^{\mathrm{N}\mathrm{G}}\right),

where D_t^{\mathrm{L}\mathrm{P}\mathrm{G}} is the forecasted LPG demand in year t, D_t^{\mathrm{N}\mathrm{G}} is the forecasted NG demand in year t, D_{t_0}^{\mathrm{L}\mathrm{P}\mathrm{G}} baseline demands of LPG in base years, D_{t_0}^{\mathrm{N}\mathrm{G}} baseline demands of NG in base years, r_T^{\mathrm{L}\mathrm{P}\mathrm{G}} annual phase-out rate of LPG in year T, where T\in \left(t_0,t\right), r_T^{\mathrm{N}\mathrm{G}} annual phase-out rate of NG in year T, where T\in \left(t_0,t\right), t any future year such that .

4.1.2 Hydrogen demand estimation

Equation (3) estimates the total national hydrogen demand in year t resulting from the displacement of LPG and NG consumption. The differences between \left(D_{t_0}^{\mathrm{L}\mathrm{P}\mathrm{G}}-D_t^{\mathrm{L}\mathrm{P}\mathrm{G}}\right) and \left(D_{t_0}^{\mathrm{N}\mathrm{G}}-D_t^{\mathrm{N}\mathrm{G}}\right) quantify the fossil energy displaced up to time t, and the adoption coefficients, {\alpha }^{\mathrm{L}\mathrm{P}\mathrm{G}} and {\alpha }^{\mathrm{N}\mathrm{G}}, represent the proportion of this displaced demand that is substituted with hydrogen. This formulation assumes that hydrogen is delivered for direct end-use (e.g., household cylinders), and therefore, no efficiency loss is incurred in the substitution, enabling a direct one-to-one energy equivalence.

(3)D_t^{{\mathrm{H}}_2}={\alpha }^{\mathrm{L}\mathrm{P}\mathrm{G}}.\left(D_{t_0}^{\mathrm{L}\mathrm{P}\mathrm{G}}-D_t^{\mathrm{L}\mathrm{P}\mathrm{G}}\right)+{\alpha }^{\mathrm{N}\mathrm{G}}.\left(D_{t_0}^{\mathrm{N}\mathrm{G}}-D_t^{\mathrm{N}\mathrm{G}}\right),

where \alpha {}^{\mathrm{L}\mathrm{P}\mathrm{G}} The fraction of displaced LPG, {\alpha }^{\mathrm{N}\mathrm{G}} represent the fraction of displaced NG, D_t^{{\mathrm{H}}_2} national hydrogen demand in the time period t.

4.1.3 Regional demand allocation

The hydrogen demand for city c in year t, denoted as D_{c,t} in Eq. (4), disaggregates hydrogen demand across cities or regions. This spatial allocation is critical for informing network-level infrastructure decisions, such as locating NDCs, RDCs and LDCs and optimising transport flows in subsequent stages of the supply chain model.

(4)D_{c,t}=D_t^{{\mathrm{H}}_2}.\left(\frac{{Pop}_c}{{\sum _{k\in c}Pop}_i}\right).

4.2 Situation analysis

The situation analysis evaluates both ideal and disruption situations for the HHSC. The ideal situation optimises distribution center locations and maximises profit, while disruption situations assess the impact of unmet demand without mitigation and with mitigation strategies. Mitigation strategies, such as rerouting, additional vehicle hiring, and maintaining safety stock, are introduced to reduce the effects of disruptions as presented in Table 5 AnyLogistix software is used to model and compare financial outcomes, evaluating total profit, penalty costs, and fulfilment rates.

4.2.1 Sets

Sets of potential locations of DCs help determine the number of DCs required over the period to satisfy the household demand for hydrogen. The sets are presented as follows:

4.2.2 Variables

Variables are essential for determining the production quantity of hydrogen at the NDCs and for modeling the flow of hydrogen from the NDCs to the RDCs and from the RDCs to the LDCs. They also specify the transport mode, such as vehicles, and the quantity of hydrogen received at designated distribution centers, including RDCs and LDCs.

4.2.3 Parameters

Parameters are crucial for determining the annual projected demand from RDCs and LDCs and assessing the capacity of NDCs and RDCs. They also play a key role in calculating estimated profit by analyzing revenue, production costs, safety stock maintenance costs, including any safety stock, and transportation costs associated with moving hydrogen from NDCs to RDCs and from RDCs to LDCs. The parameters are presented as follows.

4.2.4 Mathematical model for scenario analysis

The simplified ideal situation function presented in Eq. (5) can be expressed as follows: profit equals total revenue minus production, transportation costs. Total revenue is calculated by multiplying the quantity of hydrogen delivered to the LDCs by the selling price, as indexed in the model parameters. Production cost is similarly determined by multiplying the quantity of hydrogen produced at the NDCs. Transportation costs consist of two components: the cost of transporting hydrogen from NDCs to RDCs and from RDCs to LDCs. These components collectively contribute to the overall financial implications of the HSC. The simplified function for disruption situations without mitigation strategies presented in Eq. (6) can be expressed as follows: profit equals total revenue minus production costs, transportation costs, and penalty costs. This function is like Eq. (5), with the addition of the penalty cost. In disruption situations, the demand from the LDC may or may not be fulfilled. For any unmet demand, a penalty cost is applied to account for the shortfall. The disruption situations with the mitigation function presented in Eq. (7) reflect changes in transportation costs due to the implementation of the rerouting strategy, as well as the additional costs associated with hiring extra vehicles and maintaining safety stock. The mathematical model is presented as follows.

(5)\begin{array}{c}\mathrm{M}\mathrm{a}\mathrm{x}\text{π}=s\sum _{k=1}^K{l}_k-M\sum _{i=1}^I{P}_i-\\ \left(T_1\sum _{i=1}^I\sum _{j=1}^J{X}_{ij}+T_2\sum _{j=1}^J\sum _{k=1}^K{Y}_{jk}\right),\end{array}

(6)\begin{array}{c}\mathrm{M}\mathrm{a}\mathrm{x}\text{π}=s\sum _{k=1}^K{l}_k-M\sum _{i=1}^I{P}_i-(T_1\sum _{i=1}^I\sum _{j=1}^J{X}_{ij}+\\ T_2\sum _{j=1}^J\sum _{k=1}^K{Y}_{jk})-P\left(\sum _{k=1}^K{d}_k-\sum _{k=1}^K{L}_k\right),\end{array}

(7)\begin{array}{c}\mathrm{M}\mathrm{a}\mathrm{x}\text{π}=s\sum _{k=1}^K{l}_k-M\sum _{i=1}^I{P}_i-(T_1\sum _{i=1}^I\sum _{j=1}^J{X}_{ij}+\\ \mathrm{\Delta }{T}_1\sum _{i=1}^I\sum _{j=1}^J{X}_{ij}+T_2\sum _{j=1}^J\sum _{k=1}^K{Y}_{jk}+{\mathrm{\Delta }T}_2\sum _{j=1}^J\sum _{k=1}^K{Y}_{jk})-\\ I\sum _{j=1}^J\left(R_j-\sum _{k=1}^K{Y}_{jk}\right)-P\left(\sum _{k=1}^K{d}_k-\sum _{k=1}^K{L}_k\right)-I\left(F\sum _{k=1}^K{d}_k\right).\end{array}

4.2.5 Constraints

Equation (8) represents that the quantity of hydrogen produced at NDC i must be less than or equal to the capacity of NDC i. Equation (9) represents that the quantity of hydrogen received by RDC j must equal the quantity of hydrogen transported from NDC i to RDC j. Similarly, Eq. (10) represents that the quantity of hydrogen received by LDC k must equal the quantity of hydrogen transported from RDC j to LDC k. Equation (11) represents that the quantity of hydrogen received at LDC k must be equal to the demand of LDC k. Equation (12) represents that the production at NDC i must be greater than or equal to the quantity of hydrogen getting transported from NDC i to RDC j. Equation (13) represents that production at NDC i, the quantity of hydrogen getting transported from NDC i to RDC j and the quantity of hydrogen getting transported from RDC j to LDC k must be greater than or equal to 0. The additional constraint presented in Eq. (14) represents that if the quantity of hydrogen received at RDC j is less than the demand of LDC k, it will incur an additional penalty cost for the unmet demand. Equation (15) presents that the quantity of hydrogen delivered to LDC k may or may not be equal to the demand of LDC k. The constraint presented in Eq. (16) states that the proportion of safety stock that needs to be stored in case of mitigating route or vehicle disruptions will be a positive value. Similarly, Eq. (17) represents that the quantity of hydrogen received at RDC j must be greater than or equal to the quantity of hydrogen transported from NDC i to RDC j. The objective functions are subjected to the following constraints.

(8)P_i\le C_i;\mathrm{\forall }i,

(9)R_j=\sum _{i=1}^I{X}_{ij};\mathrm{\forall }j,

(10)L_k=\sum _{j=1}^J{Y}_{jk};\mathrm{\forall }k,

(11)L_k=d_k;\mathrm{\forall }k,

(12)P_i\ge \sum _{j=1}^J{X}_{ij};\mathrm{\forall }i,

(13)R_j\ge \sum _{k=1}^K{Y}_{jk};\mathrm{\forall }j,

(14)P\left(\sum _{k=1}^K{d}_k-\sum _{k=1}^K{L}_k\right)\ge 0;\mathrm{\forall }k,

(15)L_k\le d_k;\mathrm{\forall }k,

(16)F\sum _{k=1}^K{d}_k\ge 0,

(17)P_i,X_{ij},Y_{jk}\ge 0;\mathrm{\forall }i,\mathrm{\forall }j,\mathrm{\forall }k.

5 Results and discussions

The results and discussion demonstrate that transportation disruptions significantly impact the profitability and operational efficiency of the HHSC, with severe financial losses observed in disruption scenarios without mitigation strategies. Applying mitigation strategies, such as rerouting, hiring additional vehicles, and maintaining safety stock, helps reduce these negative effects, improving demand fulfilment and profitability. Overall, the combined mitigation approach proves crucial in enhancing the resilience of the supply chain, particularly in worst-case disruption scenarios.

5.1 Estimating the demand of hydrogen for households

The proposed long-term model for the Australian HHSC outlines a phased transition from traditional gas fuels, LPG and NG, toward a fully hydrogen-powered household energy system. This transition is aligned with national decarbonisation goals and spans from 2026 to 2090. The model targets the complete phase-out of LPG by 2045, a substantial reduction in NG use by 2050, and full NG elimination by 2080. Hydrogen is introduced gradually, aiming to replace LPG entirely by 2050 and NG by 2080, with widespread use across cooking, heating, and water systems.

The model incorporates annual projections assuming a 5% reduction in NG and LPG consumption and a corresponding 10% increase in hydrogen adoption. While these forecasts are grounded in national energy policy directions, their realization depends on several dynamic factors, including infrastructure deployment, technological maturity, public perception, and economic viability. The transition also demands evaluation of hydrogen production and logistics systems, especially concerning the environmental footprint of liquid hydrogen supply, which is beyond this paper’s scope.

Spatial analysis of household hydrogen demand reveals considerable regional variation. Urban centers such as Melbourne, Sydney, and Brisbane show higher energy demand densities, identifying them as critical nodes for LDCs. A national hydrogen demand map, supported by this model, helps guide infrastructure planning for optimal placement of NDCs, RDCs, and LDCs.

While Fig. 3 provides a visual trajectory of this fuel transition, the model’s assumptions must be interpreted with caution, particularly under changing policy, behavioral, or economic conditions. Achieving the projected hydrogen transition will require robust scenario modeling, policy coherence, and adaptive planning to address uncertainties across technical, regulatory, and societal domains.

5.2 Ideal situations

The ideal situations focus on several key aspects to ensure an effective transition to an HHSC in Australia. Firstly, the foundation phase initiates the HHSC rollout, targeting high-demand urban zones to maximise early returns. In 2026, hydrogen is supplied from a single NDC in Portland to one RDC in Melbourne and 26 LDCs across VIC and NSW. By 2030, a second RDC in Perth is added, and LDCs expand to 38. The Portland NDC supports national supply until 2045 due to its strategic location and capacity. By 2045, the network includes four RDCs, adding Brisbane and 47 LDCs, marking the transition from pilot deployment to broader operational coverage.

Secondly, the expansion phase shifts from a concentrated south-east corridor to national HHSC coverage. While the 2045 network relied on one NDC in Portland, a second NDC in Bowen is added by 2055, decentralising supply and reducing system fragility. By 2065, the HHSC comprises two NDCs, six RDCs, and 67 LDCs, extending access to Townsville, Adelaide, and Darwin. This decentralised growth supports rising demand, improves resilience, and enables regional equity. The increase from 47 to 67 LDCs strengthens NG and LPG phase-out efforts, aligning with national hydrogen strategies and long-term sustainability goals.

Finally, the maturation phase marks the HHSC’s evolution into a fully decentralised, high-resilience logistics network. By 2070, a third NDC is added in Port Hedland, joining Portland and Bowen to balance supply across east, north, and west corridors. By 2090, the network includes 3 NDCs, 8 RDCs in major cities, and 87 LDCs nationwide as presented in Appendix 4. The structure becomes multi-directional, enabling efficient last-mile delivery and reducing regional dependency.

Under ideal situations, the HHSC expands steadily from 2026 to 2090, as outlined in Appendices 4,5,6,7. The infrastructure grows from 1 NDC, 1 RDC, and 26 LDCs in 2026 to 3 NDCs, 8 RDCs, and 87 LDCs by 2090. This progression supports increasing household hydrogen demand across the country. The number of routes and vehicles increases accordingly, from 27 in 2026 to 95 in 2090, ensuring timely and efficient delivery. Appendices 2 and 3 provide the precise coordinates of RDC and NDC locations, enabling optimal network design to support full national coverage and service reliability.

5.2.1 Financial implications under ideal situations

The Australian HHSC’s Financial implication reflects phased infrastructure rollout and growing hydrogen demand as presented in Fig. 4. Profit rises from $479 million in 2026 to $88.26 billion by 2090 under a 60% mark-up pricing strategy and 5% annual cost increases. During the foundation phase (2026–2045), limited infrastructure yields modest profit. The expansion phase (2045–2070) introduces additional NDCs, RDCs, and LDCs, driving revenue growth to $26.6 billion by 2070. In the maturation phase (2070–2090), national network saturation and operational efficiency maximise profitability. This trajectory highlights the value of phased planning and cost calibration in sustainable hydrogen infrastructure development.

5.2.2 Sensitivity analysis under an ideal situation

The sensitivity analysis presented in Fig. 5 shows how changes in transportation costs affect the profitability of the Australian household hydrogen supply chain between 2026 and 2090. A 20 percent increase in transportation costs results in a reduced profit of around USD 68 billion by 2090, whereas a 20 percent decrease boosts profit to over USD 108 billion. Although the influence of transportation costs is less dramatic than that of production costs, early years such as 2040 and 2050 still experience reduced profitability under cost increases. From 2065 onwards, the gap between the increase and decrease scenarios becomes more pronounced, reflecting the combined financial impact over time. The findings suggest that controlling transportation expenses is critical, particularly during the network’s expansion phase, to safeguard long-term profitability. Effective logistical planning and cost efficiency are therefore essential to support the sustainable financial performance of the household hydrogen supply chain in Australia.

Similarly, the sensitivity analysis in Fig. 6 illustrates the impact of production cost fluctuations on the profitability of the Australian household hydrogen supply chain from 2026 to 2090. A reduction in production costs leads to a substantial increase in profit, with a 20% decrease resulting in profit exceeding USD 150 billion by 2090. In contrast, a 20% increase in costs significantly suppresses profitability, limiting returns to around USD 41 billion in the same year. During the earlier years, particularly between 2040 and 2050, even moderate cost increases result in low or negative profit, reflecting the financial vulnerability of the supply chain during its early stages. As the network expands and hydrogen adoption grows, the effect of cost variations becomes more pronounced. The analysis highlights that maintaining efficient production costs is essential for sustaining profitability, especially during the transition period when infrastructure investments and demand shifts are most sensitive to cost fluctuations.

5.3 Disruption situations without mitigation strategies

The Australian HHSC, projected from 2026 to 2090, faces several critical disruption situations that could compromise its efficiency, reliability, and profitability. These disruptions highlight significant research gaps in the existing literature, particularly regarding hydrogen logistics. Traditional supply chain studies often overlook the unique challenges posed by hydrogen transport, such as vehicle, route and combined disruptions.

5.3.1 Route disruption scenarios

Scenario R11 involves the disruption of a single NDC–RDC route for seven days without mitigation strategies, as shown in Fig. 7. In 2035, this leads to a sharp drop in demand fulfilment to 13.15%, resulting in a penalty cost of USD 456 million and a loss of USD 467 million. From 2045 onward, HHSC performance improves, demand fulfilment exceeds 82%, peaking at 94.54% in 2055, accompanied by a profit of USD 234 million. By 2085, despite a penalty cost of USD 355 million, profit reaches USD 1.37 billion, supported by higher demand and revenue.

However, in scenario R22, where three to five RDC–LDC routes are disrupted for four days without mitigation strategies, the impact on demand fulfilment is much less severe. Demand fulfilment remains consistently high, ranging from 91.40% in 2030 to 94.04% in 2085, despite an increase in the number of disrupted routes. Penalty costs, however, increase steadily, from USD 20.02 million to USD 143.38 million, while profit grows substantially from USD 13.92 million to USD 1.56 billion. This indicates that last-mile disruptions affect demand fulfilment only modestly but have a significant effect on penalty costs and profit variability. Moreover, scenario R32, which involves disruptions to both NDC–RDC and RDC–LDC routes for three to seven days without mitigation strategies, shows the most persistent pressure on HHSC performance. Demand fulfilment is lower and more variable, starting at 73% in 2030, rising to 86% in 2065, and declining to 74% by 2085. Meanwhile, penalty costs increase dramatically, from USD 5.50 million to USD 1.13 billion, and profit ranges from USD 3.82 million to USD 428.39 million. This demonstrates that disruption type, duration, and number of affected routes collectively shape the operational and financial outcomes of the HHSC without mitigation strategies.

5.3.2 Vehicle disruption scenarios

In scenario V11, each disruption affects a single NDC–RDC vehicle for four days, with no mitigation strategies applied. In 2035, demand fulfilment drops sharply to 13%, resulting in a penalty cost of USD 453.38 million and a loss of USD 464.23 million. By 2045, demand fulfilment improves to 83%, while penalty cost declines to USD 56.47 million and profit reaches USD 57.04 million. In 2055, the highest demand fulfilment of 94% aligns with a reduced penalty cost of USD 40.68 million and a profit of USD 232.88 million. In 2065 and 2075, demand fulfilment remains between 86% and 89%, but penalty costs increase to USD 133.04 million and USD 414.37 million, respectively, while profit rises to USD 498.08 million and USD 551.70 million. By 2085, demand fulfilment stays at 89%, penalty cost is USD 353.01 million, and profit peaks at USD 1.37 billion. Despite improving demand fulfilment, the absence of mitigation strategies keeps penalty costs high across the years.

However, in scenario V22, where three to five RDC–LDC vehicles are disrupted for four days without mitigation strategies, demand fulfilment remains high, ranging from 91% in 2030 to 94% in 2085. Nonetheless, penalty costs increase from USD 19.59 million to USD 140.30 million, while profit grows significantly from USD 13.62 million to USD 1.53 billion. This shows that last-mile vehicle disruptions have a modest impact on demand fulfilment but can strongly affect financial outcomes.

Moreover, in scenario V32, disruptions to both NDC–RDC and RDC–LDC vehicles for three to seven days without mitigation strategies result in lower demand fulfilment, starting at 70% in 2030 and ending at 71% in 2085. Over this period, penalty costs escalate from USD 5.32 million to USD 1.09 billion, while profit grows from USD 3.70 million to USD 414.12 million.

5.3.3 Combined disruption scenarios

In scenario C32, each disruption affects one NDC–RDC route, two RDC–LDC routes, and both one NDC–RDC and three RDC–LDC vehicles for seven days, with no mitigation measures in place. In 2035, demand fulfilment is critically low at 7.56%, causing penalties of USD 715.74 million and a loss of USD 6.22 million. By 2045, fulfilment improves to 47.71%, penalties reduce to USD 89.15 million, and profit reaches USD 65.06 million. In 2055, fulfilment rises to 54.36%, with penalties of USD 64.22 million and profit of USD 156.79 million. In 2065, fulfilment is 49.77%, penalties climb to USD 210.03 million, and profit is USD 361.73 million. In 2075, fulfilment is 51.28%, penalties surge to USD 654.15 million, with profit at USD 553.70 million. By 2085, fulfilment is 51.34%, penalties remain high at USD 557.29 million, and profit peaks at USD 985.91 million.

The impact could be even more severe if such disruptions targeted high-demand, high-dependency routes or occurred simultaneously with other network failures, potentially leading to widespread service collapse, exponential penalty growth, and significant loss of customer trust, making recovery far slower and costlier.

5.3.4 Sensitivity analysis under a disruption situation without mitigation strategies

The sensitivity analysis of transportation cost variations under the R11 scenario without mitigation highlights the vulnerability of the early years and the benefits of cost reductions in later periods. As illustrated in Fig. 8, the year 2035 records substantial losses under all conditions, ranging from a loss of about USD 470 million with a 20% cost increase to a loss of USD 461 million even with a 20% cost decrease, showing the limited effect of cost adjustments during severe disruption. By 2045, profitability begins to recover, reaching around USD 43 million under a 10% cost increase, USD 29 million under a 15% increase, and between USD 71 million and USD 98 million under 5%–15% cost decreases. From 2055 onwards, profit expands steadily, with cost reductions providing clear advantages. For example, in 2085, with mitigation, a 10% decrease generates USD 1.54 billion, and a 15% decrease yields USD 1.62 billion. Without mitigation, however, the corresponding 10% and 15% increases result in only USD 1.29 billion and USD 1.20 billion, respectively.

On the other hand, the sensitivity analysis of production cost variations under the V22 scenario without mitigation shows that profitability is highly responsive to cost changes over time. As illustrated in Fig. 9, the early years remain unstable, with 2035 recording a loss of about USD 6.44 million under a 20% increase in production cost and only marginally positive outcomes under smaller increases, compared with stronger gains of USD 28–54 million under cost reductions of 5%–20%. By 2045, the divergence widens: a 20% increase results in a loss of nearly USD 68 million, while a 20% decrease yields over USD 207 million profit. From 2055 onwards, the gap between increases and decreases becomes more pronounced. In 2085, a 20% increase in production cost limits profit to USD 714 million, while a 20% decrease generates almost USD 2.76 billion. Overall, the results confirm that reductions in production costs substantially improve profitability, while increases create significant financial risks in the absence of mitigation.

Moreover, the analysis of production cost sensitivity under the C32 scenario without mitigation illustrates significant financial vulnerability, especially in the early years. As shown in Fig. 10, in 2035, profit remains heavily negative under all conditions, ranging from a loss of USD 724 million with a 20% cost increase to a loss of USD 719 million even with a 20% decrease. This indicates that severe combined disruptions override any benefit from cost adjustments at the start of the planning horizon. By 2045, outcomes begin to improve, with a 20% increase still producing a loss of around USD 88 million, while a 20% decrease generates a modest profit of USD 61 million. In subsequent decades, profitability becomes highly sensitive to production costs. For example, in 2055, a 20% increase results in a loss of nearly USD 18 million, whereas a 20% decrease generates more than USD 240 million. By 2085, the divergence widens sharply, with profit ranging from a loss of USD 59 million to a gain of USD 1.07 billion. Overall, the results show that production cost reductions substantially improve profitability, while increases severely undermine financial stability when mitigation is absent.

5.4 Disruption situations with mitigation strategies

The disruption situations with mitigation strategies are outlined in three sections: route disruption scenarios, vehicle disruption scenarios, and combined disruption scenarios.

5.4.1 Route disruptions

The application of mitigation strategies in scenarios R11, R22, and R32 brings clear improvements in demand fulfilment, penalty cost reduction, and profitability across the HHSC network. As shown in Fig. 11, in scenario R11, the impact is most evident in the early years. For example, in 2035, without mitigation strategies, demand fulfilment is only 13.15%, penalty cost is USD 456 million, and the HHSC records a loss of USD 467 million. With mitigation strategies, demand fulfilment rises above 97%, penalty cost drops to USD 6.3 million, and profit reaches USD 91.2 million. By 2085, profit increases from USD 1.37 billion (without mitigation) to USD 1.63 billion (with mitigation), as penalty cost falls from USD 355 million to USD 93.4 million.

In scenario R22, the benefits of mitigation are also clear. For instance, in 2030, without mitigation strategies, demand fulfilment is 91%, penalty cost is USD 20.02 million, and profit is USD 13.62 million. With mitigation strategies, fulfilment improves to 97.40%, penalty cost halves to USD 10.01 million, and profit increases to USD 19.99 million. By 2085, mitigation raises profit from USD 1.56 billion to USD 1.63 billion.

In Scenario R32, applying mitigation strategies such as rerouting, vehicle hiring, and safety stock significantly improves HHSC performance. Without mitigation, demand fulfilment ranges from 73% to 86%, dropping to 74% in 2085, with penalties peaking at USD 1.13 billion and profit limited to USD 428 million. With mitigation, demand fulfilment improves to 97%–98% across all years. Penalty costs drop sharply, for example, from USD 1.13 billion to USD 112 million in 2085. Profit also rises substantially, reaching USD 1.60 billion in 2085 compared to USD 428 million without mitigation.

5.4.2 Vehicle disruptions

In scenario V11, the application of mitigation strategies, such as additional vehicle hiring and maintaining safety stock, significantly improves demand fulfilment, reduces penalty costs, and enhances profitability. For instance, in 2035, without mitigation strategies, demand fulfilment is just 13%, penalty cost reaches USD 453.38 million, and the HHSC records a loss of USD 464.23 million. With mitigation strategies, demand fulfilment exceeds 96%, penalty cost drops to USD 6.27 million, and profit turns positive at USD 17.83 million. For example, in 2085, penalty cost is reduced from USD 353.01 million to USD 92.88 million, and profit reaches USD 1.22 billion, compared to USD 1.37 billion without mitigation.

However, in scenario V22, where demand fulfilment is already relatively high, mitigation still delivers notable improvements, as shown in Fig. 12. For example, in 2030, penalty cost falls from USD 19.59 million to USD 9.80 million, and profit increases from USD 13.62 million to USD 19.57 million. In 2085, penalty cost reduces from USD 140.30 million to USD 86.08 million, and profit rises from USD 1.53 billion to USD 1.59 billion.

Moreover, in scenario V32, mitigation has the most significant impact under complex disruptions, as shown in Fig. 13. For example, in 2085, without mitigation strategies, demand fulfilment is 71%, penalty cost is USD 1.09 billion, and profit is USD 428 million. With mitigation strategies, demand fulfilment rises to 94%, penalties fall to USD 108.71 million, and profit increases to USD 1.54 billion.

5.4.3 Combined disruptions

Scenario C32 models a severe and prolonged disruption affecting both NDC-RDC and RDC-LDC links for a duration of seven days. The disruption impacts multiple routes and vehicle flows simultaneously, causing significant service delays, higher penalty costs, and losses in profitability if left unaddressed. Without mitigation, operational and financial performance declines sharply, particularly in the early years. In 2035, demand fulfilment is just 7.56%, penalty costs reach USD 715.74 million, and losses amount to USD 721.96 million. Although fulfilment improves slightly in later years, such as 51.34% in 2085, penalties remain high at USD 557.29 million, constraining profit to USD 428.62 million as presented in Fig. 14. With mitigation measures in place, including rerouting to circumvent blocked routes, additional vehicle hiring to recover lost capacity, and maintaining safety stock at RDCs to meet demand despite disruption, performance improves substantially. Demand fulfilment rises to between 92.16% and 93.82%, and penalties drop significantly, from USD 715.74 million to USD 6.62 million in 2035, and from USD 557.29 million to USD 97.52 million in 2085. These changes turn heavy losses into a USD 10.78 million profit in 2035 and boost profit to USD 1.14 billion in 2085. Overall, the introduction of mitigation strategies in Scenario C32 ensures much higher service reliability, major reductions in penalty costs, and significant gains in profitability, with the most notable improvements during years of severe disruption.

5.4.4 Sensitivity analysis under disruption situations with mitigation strategies

The sensitivity analysis of production cost variations under the V22 scenario with mitigation strategies demonstrates the strong influence of cost fluctuations on profitability. As shown in Fig. 15, profit trajectories increase steadily over time under all cost assumptions, yet the magnitude of growth differs significantly. A 20% cost decrease delivers the highest profitability, exceeding USD 2.7 billion by 2085, followed closely by 15% and 10% decreases. In contrast, cost increases dampen financial outcomes, with a 20% increase reducing profit to approximately USD 0.7 billion in 2085, compared to USD 1.6 billion under a 5% increase. Early periods, particularly 2035 and 2045, reveal substantial vulnerability, where even moderate increases in costs generate losses. However, the implementation of mitigation strategies stabilizes long-term performance, ensuring that all trajectories move into positive profit territory beyond 2055. Overall, the analysis indicates that while production cost reductions enhance resilience, even cost increases remain manageable with effective mitigation measures.

However, the analysis of transportation cost sensitivity under the C32 scenario with mitigation strategies highlights the strong influence of cost variations on profitability. As shown in Fig. 16, profit increases steadily across all cost assumptions, yet the rate of growth differs substantially. A 20% cost decrease produces the highest profitability, surpassing USD 1.5 billion by 2085, followed by 15% and 10% decreases. In contrast, increases in transportation costs reduce margins considerably, with a 20% increase yielding approximately USD 0.87 billion in 2085, compared to over USD 1.1 billion under a 5% increase. The early years, particularly 2035 and 2045, are highly sensitive, where even moderate cost increases result in negative profitability. However, mitigation strategies stabilize long-term outcomes, ensuring that all trajectories move into positive territory beyond 2055. Overall, the findings confirm that while decreasing transportation costs significantly enhances profitability, effective mitigation enables the system to remain resilient even under cost escalation.

Moreover, the results in Fig. 17 demonstrate that maintaining safety stock at RDCs reduces penalty costs but significantly increases safety stock holding costs. Penalty costs decline steadily with higher stock levels, falling from about USD 7.06 million in 2035 to USD 6.53 million with a 15% increase, and from USD 104 million in 2085 to around USD 96 million. This indicates that safety stock provides an effective buffer against disruption by improving demand fulfilment and lowering service failure penalties. However, the trade-off is the sharp rise in holding costs, which grow from USD 72 million to over USD 80 million in 2035, and from USD 390 million to more than USD 427 million in 2085. The analysis, therefore, suggests that the optimal level of safety stock lies between 5% and 10%, where penalty costs are meaningfully reduced without excessive holding expenses. Beyond this level, the additional cost outweighs the marginal savings.

5.5 Comparison between disruption situations with and without mitigation strategies

The comparison between disruption situations with and without mitigation strategies highlights the critical importance of proactive measures in sustaining supply chain resilience. The R11 scenario, representing route disruptions, shows the greatest volatility as presented in Table 6. Without mitigation, demand fulfilment falls to only 13% in 2035, resulting in severe financial losses. When rerouting and additional vehicles are introduced, fulfilment improves to nearly 99% and profitability is consistently restored. These findings are consistent with studies emphasizing the effectiveness of rerouting in minimising transport-related risks (Ivanov et al., 2017). The V22 scenario, which reflects vehicle disruptions, maintains comparatively stable performance with fulfilment levels above 90% even without intervention. Nevertheless, applying vehicle hiring and safety stock further reduces penalty costs and strengthens financial outcomes, aligning with research on flexible capacity strategies in uncertain conditions (Fazli-Khalaf et al., 2020). The V32 scenario, which indicates more volatile vehicle disruptions, records unstable fulfilment levels and negative profit in certain years when unmitigated. With targeted mitigation, however, fulfilment rises to between 93% and 97%, stabilizing performance and securing financial resilience. The C32 scenario, combining both route and vehicle disruptions, proves the most critical. Here, fulfilment drops to as low as 8% in 2035 and severe financial losses occur. Integrated strategies, including rerouting, additional vehicles, and safety stock, significantly improve fulfilment above 90% and restore profitability, reinforcing earlier findings on multi-pronged resilience planning (Almansoori and Shah, 2012). Overall, the analysis confirms that combined mitigation is indispensable for long-term stability and sustainability.

6 Practical implications

The ideal situation of Australian HHSC develops in three phases, such as foundation, expansion and maturation. Each phase introduces structural changes in the network. Disruptions at the NDC, RDC, and LDC levels can significantly affect demand fulfilment, profitability and long-term sustainability. The study shows that targeted mitigation strategies can help managers and policy makers maintain high operational continuity while advancing Sustainable Development Goals (SDGs) 7 and 9, which focus on clean energy access and resilient infrastructure (Allen et al., 2018; Vaidya and Chatterji, 2020).

In the foundation phase (2026–2045), the HHSC begins with one NDC in Portland supplying one RDC in Melbourne and 26 LDCs, increasing to two RDCs and 38 LDCs by 2035, and four RDCs and 47 LDCs by 2045. Routes increase from 27 in 2026 to 51 in 2045, with vehicles increasing in parallel. The concentration of supply in the south-east corridor makes the network highly vulnerable to single-point failures. In the R11 case, where an NDC–RDC route is disrupted, demand fulfilment drops to 13% in 2035 without mitigation, leading to high penalties and financial losses. Applying rerouting and additional vehicle hiring increases demand fulfilment to above 97% and reduces penalty costs by over 90%. This underlines the need for redundancy and rapid response in the early phase, allowing managers to protect market entry operations and policy makers to embed resilience standards into rollout plans. Similar early-phase vulnerability patterns have been identified in urban fuel supply networks (Ahmadi-Javid and Seddighi, 2013).

The expansion phase (2045–2070) marks a shift to national coverage. A second NDC in Bowen decentralises supply, the number of RDCs rises to six, and LDCs expand to 67. Routes grow from 51 in 2045 to 78 in 2070, with vehicles increasing proportionally. While decentralisation reduces single-point failure risk, RDC–LDC disruptions, such as in R22 and V22, still lower demand fulfilment to 91%–94% without mitigation. Implementing safety stock at RDCs, along with rerouting and hiring additional vehicles, lifts demand fulfilment to 97%–99% and sharply cuts penalties even during demand peaks. Managers can use these findings to refine safety stock policies and optimise transport contracts, while policy makers can invest in equitable regional coverage to reduce geographic disparities. The effectiveness of combined mitigation measures in strengthening demand fulfilment is supported by researchers (Akbari et al., 2022).

The maturation phase (2070–2090) completes the transition to a fully decentralised HHSC. By 2090, there are three NDCs in Portland, Bowen and Port Hedland, eight RDCs in major cities, and 87 LDCs nationwide. Routes increase from 78 in 2070 to 95 in 2090, matched by the number of vehicles. This multi-directional structure increases flexibility but also creates exposure to complex combined disruptions. In C32, where both routes and vehicles are disrupted, demand fulfilment can fall to 51% without mitigation. Maintaining safety stock, rerouting and hiring additional vehicles maintains demand fulfilment above 92% and reduces penalties by over 80%.

Managers would benefit from clear decision-support frameworks that specify which mitigation strategies, such as rerouting, maintaining safety stock, or additional vehicle hiring, are most effective under different disruption scenarios. These insights allow for proactive planning, resource allocation, and service continuity. Moreover, policymakers could use these results to justify long-term investments in HHSC infrastructure, including decentralised storage, route redundancy, and adaptive HHSC capabilities.

Australia presents unique geographical challenges that make such planning especially critical. The country’s vast transport distances, dispersed population centers, and uneven infrastructure distribution increase vulnerability to both isolated failures and cascading disruptions. These logistical constraints render HHSC more susceptible to systemic risk than hydrogen networks in denser or more urbanised contexts, where redundancies and alternative pathways are more readily available. The cascading failure risks observed in this study, particularly in the maturation phase where decentralisation reaches its peak, are consistent with prior findings in the resilience literature. Similar concerns have been identified in decentralised energy and supply networks, where localized disruptions can trigger widespread impacts across interdependent nodes and routes (Eskandari et al., 2024; Robles et al., 2020). This study extends existing resilience research by offering context-specific planning guidance for Australia’s spatially dispersed and decentralising HHSC, building upon earlier contributions in this domain (Almaraz et al., 2022; Dayhim et al., 2014). By modeling disruption impacts across network maturity phases, our study findings seem to reveal how targeted mitigation strategies, such as rerouting, safety stock, and additional vehicle hiring, can minimise operational losses and penalty costs (Wang et al., 2024). Without mitigation, even minor disruptions can escalate into widespread service failures, particularly in regions with long transport distances and limited redundancy (Alizadeh et al., 2022). Therefore, our study corroborates earlier literature that has discussed how ensuring a reliable HHSC is essential not only for maintaining operations but also for securing public trust in hydrogen as a household energy source (Hancock and Wollersheim, 2021). By offering an innovative perspective on this field, our study findings highlight that the alignment of mitigation with the network’s development phase would support a cost-effective and resilient planning, in turn enabling high demand fulfilment and profitability, ultimately contributing to Australia’s hydrogen transition goals.

7 Conclusions and future research opportunities

This paper investigates vehicle, route, and combined disruptions in the Australian HHSC from 2026 to 2090, using multi-period network optimisation models, scenario analysis, and targeted mitigation strategies. It assesses how these disruptions affect operational efficiency, profitability, and the ability to meet household hydrogen demand. By incorporating rerouting, additional vehicle hiring, and maintaining safety stock, the study highlights how proactive strategies significantly reduce disruption impacts and improve supply chain resilience. A key research gap addressed is the lack of comprehensive studies that examine combined disruptions in HSC. Existing models often treat vehicle disruptions and route disruptions in isolation, overlooking their combination.

Results show that combining rerouting, additional vehicle hiring, and maintaining safety stock offers substantial improvements in demand fulfilment, cost control, and overall resilience, even under severe disruption scenarios. The findings underscore the necessity of robust, multi-layered contingency plans to effectively manage vehicle, route, and combined disruptions Without such preparedness, even minor disruptions can cascade through the supply chain, leading to unmet demand, increased operational costs, and diminished profitability. Ensuring a reliable and resilient HSC is not only critical for maintaining consumer confidence but also for building public trust in hydrogen as a dependable household energy source. This is particularly important in the context of Australia’s energy transition, where hydrogen is positioned as a key low-carbon alternative, and its successful integration into everyday household use relies heavily on the perceived reliability and affordability of its delivery network.

The limitation of this study is its reliance on static models that assume transportation costs, vehicle availability, and disruption frequencies. These models may not fully capture the dynamic nature of real-world disruptions, such as fluctuating fuel prices, changes in regulatory frameworks, or unpredictable weather patterns that can affect route availability. As a result, the accuracy and practical applicability of the findings may be limited when disruptions occur under variable and evolving conditions.

Future research should investigate the role of infrastructure investments in mitigating vehicle and route disruptions, which is another promising research area. Future studies could assess the impact of building redundant transport networks, such as additional hydrogen refuelling stations and alternative transportation modes (e.g., rail, pipelines), to minimise the effects of route disruptions or vehicle disruptions. Research could focus on the cost-benefit analysis of infrastructure development to create a more resilient HSC network that is less vulnerable to combined disruptions.

8 Appendix

References

[1] Acar CDincer I (2018). Hydrogen energy. In: I. Dincer (Ed.), Comprehensive energy systems, 1: 568–605

[2] Acar C, Dincer I (2018a). Hydrogen energy. Comprehensive Energy Systems, 1: 568–605. Available at the website of books.google.com.au

[3] Ahmadi-Javid A, Seddighi A H, (2023). A location-routing problem with disruption risk. Transportation Research Part E, Logistics and Transportation Review, 53: 63–82

[4] Ahmed F, Huynh N, Ferrell W, Badyal V, Padmanabhan B, (2024). Vehicle re-routing under disruption in cross-dock network with time constraints. Expert Systems with Applications, 237: 121517

[5] Akbari F, Valizadeh J, Hafezalkotob A, (2022). Robust cooperative planning of relief logistics operations under demand uncertainty: A case study on a possible earthquake in Tehran. International Journal of Systems Science: Operations & Logistics, 9( 3): 405–428

[6] Aldrighetti R, Battini D, Ivanov D, Zennaro I, (2021). Costs of resilience and disruptions in supply chain network design models: A review and future research directions. International Journal of Production Economics, 235: 108103

[7] Alizadeh T, Clements R, Legacy C, Searle G, Kamruzzaman M, (2022). Infrastructure governance in times of crises: A research agenda for Australian cities. Urban Policy and Research, 40( 1): 1–14

[8] Allen C, Metternicht G, Wiedmann T, (2018). Initial progress in implementing the Sustainable Development Goals (SDGs): A review of evidence from countries. Sustainability Science, 13( 5): 1453–1467

[9] Ally J, Pryor T, Pigneri A, (2015). The role of hydrogen in Australia’s transport energy mix. International Journal of Hydrogen Energy, 40( 13): 4426–4441

[10] Almansoori A, Shah N, (2012). Design and operation of a stochastic hydrogen supply chain network under demand uncertainty. International Journal of Hydrogen Energy, 37( 5): 3965–3977

[11] Almaraz S D L, Azzaro-Pantel C, Montastruc L, (2015). Boix M. Deployment of a hydrogen supply chain by multi-objective/multi-period optimisation at regional and national scales. Chemical Engineering Research & Design, 104: 11–31

[12] Almaraz S D L, Rácz V, Azzaro-Pantel C, Szántó Z O, (2022). Multiobjective and social cost-benefit optimisation for a sustainable hydrogen supply chain: Application to Hungary. Applied Energy, 325: 119882

[13] Azadnia A H, McDaid C, Andwari A M, Hosseini S E, (2023). Green hydrogen supply chain risk analysis: A european hard-to-abate sectors perspective. Renewable & Sustainable Energy Reviews, 182: 113371

[14] Ball M, Wietschel M, Rentz O, (2007). Integration of a hydrogen economy into the German energy system: an optimising modelling approach. International Journal of Hydrogen Energy, 32( 10–11): 1355–1368

[15] Bayram V, (2016). Optimization models for large scale network evacuation planning and management: A literature review. Surveys in Operations Research and Management Science, 21( 2): 63–84

[16] Browning T, Kumar M, Sanders N, Sodhi M S, Thürer M, Tortorella G L, (2023). From supply chain risk to system-wide disruptions: research opportunities in forecasting, risk management and product design. International Journal of Operations & Production Management, 43( 12): 1841–1858

[17] Chung S H, Tse Y K, Choi T M, (2015). Managing disruption risk in express logistics via proactive planning. Industrial Management & Data Systems,, 115( 8): 1481–1509

[18] Çimen M, Benli D, İbiş Bozyel M, Soysal M, (2024). A review on sustainability, Industry 4.0 and collaboration implications in vehicle allocation operations. International Journal of Logistics Management, 35( 3): 943–978

[19] Cox E (2021). Resilience of the future energy system: Impacts of energy disruptions on society (UKERC Working Paper). UK Energy Research Centre. Available at the website of ukerc.rl.ac.uk

[20] Datta P P (2007). A complex system, agent-based model for studying and improving the resilience of production and distribution networks (Doctoral dissertation, Cranfield University). Cranfield University Repository. Available at the website of dspace.lib.cranfield.ac.uk

[21] Dawood F, Shafiullah G, Anda M, (2020). A hover view over Australia’s Hydrogen Industry in recent history: the necessity for a hydrogen industry knowledge-sharing platform. International Journal of Hydrogen Energy, 45( 58): 32916–32939

[22] Dayhim M, Jafari M A, Mazurek M, (2014). Planning sustainable hydrogen supply chain infrastructure with uncertain demand. International Journal of Hydrogen Energy, 39( 13): 6789–6801

[23] Edwards R L, Font-Palma C, Howe J, (2021). The status of hydrogen technologies in the UK: A multi-disciplinary review. Sustainable Energy Technologies and Assessments, 43: 100901

[24] Emenike S N, Falcone G, (2020). A review on energy supply chain resilience through optimization. Renewable & Sustainable Energy Reviews, 134: 110088

[25] Eskandari M, Gilani H, Sahebi H, Sharifi S, (2024). Design and planning of global sustainable bio-hydrogen supply chain with uncertainty: A transportation-oriented robust model. Chemical Engineering Science, 283: 119365

[26] Esmizadeh Y, Mellat Parast M, (2021). Logistics and supply chain network designs: incorporating competitive priorities and disruption risk management perspectives. International Journal of Logistics,, 24( 2): 174–197

[27] Fazli-Khalaf M, Naderi B, Mohammadi M, Pishvaee M S, (2020). Design of a sustainable and reliable hydrogen supply chain network under mixed uncertainties: A case study. International Journal of Hydrogen Energy, 45( 59): 34503–34531

[28] Fitt H, (2022). Boring and inadequate? A literature review considering the use of electric vehicles in drive tourism. Current Issues in Tourism, 25( 12): 1920–1946

[29] Foorginezhad S, Mohseni-Dargah M, Falahati Z, Abbassi R, Razmjou A, Asadnia M, (2021). Sensing advancement towards safety assessment of hydrogen fuel cell vehicles. Journal of Power Sources, 489: 229450

[30] Golan M S, Jernegan L H, Linkov I, (2020). Trends and applications of resilience analytics in supply chain modeling: systematic literature review in the context of the COVID-19 pandemic. Environment Systems & Decisions, 40( 2): 222–243

[31] Gorji S A, (2023). Challenges and opportunities in green hydrogen supply chain through metaheuristic optimization. Journal of Computational Design and Engineering, 10( 3): 1143–1157

[32] Hancock L, Wollersheim L, (2021). EU Carbon Diplomacy: Assessing Hydrogen Security and Policy Impact in Australia and Germany. Energies, 14( 23): 8103

[33] Hassan Q, Algburi S, Sameen A Z, Jaszczur M, Salman H M, (2024). Hydrogen as an energy carrier: properties, storage methods, challenges, and future implications. Environment Systems & Decisions, 44( 2): 327–350

[34] Humagain S, Sinha R, Lai E, Ranjitkar P, (2020). A systematic review of route optimisation and pre-emption methods for emergency vehicles. Transport Reviews, 40( 1): 35–53

[35] Ivanov D, Dolgui A, Sokolov B, Ivanova M, (2017). Literature review on disruption recovery in the supply chain. International Journal of Production Research, 55( 20): 6158–6174

[36] Katsaliaki K, Galetsi P, Kumar S, (2022). Supply chain disruptions and resilience: A major review and future research agenda. Annals of Operations Research, 319( 1): 965–1003

[37] Kljaić Z, Pavković D, Cipek M, Trstenjak M, Mlinarić T J, Nikšić M, (2023). An Overview of Current Challenges and Emerging Technologies to Facilitate Increased Energy Efficiency, Safety, and Sustainability of Railway Transport. Future Internet, 15( 11): 347

[38] Kuby M JMartinez A SKelley S B & Tal G (2023). Hydrogen station location analysis and optimization: Advanced models and behavioral evidence. Hydrogen Economy, 315–380

[39] Moudio M P E, Bolin R, Carpenter A, Reese S B, Shehabi A & Rao P, (2022). Characterizing manufacturing sector disruptions with targeted mitigation strategies. Environmental research: infrastructure and sustainability, 2( 4): 042001

[40] Nazib R A, Moh S, (2020). Routing protocols for unmanned aerial vehicle-aided vehicular ad hoc networks: A survey. IEEE Access : Practical Innovations, Open Solutions, 8: 77535–77560

[41] Ng M T, Hernandez A, Durango-Cohen P L, Mahmassani H S, (2024). Trading off energy storage and payload–An analytical model for freight train configuration. Transportation Research Part E, Logistics and Transportation Review, 187: 103601

[42] Okonkwo E C, Al-Breiki M, Bicer Y, Al-Ansari T, (2021). Sustainable hydrogen roadmap: A holistic review and decision-making methodology for production, utilisation and exportation using Qatar as a case study. International Journal of Hydrogen Energy, 46( 72): 35525–35549

[43] Pando V, San-José L A, Sicilia J, Alcaide-López-de-Pablo D, (2024). An inventory model with price-and stock-dependent demand and time-and stock quantity-dependent holding cost under profitability maximization. Computers & Operations Research, 164: 106520

[44] Patnala P K, Regehr J D, Mehran B, Regoui C, (2024). Resilience for freight transportation systems to disruptive events: a review of concepts and metrics. Canadian Journal of Civil Engineering, 51( 3): 237–263

[45] Reuß M, Dimos P, Léon A, Grube T, Robinius M, Stolten D, (2021). Hydrogen road transport analysis in the energy system: A case study for Germany through 2050. Energies, 14( 11): 3166

[46] Robles J O, Azzaro-Pantel C, Aguilar-Lasserre A, (2020). Optimization of a hydrogen supply chain network design under demand uncertainty by multi-objective genetic algorithms. Computers & Chemical Engineering, 140: 106853

[47] Sabio N, Gadalla M, Guillén-Gosálbez G, Jiménez L, (2010). Strategic planning with risk control of hydrogen supply chains for vehicle use under uncertainty in operating costs: a case study of Spain. International Journal of Hydrogen Energy, 35( 13): 6836–6852

[48] Sgarbossa F, Arena S, Tang O, Peron M, (2022). Renewable hydrogen supply chains: A planning matrix and an agenda for future research. International Journal of Production Economics, 250: 108712

[49] Smit R, Helmers E, Schwingshackl M, Opetnik M, Kennedy D, (2024). Greenhouse Gas Emissions Performance of Electric, Hydrogen and Fossil-Fuelled Freight Trucks with Uncertainty Estimates Using a Probabilistic Life-Cycle Assessment (pLCA). Sustainability, 16( 2): 762

[50] Vaidya H & Chatterji T (2020). SDG 11 Sustainable Cities and Communities: SDG 11 and the New Urban Agenda: Global Sustainability Frameworks for Local Action. Actioning the Global Goals for Local Impact: Towards Sustainability Science, Policy, Education and Practice, 173–185

[51] Valença G, Moura F & de Sá A M, (2021). Main challenges and opportunities to dynamic road space allocation: From static to dynamic urban designs. Journal of urban mobility, 1: 100008

[52] Visentini M S, Borenstein D, Li J Q, Mirchandani P B, (2014). Review of real-time vehicle schedule recovery methods in transportation services. Journal of Scheduling, 17( 6): 541–567

[53] Wang G, Gunasekaran A, Ngai E W, Papadopoulos T, (2016). Big data analytics in logistics and supply chain management: Certain investigations for research and applications. International Journal of Production Economics, 176: 98–110

[54] Wang X, Wu Y, Wen Z, Cui Z, Wang Y, (2024). A New Transportation Route Planning Method for Wind-Based Hydrogen Supply Chains. ACS Sustainable Chemistry & Engineering, 12( 22): 8436–8452

[55] Xu J, Bo L, (2024). Enhancing supply chain efficiency and resilience using predictive analytics and computational intelligence techniques. IEEE Access, 12: 183451–183466

[56] Zhu Y, Jin J G, Wang H, (2024). Path-choice-constrained bus bridging design under urban rail transit disruptions. Transportation Research Part E, Logistics and Transportation Review, 188: 103637

RIGHTS & PERMISSIONS

Higher Education Press