Relationship Between Area and Capacity of Hydrogen Refueling Stations and Derivation of Design Recommendations

Abstract

Hydrogen plays a pivotal role in the decarbonization of the transport sector, necessitating the development of an adequate infrastructure in the form of hydrogen refueling stations (HRSs) to support hydrogen-powered vehicles. This study investigates the characteristics of hydrogen refueling stations to optimize their spatial design and provide key performance indicators for spatial efficiency. An overview of HRS components and their operational requirements is provided, alongside the classification of stations into distinct categories. The primary focus is on analyzing the relationship between station area and capacity. Utilizing spatial data from hydrogen stations, areas are determined through Google Maps analysis. Linear and power regression models are applied to quantify the relationship, with both models proving effective for capturing these dynamics. Based on the findings, spatially efficient design recommendations are proposed, supplemented by examples and a conceptual blueprint for optimized HRS construction, which are then summarized in a morphological design catalog.

1. Introduction

The utilization of hydrogen in order to achieve climate and decarbonization goals is very promising. One crucial use case for hydrogen is as fuel for heavy-duty and potentially also for light-duty or passenger vehicles. In order to facilitate the use of hydrogen, the creation of a coherent refueling station network is essential. This conflicts with rising land prices, especially in densely populated areas. It is thus unavoidable to create a system of target values for estimating the spatial efficiency of a hydrogen refueling station design in the early phase of development. This study builds on the insights gained about the early-phase product development of HRSs in the study by Nolte et al., where key design characteristics, the operation of HRSs, and concepts have been identified [1].

2. State of the Art

This chapter reviews the state of the art in hydrogen refueling station (HRS) infrastructure, operations, key technical concepts, regulations, and characteristics.

2.1. Hydrogen Refueling Station Infrastructure

For the long-term implementation of fuel cell vehicles (FCVs), establishing a nationwide hydrogen refueling infrastructure is essential. In Germany, significant progress is evident, with 87 operational 700-bar hydrogen refueling stations and 27 under development, primarily concentrated in the southern and western regions and metropolitan areas like Hamburg and Berlin. Europe hosts 162 operational 700-bar stations, and there are 60 under construction, with Germany leading the continent, while other European countries, such as the Netherlands or Switzerland, have also developed larger networks of over a dozen stations. Figure 1 illustrates hydrogen stations operated by H2 Mobility, detailing stations in Germany (left) and the count per country (right). White numbers indicate operational stations, and blue numbers represent those under construction. For 350-bar stations, Germany has 37 operational and 41 under construction, often as extensions of 700-bar facilities, totaling over 90 stations nationwide [2]. Globally, Asia leads with 275 operational stations, while the USA has 75, 50 of which are in California [3].

To ensure a seamless hydrogen supply, further infrastructure expansion is required. In 2022, H2 Mobility announced plans to increase Germany’s hydrogen stations to 300 by 2030, including smaller stations to support heavy-duty vehicles [4]. Research by Krieg [5] and Mayer [6] addressed the optimal number of stations in Germany. Krieg’s study suggests approximately 10,000 stations, below the current number of conventional refueling stations, while Mayer estimates around 5000 stations with a capacity of 1100 kg/day. These findings highlight the cost-saving potential of an optimally planned hydrogen refueling network [5,6].

2.2. Hydrogen Refueling Station Operation

Hydrogen refueling station operations describe the delivery mode and capacity or the on-site production mode and capacity of hydrogen to the HRS, the state in which the hydrogen is delivered, and the storage and dispensing characteristics of the hydrogen. A detailed description of HRS operations is found in the study by Nolte et al. [1].

2.3. Hydrogen Refueling Station Characterization

Based on the analysis of multiple studies by Nolte et al., it clearly follows that area and capacity are among the most relevant characteristics of hydrogen refueling stations [1]. The relevance of the further investigation of the relationship between the area and capacity of HRSs is also showcased by the study of Šimunović et al., who investigates exactly this relationship for a special use case in great detail [7].

This study compares its findings on spatial requirements with prior research. Recommendations from [3] suggest areas of 80–250 m² for S-stations, differing slightly in starting and upper limits. Power regression aligns more closely with these values, making it more representative. For M-stations, recommended areas range from 200–350 m², but this study’s results exceed these limits, likely due to strict size category separations. For L-stations, recommended areas of 250–800 m² match this study’s intervals, though significant variability within this size class indicates strict boundaries may not always be appropriate. The “Tankstelle der Zukunft” project addresses transitioning from diesel to hydrogen for trucks, incorporating space for waiting areas. The coefficients are higher than in this study due to the dataset focusing on truck stations with capacities ≥ 400 kg/d, unlike this study, where <25% of stations meet this threshold. The linear relationship becomes relevant for larger capacities [8]. A study by NOW-GmbH examined hydrogen infrastructure for bus fleets. The linear slope is lower than in this study, indicating slower area growth with capacity. However, the minimum station size of 372.05 m² is larger than this study’s (~240 m²). Despite the differences, the results show some alignment, noting that these findings apply to bus rather than car stations [9]. Nolte’s work, using literature-based data, derives a slope coefficient by averaging area-to-capacity ratios, resulting in 1.6 m²/kg. This coefficient is larger than that of this study but lacks a defined minimum area, aligning well within specific intervals. However, caution is advised, as it relies on averages, not raw data [1].

2.4. Research Questions

Based on the evaluation of the literature shown in Section 2.1, Section 2.2 and Section 2.3, two aims can be formulated for this research paper:

Question 1: Is there a significant relationship between the surface area and capacity of HRSs?

Question 2: Is it possible to derive morphological characteristics for HRSs based on the data gathered from Question 1?

3. Materials and Methods

3.1. Data Collection

This study utilized two datasets. The first dataset, from the H2Tools website of the “Hydrogen Analysis Resource Center”, includes data on hydrogen technologies [10]. Updated on 30 September 2022, it covers 546 operational and 89 planned hydrogen refueling stations worldwide, detailing location, operator, commissioning date, and capacity. The second dataset, obtained from “Clean Hydrogen Partnership”, features 187 European stations as of 13 May 2024, with additional information on location, operator, date, capacity, opening hours, and payment methods [11].

Since neither dataset provided spatial dimensions for the stations, these were manually determined using Google Maps. Addresses were entered into the satellite view, and the “measure distance” tool was used to calculate areas, including refueling zones and technical components. The first step after identifying the image was to determine the overall area, as shown on the left of Figure 2. After evaluating the dataset, it was apparent, due to the high variance in building solutions of HRSs, that the scope of the analysis should be narrowed down to only the crucial components of an HRS. Challenges arose when stations were integrated into existing gas stations or shared premises with other businesses. Thus, only technical facilities and the refueling area have been considered for further investigation to make the gathered data comparable, as shown in Figure 2. Entrance, exit, and total property areas were recorded separately. Google Street View was used when available, but image updates were rare, as many stations were recently constructed.

Additional issues arose when hydrogen refueling stations could not be located at the provided addresses or nearby, likely due to outdated satellite images, as many stations are newly constructed. It was also unclear in some cases which areas belonged to the station, partly due to poor image quality available on Google Maps. To address these limitations, a quality scale from 1 to 4 was introduced to rate the accuracy of spatial measurements, with lower values indicating higher data quality.

Table 1. Data quality of the area measurements.

Data Quality	1	2	3	4
Amount	30	22	3	52

presents the distribution of the image quality levels. Data quality was used, as shown in Appendix A, to further investigate whether results change upon changes in data quality. The results in Appendix A show that results remain constant irrespective of data quality.

3.2. Characterization of the Dataset

Of the initial 733 hydrogen refueling stations, only 134 remain due to missing capacity data, incorrect addresses, or poor image quality. Dataset overlaps were resolved by prioritizing capacity values from the Hydrogen Analysis Resource Centre, as they provided exact figures. From the remaining 134 stations, South Korea accounts for 85 stations, the largest share, followed by 14 in France, 10 in the Netherlands, and smaller numbers in other European countries [12].

Capacity adjustments were necessary for the Clean Hydrogen Partnership dataset, which provided ranges as follows: <100 kg/d, 100–500 kg/d, and >500 kg/d. Conservative estimates set <100 kg/d to 70 kg/d and >500 kg/d to 600 kg/d unless additional data were found. The mid-range was averaged at 300 kg/d.

Table 2. Overview of the dataset.

Variable	Min.	1. Quantile	Median	Mean	3. Quantile	Max.	NAs
Data Quality	1	2	3	2.791	4	4	0
Area [m2]	60	230	322.5	352.3	427.5	1100	0
Capacity [kg/d]	20	250	250	330.5	300	2000	0
Number of 350-bar Dispensers	0	0	0	0.44	1	3	0
Number of 700-bar Dispensers	0	1	1	1.2	1	3	69
Total Number of Dispensers	1	2	2	2.108	3	5	69
Start Time	2012	2019	2021	2020	2021	2022	40

summarizes dataset variables, showing minimum, maximum, quantiles, and missing data.

Technical component areas range from 60 to 1100 m², with an average of 352.3 m² and 75% exceeding 230 m². Entry/exit areas (82 missing measurements) span 80–1650 m², with a median of 455 m² and an average of 554.6 m², skewed by outliers. Property sizes range widely up to 4800 m², with a median of 1600 m² and an average of 1883 m². Station capacities vary from 20 to 2000 kg/d, averaging 330.5 kg/d. Categorization by size yields 24 small (S), 87 medium (M), 20 large (L), and 3 extra-large (XL) stations. HRS t-shirt sizes, from S to XL, have been defined by Nolte et al. [1]. Dispenser data shows 44% offer 350-bar refueling, while stations average 1.2 700-bar dispensers and 2.108 total dispensers. The largest station, in For sur Mer, France, has 5 dispensers. Opening years range from 2012 to 2022, with a median of 2021, indicating half were commissioned in the last two years.

The average values for hydrogen stations by size are analyzed, dividing them into the four predefined size categories. Averages for area and dispenser count are calculated and compared with H2 Mobility’s assumptions, which use the same size classification, providing a reliable benchmark for data quality. The comparison is shown in

Table 3. Categorization of HRS into t-shirt sizes.

Size	S	M	L	XL
Number of Stations	24	87	20	3
Average Number of Dispensers	1.7	2.16	2.8	3
Number of Dispensers (per [3])	1	2	2–3	2–4
Average Area [m2]	260.42	321.25	566.75	560
Space Requirement (per [3], [m2])	80–250	200–350	250–800	-

.

S-stations have a slightly higher average dispenser count, potentially due to missing data or an overrepresentation of S-stations with two dispensers in the dataset. M-station averages closely align with literature recommendations, while L-stations fall within the recommended range. XL-station data is limited to one station, making it non-representative despite agreement with literature dispenser counts. A general trend shows increasing dispensers with station size. Regarding area, S-stations slightly exceed the recommended maximum, possibly due to a lack of mobile stations in the dataset. Mobile stations, designed for quick deployment and low capacity, are often used to establish hydrogen availability in new regions [13]. M- and L-station areas match literature values, while XL-station sizes are not explicitly defined in the literature but depend on hydrogen technology. Nonetheless, a trend of increasing area with station size is evident, except for the non-representative XL station.

The data presented in the table above are visualized in Figure 3 to illustrate the trends and relationships between station size, t-shirt size, and area requirements.

4. Evaluation of the Dataset

The following chapter contains the formal analysis of correlation possibilities for the dataset and the possible interpretation of the identified results.

4.1. Analysis of the Identified Data

The regressions were conducted using R, a programming language specialized in statistical analysis [14]. Results are presented in

Table 4. Analyzed correlations (**: p ≤ 0.01; ***: p ≤ 0.001).

Model	y	x	a	b	c	Significance	${\mathit{R}}_{\mathit{a}\mathit{d}\mathit{j}}^2$
Linear	Area	Capacity	-	0.364	232.13	***	0.3195
Quadratic	Area	$\mathrm{Capacity}+{\mathrm{C}\mathrm{a}\mathrm{p}\mathrm{a}\mathrm{c}\mathrm{i}\mathrm{t}\mathrm{y}}^2$	−0.0002	0.664	171.7	**	0.3613
Logarithmic	Area	ln(Capacity)	-	120.19	−312.89	**	0.2777
Power	ln(Area)	ln(Capacity)	-	3.7093	0.368	***	0.3189

. The first column specifies the regression type, while the second and third columns describe variable modifications for compatibility with linear regression. Columns four, five, and six list the respective coefficients. The penultimate column indicates result significance, with stars denoting the probability of rejecting the null hypothesis: three stars (>99.9%), two stars (99–99.9%), and one star (95–99%). The final column reports the regression quality as the adjusted R² value.

All explored correlations between station area and capacity across the tested models did not yield statistically significant results. While some models demonstrated moderate adjusted R² values, these were insufficient to establish robust predictive relationships, indicating that additional factors may influence the observed variance. This is shown in Figure 4. Further approaches to analyzing the dataset by segmenting it are shown in Appendix A, where no significant correlation has been found as well.

4.2. Explanations for the Dataset

The lack of correlation between the area and capacity of hydrogen refueling stations can be attributed to several factors. In rural or suburban areas, where land is relatively inexpensive compared to densely populated regions, the station’s area is often not a critical design consideration or constraint. Additionally, irregularly shaped plots allocated for station development may lead to less efficient designs, further weakening the relationship. The consistent capacity increments across stations are likely due to the use of standardized, mass-produced storage equipment, which operates in discrete capacity steps. The integration of hydrogen refueling stations within conventional fuel stations may also distort the data. Moreover, dataset inconsistencies stemming from challenges in collecting accurate area and capacity data via Google Maps and original sources further contribute to the observed lack of correlation. A further possibility for the state of the dataset is the consideration of policies and standards, such as safety regulations like BetrSichV, ISO 20100, or SAE J2601, which mandate certain spatial design constraints for HRSs, depending on hydrogen capacity [15,16,17].

5. Design Suggestions for HRSs

The evaluation of data provides no clear guidance for building a new hydrogen station. However, analyzing existing stations and implementing design considerations based on the gathered data supports the creation of efficient HRS designs. This chapter examines real-world stations and concludes with a design proposal based on the gathered insights. The chosen HRSs possess a high spatial efficiency and are thus to be considered as suitable design examples for further HRSs. Still, it is important to mention that each HRS is the result of an individual design process that is supported by the identified criteria. The design of HRSs will be evaluated on the basis of spatial efficiency, a metric that is the ratio of HRS capacity to HRS area.

5.1. Design Examples for Small-Sized HRSs

The relation between capacity and spatial efficiency (SE), as well as the relation between area and SE for S-sized HRSs, has been investigated to derive further insight regarding the design of HRSs; this is illustrated in Figure 5.

While no clear correlation is evident between capacity and SE, for the relation between area and SE, a clear correlation is visible, which shows that an increase in area for S-sized HRSs leads to a decreasing SE. Based on the right graph in Figure 5, the top three highest-SE HRSs were selected for further analysis. For an HRS to be in the top 10% of the S-size class regarding SE, an SE value of 0.69 ${\displaystyle \frac{\mathrm{k}\mathrm{g}/\mathrm{d}}{{\mathrm{m}}^2}}$ is required. The average HRS in the M category has a spatial efficiency of 0.37 ${\displaystyle \frac{\mathrm{k}\mathrm{g}/\mathrm{d}}{{\mathrm{m}}^2}}$.

The hydrogen refueling station marked with number 1 in Figure 6 is situated in Douains, France. Technical components are located at a roundabout featuring a wide entryway that splits into two lanes. The left lane provides direct access to the dispenser. It has an area of 100 m² and a capacity of 70 kg/d, giving it a spatial efficiency of 0.7 ${\displaystyle \frac{\mathrm{k}\mathrm{g}/\mathrm{d}}{{\mathrm{m}}^2}}$. The hydrogen refueling station marked with number 2 in Figure 6 is situated in Paris, France. A small, covered section is designated for refueling, allowing vehicles to be refueled sequentially. It has an area of 100 m² and a capacity of 70 kg/d, giving it a spatial efficiency of 0.7 ${\displaystyle \frac{\mathrm{k}\mathrm{g}/\mathrm{d}}{{\mathrm{m}}^2}}$. The hydrogen refueling station marked with number 3 in Figure 6 is situated in Göteborg, Sweden. A small, covered section is designated for refueling, allowing vehicles to be refueled sequentially. It has an area of 80 m² and a capacity of 70 kg/d, giving it a spatial efficiency of 0.88 ${\displaystyle \frac{\mathrm{k}\mathrm{g}/\mathrm{d}}{{\mathrm{m}}^2}}$.

During the development of the identified HRSs, a compact design was chosen that incorporated all essential components. The exclusive use of 350-bar refueling reduces costs and space needed for compressors and storage. Delivery likely involves small gas cylinders, as no large tanks or truck parking areas are apparent. If these cylinders are pressurized to at least 350 bar, only a small compressor is needed for refueling management. This design allows easy integration into various locations but is limited by the low refueling pressure. Based on Table 1, it is sensible to design the overall storage capacity to be below 3 t. The suggested design baseline value for spatial efficiency is 0.69 ${\displaystyle \frac{\mathrm{k}\mathrm{g}/\mathrm{d}}{{\mathrm{m}}^2}}$ to be among the 10% most spatially efficient designs.

5.2. Design Examples for Medium-Sized HRSs

The relation between capacity and spatial efficiency (SE), as well as the relation between area and SE, for M-sized HRSs has been investigated to derive further insight regarding the design of HRSs; this is illustrated in Figure 7.

While no clear correlation is evident between capacity and SE, for the relation between area and SE, a clear correlation is visible, which shows that an increase in area for M-sized HRSs leads to a decreasing SE. Based on the right graph in Figure 7, the top three highest-SE HRSs were selected for further analysis. For an HRS to be in the top 10% of the M-size class regarding SE, an SE value of 1.50 ${\displaystyle \frac{\mathrm{k}\mathrm{g}/\mathrm{d}}{{\mathrm{m}}^2}}$ is required. The average HRS in the S category has a spatial efficiency of 0.97 ${\displaystyle \frac{\mathrm{k}\mathrm{g}/\mathrm{d}}{{\mathrm{m}}^2}}$.

The station marked with number 1 in Figure 8 is on the premises of a conventional refueling station located in Berlin, Germany. It has an area of 150 m² and a capacity of 300 kg/d, giving it a spatial efficiency of 2 ${\displaystyle \frac{\mathrm{k}\mathrm{g}/\mathrm{d}}{{\mathrm{m}}^2}}$. Unlike the previous station, the one marked with number 2 in Figure 8 is standing independently, located in Paray-Vielle-Poste in France. The station has an area of 150 m² and a capacity of 300 kg/d, giving it a spatial efficiency of 2 ${\displaystyle \frac{\mathrm{k}\mathrm{g}/\mathrm{d}}{{\mathrm{m}}^2}}$. The HRS marked by number 3 in Figure 8 is standing independently, as well. The station has an area of 180 m² and a capacity of 500 kg/d, giving it a spatial efficiency of 2.78 ${\displaystyle \frac{\mathrm{k}\mathrm{g}/\mathrm{d}}{{\mathrm{m}}^2}}$.

To meet higher hydrogen demand, station capacity must be increased. In Germany, it is common for hydrogen to be offered alongside conventional fuels, often by retrofitting or adding a dispenser near existing ones, typically feasible for S- or M-stations. This type of station includes more components than the smaller S-category station, such as a standalone dispenser, a roof, a compressor, and larger storage tanks. It is also clearly visible that medium-sized HRS can achieve higher spatial efficiency by increasing capacity more significantly than its spatial requirements. The suggested design baseline value for spatial efficiency is 1.50 ${\displaystyle \frac{\mathrm{k}\mathrm{g}/\mathrm{d}}{{\mathrm{m}}^2}}$.

5.3. Design Examples for Large- and Extra-Large-Sized HRSs

The relation between capacity and spatial efficiency (SE), as well as the relation between area and SE, for L-sized HRSs has been investigated to derive further insight regarding the design of HRS; this is illustrated in Figure 9. Due to a lack of sufficient data, stations of size XL are not visualized.

While no clear correlation is evident between capacity and SE, for the relation between area and SE, a clear correlation is visible, which shows that an increase in area for L-sized HRSs leads to a decreasing SE. Based on the right graph in Figure 9, the top three highest-SE HRSs were selected for further analysis. For an HRS to be in the top 10% of the L-size class regarding SE, an SE value of 1.97 ${\displaystyle \frac{\mathrm{k}\mathrm{g}/\mathrm{d}}{{\mathrm{m}}^2}}$ is required. The average HRS in the L category has a spatial efficiency of 1.38 ${\displaystyle \frac{\mathrm{k}\mathrm{g}/\mathrm{d}}{{\mathrm{m}}^2}}$.

For opening an L- or XL-sized station to meet higher demand, dedicated land is recommended. The L-sized HRS shown in Figure 10 on the left is located in Changwon-Si, South Korea, and has an area of 500 m² and a capacity of 1000 kg/d, giving it a spatial efficiency of 2 ${\displaystyle \frac{\mathrm{k}\mathrm{g}/\mathrm{d}}{{\mathrm{m}}^2}}$. The L-sized HRS shown in the middle of Figure 10 is located in Changwon, South Korea, and has an area of 325 m² and a capacity of 640 kg/d, giving it a spatial efficiency of 1.97 ${\displaystyle \frac{\mathrm{k}\mathrm{g}/\mathrm{d}}{{\mathrm{m}}^2}}$. The HRS shown in the right of Figure 10 is located in Goesan, South Korea, and has an area of 210 m² and a capacity of 600 kg/d, giving it a spatial efficiency of 2.86 ${\displaystyle \frac{\mathrm{k}\mathrm{g}/\mathrm{d}}{{\mathrm{m}}^2}}$. It is visible that larger HRSs possess a higher spatial efficiency than smaller-sized HRSs. The average HRS in the L category has a spatial efficiency of 1.38 ${\displaystyle \frac{\mathrm{k}\mathrm{g}/\mathrm{d}}{{\mathrm{m}}^2}}$. The suggested design baseline value for spatial efficiency is 1.97 ${\displaystyle \frac{\mathrm{k}\mathrm{g}/\mathrm{d}}{{\mathrm{m}}^2}}$.

5.4. Overall Analysis of Spatial Efficiency

An overview of the SE of the complete dataset is provided in Figure 11.

By analyzing Figure 11, it is visible that the trends identified for the t-shirt sizes regarding SE do not hold true for the entirety of the dataset, further validating the use of t-shirt sizes as measures for categorizing HRSs. The main conclusion of the analysis of SE has been the definition of a suggested design target to optimize the land use of the HRS designs. Another overview of the SE data is provided by the boxplots in Figure 12.

Size S has the smallest spread, indicating low variability, with no apparent outliers. Size M has a larger interquartile range (IQR) and multiple outliers above the upper whisker, suggesting higher variability and the presence of extreme values. Size L has a similar spread but is slightly larger compared to Size M and appears more symmetrically distributed with fewer outliers. Overall, Size M shows the highest dispersion and most outliers, while Size S has the least variation. A possible approach for interpreting this result is that for S-sized HRSs, space as a design constraint plays a much larger role while still having a lower SE value. This could be explained by the fact that small stations are often used in urban environments, often with small but inefficient spaces available for HRS deployment. Larger HRSs can be built outside of dense, urban settings, usually on freely plannable areas; furthermore, larger HRSs profit from a scaled sizing, having the benefit of having just one of the required technical components while having multiple dispensers, thus saving additional space.

5.5. Morphological Design Catalogue

Based on the analysis of the identified best, most efficient example HRSs, a morphological design catalog (MDC) has been developed to support the very early concept development of HRSs by enabling a selection of a discrete number of characteristics for each relevant design aspect of an HRS. The morphological design catalog is shown in Appendix B. The MDC also contains selection examples based on the identified most spatially efficient HRSs.

5.6. Investigating Design Regulations

Hydrogen refueling station construction can follow three models: independent management for full control (suitable for multiple stations), outsourcing to an operator for minimal involvement (ideal for single stations), or commissioning construction with subsequent operation by the client, who controls supply logistics [18]. Key planning factors include hydrogen volume, refueling frequency, vehicle types, location, and applicable government subsidies, which vary by station, country, and capacity [19]. Standards like ISO 20100 or ISO 19880-1:2015 govern safety and environmental requirements, supported by H2 Mobility’s guidelines on detectors, safety distances, and components [20,21]. Furthermore, the main driving factor in the layout design of HRSs is governmental regulation, which differs from country to country, such as the NFPA 2: Hydrogen Technologies Code or the OSHA Standard 1910.103 in the USA, the EIGA Doc 15/20 in the EU (while every country has further specific regulation), and Korea’s Hydrogen Economy Policy document for South Korea, which all contain regulations regarding relevant placement distances of infrastructure [22,23,24,25]. On-site production demands further permits, with timelines from one to seven months [18].

This scope of regulation clearly shows that it has a significant influence on the final layout of HRSs, mainly in regard to the distance from other non-HRS-relevant structures. While the influence of the regulation can never be completely excluded as an influencing factor for spatial efficiency for HRSs, the described HRS area determination method, where just necessary installations are counted while areas such as drive-in and drive-out are excluded, mitigates the influence of many regulations, since those focus on exactly the areas that provide distance from other infrastructure. While the data are even more limited when only considering certain regions, no clear pattern has been identified.

6. Results

This study provides valuable insights into a relatively underexplored field, delivering foundational data and actionable findings to advance the understanding of spatial efficiency in hydrogen refueling station (HRS) design. One of the primary observations was the lack of a clear and consistent relationship between the total area of an HRS and its refueling capacity. While capacity inevitably influences design considerations, the spatial requirements appear to be determined more by site-specific variables and design strategies than by a straightforward correlation with capacity. These findings challenge conventional assumptions and highlight the complexity of designing HRS systems tailored to diverse operational contexts. Despite the absence of a direct relationship between area and capacity, the analysis revealed a notable trend: spatial efficiency increases with larger-capacity stations. Larger-capacity systems, when optimized, make more effective use of available space, suggesting that economies of scale can be achieved not only in operational costs but also in spatial utilization. Through detailed analysis, the study also identified design requirements and established target values for spatial efficiency. The identified benchmarks provide a valuable reference for evaluating existing HRS designs and allowed for the creation of an MDC, formalizing the early-phase development approach.

7. Discussion

This study offers critical insights into the spatial efficiency of hydrogen refueling station (HRS) design despite notable limitations in data availability and quality. A primary challenge was the limited dataset, with minimal publicly accessible information on HRS capacity and area. Area measurements relied heavily on satellite imagery, which was often hampered by cloud cover, outdated images, or insufficient resolution, reducing the precision of area assessments. The large variety of HRS designs and layouts, as well as the integration with other facilities and businesses, made it very difficult to clearly define the station area, leading to the focus on the main station area. Moreover, the dataset’s heavy reliance on stations from East Asia raises concerns about geographic representativeness, as regional design practices may influence the observed trends and limit global applicability. The design recommendations and target values derived from these data are similarly subject to potential bias, as they are based on the limited visual and quantitative information available. However, the robustness of certain findings mitigates these concerns. The analysis confirms with consistency that the relationship between HRS area and capacity is weak and lacks a clear mathematical correlation. This suggests that spatial design depends on complex, context-specific factors rather than simple scaling based on capacity. A stable observation is the increase in spatial efficiency with larger HRS capacities. This trend highlights the potential for improved land-use optimization in higher-capacity stations, reflecting the benefits of scale when supported by effective design strategies. While these findings establish a foundational understanding of HRS spatial efficiency, the study underscores the need for more extensive and diverse datasets, alongside improved spatial measurement techniques, to refine and validate the results. Despite these limitations, the observed trends provide a valuable basis for advancing HRS design and optimizing the use of space in future infrastructure development. Due to the large variety of regulations and standards that influence the design of HRSs, a recommended future study for determining the ideal spatial layout of the station is a rigorous analysis of regulations to determine all relevant constraints and define a “minimum” area from a regulations standpoint as a lower baseline for design.

8. Conclusions

Future research should prioritize direct data acquisition from HRS operators to obtain precise information on station capacity, area, and design parameters, addressing limitations in current open-source data. Additionally, incorporating the broader context of station surroundings—such as adjacent land use, infrastructure integration, and accessibility—will enhance the understanding of spatial efficiency. Categorizing HRS designs by geographical context and urban, rural, or outskirts settings is essential for identifying location-specific trends and developing tailored design frameworks. These approaches will yield more accurate, globally applicable insights to inform the efficient development of hydrogen refueling infrastructure. Additionally, the MDC will be adjusted, along with consultations with representatives from the industry, to validate design concepts and fit solutions closer to industry needs. A further aim is to analyze the influence of regulations and standards on the effects of spatial design of HRSs to identify potential design efficiency limitations arising from these policies and regulations. The results show that further research into the direction of design criteria, as shown by Nolte et al., is required, but this analysis must be created in a multifactorial approach, taking multiple potentially relevant factors, such as traffic intensity, user demand, and surrounding infrastructure development, into consideration [1].