Comparative Analysis of Apron Capacity with the Progressive Introduction of Hydrogen-Powered Aircraft

Abstract

Aviation is currently facing one of its greatest challenges: reconciling growing traffic demand with the need to drastically reduce climate-altering emissions. Hydrogen has emerged as one of the most promising alternatives to decarbonize air transport. However, it poses significant challenges related to cryogenic storage, safety, and the adaptation of the airport infrastructure. Aprons represent a critical issue, as the increased volume of fuel tanks and different refueling protocols directly impact airport operational capacity. This research fits within this framework by analyzing a Code 4E Italian airport over three time horizons: 2025, with an all-kerosene fleet; 2035, with a 25% penetration of hydrogen-powered class A and B aircraft; and 2045, with a further increase in the hydrogen share (75% class A and B and 15% class C). The study evaluates apron capacity using fast-time simulation and compares the outcomes with an analytical model. The results show good consistency between theoretical and simulated capacity. The 2035 and 2045 scenarios with the introduction of hydrogen-powered aircraft show a reduction in apron capacity between 16% and 5% compared to conventional scenarios.

1. Introduction

According to IATA’s Chart of the Week, hydrogen-powered aircraft could represent 18% of the global fleet by 2050, and small regional aircraft would make up ~54% of that hydrogen fleet [1]. Hydrogen is one of the most promising solutions for decarbonizing air transport, as it eliminates carbon dioxide emissions during flight when utilized in fuel cells or direct combustion [2]. Additionally, it offers a higher energy density per unit mass than kerosene (

Table 1. Properties of Jet A-1, liquid (LH₂), and gaseous (GH₂) hydrogen.

Property	Jet A-1	LH2	GH2 (350 Bar)	GH2 (700 Bar)
Specific energy (MJ/kg)	~43	~120	~120	~120
Energy density (MJ/L)	~35	~8.5	~2.9	~4.8
Storage temperature (K)	ambient	~20	ambient	ambient
Storage pressure (bar)	ambient	~2	350	700
Tank gravimetric efficiency (%)	~100	30–90	1–15	1–15

) [3,4]. On the other hand, the low volumetric energy density of liquid hydrogen and the requirement for storage under cryogenic conditions introduce substantial design constraints, resulting in the need for larger tanks, alternative propulsion integration into the airframe, and dedicated safety systems as compared to contemporary aircraft powered by Jet A-1 [4,5,6]. In particular, hydrogen fuel cells generally achieve higher electrical efficiencies and near-zero direct emissions, but they suffer from lower power density and more complex thermal and architectural integration compared with hydrogen combustion turbines, which benefit from lighter and more compact and technologically mature configurations, albeit with residual NO_x emissions [7]. Moreover, while fuel cells are particularly suited to smaller-scale or distributed propulsion applications, combustion-based solutions appear more compatible with higher power requirements and potential retrofitting strategies, though at the expense of lower overall efficiency [8].

Major manufacturers and international institutions see hydrogen as a good solution for short- and medium-range traffic. The first commercial aircraft is expected to enter service around the mid-2030s, if the technological and infrastructural challenges are addressed [9,10]. Several conceptual studies and demonstration programs have indicated that hydrogen-powered configurations are particularly well suited to regional and single-aisle segments. In such cases, mission profiles and range requirements can be met with acceptable trade-offs in terms of payload and operating costs [11,12]. In this context, it is probable that airports will host mixed fleets, with Jet A-1- and hydrogen-powered aircraft coexisting for a protracted transition period [13,14]. Although hydrogen is widely regarded as a leading long-term option for aviation decarbonization, its large-scale adoption depends on simultaneous progress in low-carbon hydrogen production, airport infrastructure, onboard storage, and aircraft redesign, meaning that commercial deployment is expected only through a phased introduction extending toward mid-century rather than in the near term [15,16]. From both an industry and technological perspective, recent studies highlight that, while sustained investments and coordinated roadmaps support the feasibility of hydrogen-powered flight, significant challenges remain in aircraft-level implementation and certification constraints, which may slow down large-scale deployment despite ongoing progress [17]. From an industry perspective, the continued investment and technological milestones achieved by major manufacturers suggest that hydrogen-powered aviation is likely to become a tangible reality, even if timelines shift as engineering and infrastructure challenges are progressively resolved [18]. However, the repeated postponement of entry-into-service targets also highlights persistent uncertainties—particularly around certification, supply infrastructure, and cost competitiveness—indicating that while hydrogen technology is credible, its large-scale commercial adoption may remain slower and more complex than initially anticipated [19].

Several technological and economic research questions remain open. These include whether current fuel-cell stacks meet the required power-density targets for aviation applications, their operational lifespan, overall system costs, and the timeline for technological maturity. Moreover, other unresolved questions arise, concerning NO_x emissions, hydrogen storage concerning gravimetric and volumetric power density, the state of the art of liquid-hydrogen (LH₂) tank integration, the positioning of storage systems within the aircraft, and the extent to which existing airframes would require structural redesign. While manufacturers are actively addressing many of these challenges, the pace of progress and the principal factors causing delays are still under discussion. These topics are highly relevant to the broader transition toward hydrogen aviation; however, they lie outside the primary scope of this study and need to be explored in another paper. Indeed, the present manuscript does not aim to assess propulsion technologies, on-board storage solutions, aircraft design, or economic performance. Instead, it focuses exclusively on apron capacity with the progressive introduction of hydrogen-powered aircraft.

From an airport perspective, the introduction of hydrogen primarily affects production, storage, distribution, and refueling operations with direct effects on apron configuration and capacity [20,21]. Hydrogen can be supplied in liquid or gaseous form, with production occurring on site or via external facilities. Its conditioning and storage take place in dedicated infrastructure, prior to its transfer to aircraft via cryogenic tankers or hydrant systems [22]. For airports with limited initial demand, refueling concepts based on mobile cryogenic tankers are the most viable solution in the short term, while hydrant systems become more attractive as traffic volumes and consumption increase [23,24]. The process of refueling hydrogen differs from refueling Jet A-1 in several aspects. These include the management of cryogenic temperature and boil-off gas, and the utilization of larger-diameter pipes. Furthermore, there are specific purging and chill-down phases, which necessitate revised safety perimeters and dedicated ground handling protocols [5,25]. However, recent studies have indicated that, under optimized conditions, the utilization of liquid hydrogen for refueling can result in turnaround times that are comparable to or even shorter than those achieved with kerosene, particularly when refueling processes are automated and integrated with other ground handling facilities [5,26]. The overall impact on turnaround is therefore not solely determined by refueling; rather, it is a consequence of the combined effect of stand segregation, additional safety constraints, and the coordination of apron operations [18].

In operational terms, hydrogen-powered aircraft may require dedicated or multi-fuel stands, increased separation from terminal buildings and other traffic, and specific routing of ground support equipment, all of which influence apron flexibility and the effective utilization of available parking positions [2,27]. As the fleet of hydrogen-powered aircraft expands, the coexistence of various refueling systems and aircraft categories may impede the interoperability of different stand systems and increase sensitivity to periods of high demand. This potentially decreases the effective apron capacity, even in the presence of an unchanged or growing number of stands.

Furthermore, the potential associated with hydrogen-powered operations includes additional areas of study, currently underway, with a particular focus on three main issues:

• Hydrogen-powered ground support equipment (GSE) reducing apron congestion and emissions, improving overall coordination without new stands.
• Zero-loss liquid hydrogen (LH2) systems minimizing boil-off, enabling higher refueling throughput and stand utilization.
• Robotic refueling for further turnaround gains and safety zone optimization.

Hydrogen-powered GSE, such as baggage tugs, pushback tractors, and belt loaders powered by hydrogen fuel cells, offers zero-emission operation on the apron, eliminating exhaust from diesel vehicles that contribute to local air pollution and safety hazards near passengers and personnel. These vehicles produce only water vapor, aligning with airport sustainability goals and reducing the carbon footprint during ground operations [28], which account for a significant portion of airport emissions [29]. Zero-loss LH2 systems minimize boil-off losses, boosting refueling throughput and apron stand utilization. Advanced designs with integrated refrigeration recover boil-off gas for recooling, ensuring stable density and high flows. This sustains rapid chill-down and refueling for regional jets, matching turnaround requirements, although safety is a major challenge in the development of this technology [30,31]. Since its first implementation at Munich Airport in 1999, robotic refueling has streamlined hydrogen-powered aircraft operations by shortening turnaround times and refining safety perimeters on the apron. Automated systems deftly manage cryogenic hose connections with pinpoint accuracy, slashing connection durations and enabling seamless integration of purging and fueling phases [5].

On aprons, zero-loss systems enable reliable hydrant delivery without venting interruptions, potentially increasing peak stand utilization by 15–20% through more stable hydrogen supply under mixed-fleet operations. For short-haul Italian airports, these systems expand capacity envelopes, averting saturation from quick cycles per Mirković models. Scalable to 320 t LH2 daily, these cut costs by 25% through energy recovery, enhancing flexibility without new infrastructure.

Airport capacity is commonly defined as the maximum number of arrivals and departures that can be handled safely while maintaining an acceptable level of service (LoS). Institutions such as the FAA, ACI, and Eurocontrol define capacity in slightly different ways, distinguishing between theoretical and practical levels based on acceptable delays and operational conditions [32,33,34]. These definitions highlight that capacity is dynamic, because it depends on performance objectives, traffic characteristics, and external constraints. Runway configuration and operating procedures constitute the primary determinants of airport capacity [35,36,37,38], as they govern the maximum number of allowable movements per hour. Improvements such as reduced runway occupancy time (ROT) [39] and rapid-exit taxiways can significantly enhance throughput [40,41]. Moreover, apron capacity is equally important: the number and flexibility of stands, together with turnaround efficiency, shapes the airport’s ability to accommodate traffic peaks. Digital coordination systems like A-CDM can help reduce ground times, improving overall performance [34]. Mirković and Tošić [42] propose the use of an apron capacity envelope, which describes capacity not as a single value but as a range of possible configurations depending on the proportion of small-, medium-, and wide-body aircraft. The same study introduces the notion of apron flexibility, defined as the system’s ability to adapt to changes in demand or traffic mix. A flexible apron with multi-purpose, reconfigurable stands is more resilient than one with rigid stands dedicated to specific aircraft categories. Apron saturation often occurs before runway saturation, particularly at airports dominated by short-haul traffic, where rapid turnarounds generate stand congestion [43]. In such cases, stand shortages affect ground delay and become the main operational bottleneck, causing cascading delays affecting the entire airport system [36,44]. Delays can occur at different stages of an aircraft’s movement [37,42] and constitute one of the main key performance indicators (KPIs) for evaluating the level of service at an airport [37,44]. According to the FAA [34], a delayed flight deviates by at least 15 min from its scheduled time. Other studies [38,41] identify three service levels based on average delay: optimal service if the average delay per flight is 3–4 min; critical capacity conditions with average delays of 10–15 min; and system saturation when delays exceed 20 min.

This study aims to assess through a fast-time simulation how the gradual introduction of hydrogen-powered aircraft may affect the apron capacity of an Italian airport herein not disclosed due to privacy reasons. The results offer insights for infrastructure planning and for supporting the aviation sector’s energy transition.

2. Materials and Methods

The evaluation methodologies of airport capacity can be classified into four categories:

• Historical analyses—These use recorded traffic and delay data at the airport to estimate actual capacity under real conditions [35]. They are simple to apply but cannot represent future scenarios or infrastructure changes.
• Analytical models—Based on formulas and theoretical approaches (e.g., queuing theory, Poisson distributions) [36,39].
• Real-time simulation (RTS)—These reproduce operations with a 1:1 correspondence between real and simulated time. They offer a high level of detail but are costly and not very flexible for analyzing multiple scenarios [42].
• Fast-time simulation (FTS)—This allows operational dynamics to be reproduced under accelerated simulation times, enabling the analysis of entire traffic days in just a few minutes [45]. This well-established approach is the most effective tool for assessing apron capacity, as it allows simultaneous consideration of aircraft movements, turnaround operations, and interactions with ground support equipment [46,47].

This study presents the evaluation of the apron capacity of an Italian airport by applying analytical modeling and fast-time simulation.

2.1. Analytical Models

In this study, the analytical model proposed by Horonjeff [35] was applied, as it best captures the composition of the apron. System capacity is reached when the demand for stand occupancy equals the actual availability of the apron (Equation (1)):

$$C_{gi}={\displaystyle \frac{{\mu }_{k,i}·N_{k,i}}{E\left(T_g\right)}}$$

(1)

where ${\mu }_{k,i}$ represents the gate utilization ratio, defined as the ratio between the average number of occupied stands and the total number of available stands of type-k gates; $N_{k,i}$ is the number of type-k gates available for aircraft of type i; $C_{gi}$ is the capacity of type-k gates, expressed in aircraft per hour; and $E\left(T_g\right)$ is the expected gate occupation time required by aircraft that may use type-k gates (Equation (2)):

$$E\left(T_g\right)= \sum m_i·T_{gi}$$

(2)

where $m_i$ is the percentage of type-i aircraft in the fleet mix using the gates at airport, and $T_{gi}$ is the gate occupancy time required for type-i aircraft.

The model adopts a constrained stand-usage strategy: aircraft belonging to a given category or, as in this case, powered by a specific fuel, may only use stands designed for that or higher categories. The apron is therefore divided into homogeneous stand groups, each able to accommodate only certain types of aircraft and thus having a partial capacity $C_{gi}$.

However, these groups do not operate independently: some may reach saturation quickly, while others remain partially unused. Consequently, the overall apron capacity is limited by the group that saturates first (Equation (3)):

$$C=\mathrm{m}\mathrm{i}\mathrm{n}\left(C_{gi}\right)$$

(3)

Operationally, this condition implies that:

• Each stand group has its own capacity, determined by the number of available positions and the associated average occupation time.
• The overall system can operate without issues only as long as all groups have available stands.
• Once one group reaches saturation, the total apron capacity stabilizes at the minimum value corresponding to the limiting group.

In this way, the model identifies the dominant structural or operational constraint that determines the maximum sustainable apron capacity under continuous traffic demand.

2.2. Fast-Time Simulation

The distinctive feature of the FTS approach is its ability to capture nonlinear interactions among the different system components, aircraft, infrastructure, operational constraints, and traffic procedures. In complex systems like airports, there is no direct or proportional relationship between increases in traffic demand and overall performance. Operational experience shows that small increases in traffic can generate amplified and disproportionate effects, especially near capacity limits. Even a seemingly modest increase in movements can cause taxiway congestion, departure queues, or stand saturation, with much greater impact than the initial variation. FTS allows these phenomena to be represented by simulating in detail the behavior of each aircraft and the evolution of operations over time. Delays, conflicts, and congestion thus emerge from the model, providing a realistic representation of the airport performance under various traffic conditions.

AirTOP 5.1.3 [48] is a common platform based on this approach. It allows integrated modeling of both airside components (runways, separation procedures, arrivals, and departures) and ground components (taxiways, aprons, and stands), producing a wide range of synthetic indicators, including maximum hourly and daily movements; average and total delay by type (runway, taxi, and apron); apron saturation levels and stand utilization rates; and turnaround times and associated variability. For apron capacity assessment, AirTOP simulates the entire turnaround cycle, including disembarkation and boarding, refueling, catering, and light maintenance. The results are then aggregated to estimate stand saturation levels and identify the operational bottlenecks. Effective capacity is assessed based on the average delay: low values indicate available traffic growth margins, while increasing delays signal the approach to the practical capacity limit. AirTOP records values as delay and throughput during simulation, and those can be used to generate capacity curves relating hourly movements to average delay and analyze the system performance through KPIs. This approach is particularly relevant for analyzing innovative scenarios, such as the introduction of hydrogen-powered aircraft.

3. Case Study

3.1. Description of the Airport

This paper examines a case study of a Code 4E Italian airport [49] that plays a strategic role for both passenger and cargo traffic (Figure 1). The airport has two parallel runways, with a centerline separation of 202 m, which prevents their simultaneous use.

The main runway is 3300 m long, 45 m wide, and equipped with 15 m shoulders on each side. It is fitted with an Instrument Landing System (ILS) Category III B. The second runway is not used for regular commercial traffic and primarily serves as a parallel taxiway to the main runway. Its dimensions are 2780 m × 45 m, with 7.5 m shoulders on each side. The approach path is Category I.

The apron can accommodate both narrow-body and wide-body aircraft and includes two main types of stands: contact stands, directly connected to the terminal via loading bridges, and remote stands. The stands are divided into five main areas (i.e., 100, 200, 300, 400, and 500) (Figure 2):

• Area 100 consists of five MARS (Multiple Aircraft Ramp System) stands and seven narrow-body remote stands. The MARS stands can alternately accommodate five class D or E wide-body aircraft, or ten class C narrow-body aircraft, providing significant operational flexibility.
• Area 200 consists of four remote stands primarily for class C aircraft. Passenger boarding and disembarking are carried out via shuttle buses.
• Area 300 is a mixed zone, with a total of sixteen stands, comprising seven contact stands, including a MARS stand, for class up to D and E aircraft and nine stands for class C or lower aircraft.
• Area 400 includes one MARS stand, different stands for class C or lower aircraft, and one stand for class D or E.
• Area 500 is the general aviation apron and includes three stands for class A or B aircraft.

The entire apron is configured for pushback procedures in tractor-assisted mode.

3.2. Traffic Data

For the capacity assessment, a representative day in July 2025 was selected, as this month exhibits the peak air traffic intensity according to official Eurocontrol data. The selected day shows values slightly above the monthly average, providing a representative basis and adding a reasonable margin to the assessment of the airport operational capacity.

Table 2. Percentages of aircraft by category.

Code	%	Group %
A	1.29	6.43
B	5.14
C	85.21	85.21
D	3.86	8.36
E	4.50
F	0.00

shows the percentage distribution of the aircraft fleet by ICAO code. The aircraft were subsequently grouped into categories as the stands are not differentiated between classes A and B, nor between classes D, E, and F.

3.3. Airport Development and Hydrogen Implementation Scenario by 2035

By 2035, the aprons will be expanded and reorganized as in Figure 3. The total number of stands will be increased compared to the current scenario:

• Area 100: Addition of one MARS stand and two narrow-body stands and redesign of the narrow-body stands after the removal of one taxiway.
• Area 200: One additional MARS stand.
• Area 300: Additional stands for narrow-body aircraft in the north area of the apron, both self-maneuvering and pushback. Some terminal stands are reorganized to accommodate more wide-body aircraft.
• Area 400: Adjusted stand geometries to accommodate some wide-body aircraft.
• Area 500: The layout remains unchanged.

In the 2035 hydrogen scenario of this airport layout, 4% of total traffic (i.e., 25% of aircraft categories A and B) will be replaced with hydrogen-powered aircraft of the same class. This area is at a sufficient distance from the passenger terminal to ensure safe cryogenic operations and fuel handling. The stands in this area have longitudinal dimensions suitable for accommodating hydrogen-powered aircraft with longer fuselage than current short-range models, ensuring operational flexibility. The hydrogen-powered aircraft considered in this scenario have storage tanks located at the rear of the fuselage, near the tail. Refueling will take place on the apron 500 stands, via mobile cryogenic tanks, which will draw liquid hydrogen from storage tanks located in a dedicated area of the airport and transfer it directly to the aircraft. However, a change in the aircraft’s wingspan has not been considered at this stage, assuming that future hydrogen-powered configurations will maintain dimensions similar to those of conventional aircraft of the same category.

Traffic estimates for 2035 derive from the project forecasts [50] in

Table 3. Traffic forecast [50].

	2025	2026	2027	2028	2029	2030	2031	2032	2033	2034	2035
Passengers (millions)	11.9	12.6	13.0	13.6	14.1	14.7	15.5	16.5	17.3	17.8	18.4
Freight (tons × 1000)	64.5	65.8	67.1	68.4	69.8	71.2	72.6	74.1	75.5	77.1	78.6
Movements (×1000)	87.4	91.8	93.3	98.7	100.5	106.2	109.6	117.5	120.9	126.2	127.8

. The fleet mix was kept unchanged compared to the 2025 scenario.

3.4. Airport Development and Hydrogen Implementation Scenario by 2045

By 2045, the airport’s apron system (Figure 4) will be further expanded and reconfigured to accommodate the estimated traffic growth. The new configuration prioritizes operational flexibility and the efficiency of taxiing flows.

The main changes compared to the 2035 scenario consist of the reorganization of Area 400 to accommodate four stands for class D, E, and F aircraft, and the expansion of Area 500 to a total of nine stands. A cargo area is added on the eastern side of the airport, far from the other aprons. It consists of five stands for class D, E, and F aircraft. This area is designed for logistics and cargo operations, with direct access to the airside network.

In the 2045 hydrogen scenario for this layout, hydrogen-powered aircraft are 16% of the operating fleet. Hydrogen-powered aircraft account for approximately 75% of the total fleet in classes A and B and 15% of the class C fleet. Hydrogen-powered aircraft in classes A and B will be parked in Area 500, which remains separated and sufficiently distanced from the passenger terminal to ensure the necessary safety standards for the handling of cryogenic fuels. Hydrogen-powered aircraft of class C will be allocated in the north self-maneuvering stands of Area 300, where the same segregation and cryogenic configuration considerations apply.

Concerning the refueling methods, class A and B aircraft will be refueled using mobile cryogenic tankers, following the model described in 3.6, and class C aircraft will use a partial cryogenic hydrant system integrated into the stands. Underground pipelines will connect the hydrogen stands to the main storage tanks, whose position is not herein considered. The designated stands have adequate longitudinal dimensions for aircraft with extended fuselages. However, no changes to wingspan have been considered.

In the absence of specific forecasts for the individual airport, the traffic volume for 2045 was derived by applying an average annual growth rate of 1.2% compared to the reference traffic of 2025, in line with Eurocontrol forecasts [51]. According to this report, the average increase in IFR European flights between 2019 and 2050 is estimated at around 1.2% per year, with slightly higher dynamics for passenger and cargo connections.

3.5. AirTOP Model Construction

The creation of the model in AirTOP requires the following steps:

• Definition of all structural and operational elements in the airport: runway layout, taxiway system, apron stands, and the logical connections between them.
• Input of all simulated flights during the sample period including aircraft type, airline, arrival and departure times, destination, origin, and assigned stand.
• Use of the Eurocontrol aircraft performance database, which includes all parameters (such as maximum takeoff weight, climb and descent routes, operating speeds, fuel consumption, and engine power curves at different flight stages) required to calculate taxi times, takeoff and landing distances, climb gradients, and runway occupancy times.
• Definition of the minimum separation rules (longitudinal, lateral, and temporal) among aircraft; holding areas (in-flight waiting); ground waiting points (holding points); and sequencing zones for approach and takeoff regulation. These parameters allow the simulator to manage traffic conflicts, reduce delays, and analyze the impact of new operational configurations or infrastructure changes on the airport performance.
• Description of runway, taxiway, and apron configurations. The runways are defined by specifying operational directions, intersections, lengths, reference ICAO category, and usage rules (arrivals only, departures only, or mixed). The taxiways are represented as a network of nodes and segments, to which parameters such as maximum allowed speed, directions of travel, usage priorities, and any restrictions related to aircraft wingspan are assigned. The aprons include the stands, operational entry and exit modes (power-out or pushback), and compatibility with different aircraft classes, along with any restrictions due to wingspan and/or aircraft length, to avoid interference with adjacent areas or limit the simultaneous use of multiple stands.
• Description of Standard Instrument Departure (SID) and Standard Instrument Arrival (STAR) procedures. The simulation is set up to replicate in detail the arrival, parking, and departure operations of aircraft, to observe the effective apron capacity by varying demand conditions.

The traffic is managed by the software’s built-in Traffic Increase feature, which generates copies (“clones”) of the flights included in the base scenario, incremented with progressive values (+10%, +20%, +40%, etc.) until saturation is reached. The clones are distributed randomly throughout the operational day, with slight modifications to arrival and departure time slots, but keeping all original operational parameters (aircraft type, airline, parking duration, and stand compatibility) unchanged. In this way, each subsequent simulation cycle represents a scenario with increasing demand. The system is assumed to be at its overall capacity when at least one aircraft category can no longer be serviced by the apron. The analysis was conducted by monitoring the maximum number of hourly movements, the average stand occupancy time and its variation as demand increases, and the utilization ratio of stands.

The apron capacity study analyzed five traffic scenarios: 2025 (i.e., the current one); 2035 (i.e., a 2035 scenario with only kerosene-powered aircraft); 2035_H (i.e., a 2035 scenario with 25% of aircraft types A and B replaced by hydrogen-powered aircraft); 2045 (i.e., a 2045 scenario with only kerosene-powered aircraft); and 2045_H (i.e., a 2045 scenario with 75% of aircraft types A and B and 15% of aircraft type C replaced by hydrogen-powered aircraft).

Finally, the theoretical capacity of the apron was calculated by applying the Horonjeff analytical model [35] and then compared with the maximum value of operations managed by the apron as obtained from FTS.

3.6. Hydrogen-Powered Aircraft Operations

Several hypotheses have been proposed regarding how hydrogen-powered operations will integrate into airport environments and change air transport (

Table 4. Probability changes for airplane characteristics with hydrogen implementation [2].

	Likely to Change	Unknown Whether It Will Change	Unlikely to Change
Wingspan			x
Main gear wheel span, turn radius		x
Fuselage length	x
Turnaround time			x *
Required runway distance, taxiway geometry		x
Servicing design		x

* Will likely change in the first period of implementation and gradually decrease and return to normal or lower values.

) [2]. A NASA study [24] evaluated two options, trucks and pipelines, and concluded that pipelines represent the most economically viable long-term solution, while trucks provide a practical short-term approach during the initial deployment of hydrogen-powered aircraft. This study assumes that hydrogen is already available at the airport in its final gaseous or liquid form.

Although the fuselage is expected to be approximately 5 to 10 m longer, this constraint can be accommodated within the existing stand dimensions in the modeled airport.

However, the introduction of liquid hydrogen refueling likely implies specific turnaround challenges, including longer stand times, wider safety zones, and the need for new equipment, although Fly Zero indicates that these can be mitigated [22] in the long term. With LH₂, refueling may move onto the critical path of turnaround, and because LH₂ has lower volumetric energy density than kerosene, in the first years of implementation, refueling at equivalent hose diameter and flow rate would take longer [22]. Following these findings, hydrogen refueling was treated as a non-simultaneous activity, separated from other turnaround operations, adding from 10 to 20 min to the turnaround of hydrogen operations, depending on the refueling mode for each aircraft type.

Depending on the airport size and hydrogen traffic demand, the implementation of a full-hydrant system can occur at different time rates, with large airports capable of considering hydrants starting from 2040 and medium airports from 2045, while small airports will probably continue to use trucks [22].

In this study, different systems are used by varying aircraft sizes. For small and regional aircraft, mobile cryogenic tanks draw liquid hydrogen from storage tanks in a dedicated airport area and transfer it directly to the aircraft. In the software, this is modeled with a single truck that fuels aircraft one at a time; if an aircraft is still refueling, other hydrogen-powered aircraft should wait until the truck is available. For narrow-body aircraft, a partial cryogenic hydrant system will be modeled on one of the airport aprons. This system connects the main storage tanks to the dedicated stands via underground pipelines. By adding time to the parking duration of hydrogen-powered aircraft, we simulate the impact of these operations on capacity and delay.

4. Results

4.1. Number of Operations over Time

Figure 5a–e shows the number of on-block and off-block movements over time of the saturated run for the five scenarios (2025, 2035, 2035_H, 2045, and 2045_H, respectively), obtained through FTS. The diagrams allow identification of periods with higher operational intensity and assessment of the balance between arrival and departure phases, highlighting any saturation conditions of the apron. The red lines also indicate the apron utilization ratio (i.e., the ratio between the used and the total available stands on the apron).

Across all scenarios, Figure 5 shows a broadly stable daily distribution, with a morning peak between 07:00 and 09:00 followed by sustained activity until early afternoon, with a value oscillating between 20 and 30 hourly operations, gradually decreasing throughout the last hours of the day.

In the 2025 scenario, traffic is primarily concentrated during daytime hours, with a peak between 8:00 a.m. and 4:00 p.m. During this period, the balance between arrivals and departures is generally maintained, as evidenced by the apron utilization ratio, which reaches values close to 70–80%, without exhibiting critical saturation conditions. As the day progresses, the throughput of the apron exhibits a marked decline, leading to a gradual emptying of the apron.

In the 2035 scenario, the increase in demand leads to an almost constant increase in movements throughout the entire operating day. The utilization ratio of the apron remains elevated for an extended period, with values frequently exceeding 75%. The persistence of elevated utilization ratios reveals a diminished system resilience, particularly during morning and afternoon peak hours.

Although the shape of the graphs remains similar, a decrease in the utilization ratio is noted in the 2045 scenario. In this scenario, the system demonstrates a lower level of occupancy due to infrastructure modifications and an augmented level of stand availability. This results in a more distributed use of resources and a gradual reduction in the average utilization ratio, indicating that the apron expansion leads to a reduction in saturation conditions.

However, the introduction of hydrogen-powered aircraft modifies the throughputs (Figure 5c,e). In the 2035_H and 2045_H scenarios, a general decrease in the utilization ratio is observed, more pronounced than in the corresponding conventional scenarios. While maintaining the same daily operational profile, the presence of hydrogen-powered aircraft affects the overall system efficiency, primarily due to operational constraints related to the availability of compatible stands and longer turnaround times. As a result, apron areas are utilized less uniformly and there is a slight overall slowdown in rotations.

The trend of throughput in the different scenarios shows a consistent and easily interpretable behavior of the overall capacity of the airport system (Figure 6a–e).

In all scenarios, the throughput remains nearly constant throughout the day and across the different projections. Activity peaks always concentrate between the morning and early afternoon, and no significant increase in the number of operations processed is observed. This indicates that the airport reaches and maintains a constant level of saturation, beyond which it is unable to increase its operational capacity, regardless of the growth in demand. This is related to the operational runway capacity limits, which remain unaffected by apron changes and represent the bottleneck of the entire system.

4.2. Capacity Estimation by Simulation

Different operational configurations of the system can be distinguished based on the balance between on-block and off-block operations:

• Mixed-Mode: The number of on-block operations equals the number of off-block operations. This configuration achieves the optimal operational equilibrium condition.
• Hub-Out Mode: Combinations with a predominance of off-block operations.
• Hub-In Mode: Combinations with a higher number of on-block operations.

In the Pareto diagrams (Figure 7a–e), the pairs of on-block/off-block values represent the number of movements per hour, calculated for all simulated traffic levels.

The light blue points represent all the observed combinations, while the dark blue line marks the Pareto frontier, which is the set of configurations that maximize hourly movements while maintaining acceptable delays: they are representative of the maximum operational capacity of the airport apron [52]. In Figure 7, the point distribution and the shape of the frontier do not undergo substantial changes over the years, although the prevailing operational configuration varies. In the 2025 scenario (Figure 7a), the Pareto frontier shows balanced results, with maximum values of about 22 on-block and 23 off-block operations per hour. The slope of the curve is moderate, indicating that the airport can manage both types of movements in a balanced way without significant operational asymmetries. This condition represents a mixed-mode operation configuration. In the 2035 scenario (Figure 7b), the frontier shifts slightly to the right, indicating an increase in overall operational capacity. In this case, the slope of the frontier becomes steeper, and the points along it corresponding to mixed-mode operations no longer guarantee the maximum number of total operations. The maximum is reached in the hub-out configuration, with 12 on-block and 24 off-block movements.

In the 2045 and 2045_H scenarios (Figure 7d and Figure 7e, respectively), a trend similar to that of 2035 and 2035_H is observed. In this case, for the conventional scenario, the slope of the frontier is steeper, with a more pronounced tendency to prioritize inbound flows during peak hours, typical of a hub-in configuration.

Table 5. Apron capacity values resulting from AirTOP.

Scenario	Apron Capacity (op/h)	Mode
2025	32	Mixed
2035	36	Hub-out
2035_H	30	Mixed
2045	38	Hub-in
2045_H	36	Hub-in

summarizes the capacity values obtained from the AirTOP simulations, expressed in the operational configuration that ensures the maximum number of overall operations for the system.

4.3. Analytical Assessment of Capacity

For each scenario, the apron theoretical capacity was calculated according to Equation (1). A constant value of ${\mu }_{k,i}$ was adopted to reproduce the system’s saturation condition. The average stand occupancy time was obtained from the simulation results, analyzing turnaround times. To represent peak operating conditions, the 95th percentile of the stop time distribution for each category was used, thus avoiding underestimation and obtaining a more realistic assessment of the theoretical capacity. The red curves in Figure 8 show the trends in ${\mu }_{k,i}$ over time for each scenario; the horizontal gray band shows the typical values used for this variable [42].

In the last hours of the day, ${\mu }_{k,i}$ is not fully representative: many aircraft finish their operations and have no flights scheduled for the following day; therefore, they are removed from the simulation. This condition produces an apparent reduction in apron utilization that does not reflect actual operational availability (Figure 8). For this reason, the utilization ratios considered for the calculation of theoretical capacity according to [35] are higher than the actual average of the values recorded for the entire day.

The parameters in

Table 6. Turnaround and apron capacity values (runs with saturated apron).

Scenario	Aircraft Code	Average Turnaround		Apron Capacity
		min	h	${\mathit{N}}_{\mathbf{k}}$	${\mathit{E}\left(\mathit{T}\right)}_{\mathit{g}}$	${\mathit{C}}_{\mathit{g}\mathit{i}}$
2025	A, B	133.93	2.23	42	1.52	17.57
	C	78.74	1.31	39	1.37	18.01
	D, E, F	183.73	3.06	10	0.26	24.79
2035	A, B	221.86	3.70	46	1.95	17.65
	C	101.50	1.69	43	1.71	18.79
	D, E, F	192.06	3.20	12	0.27	33.49
2035_H	A, B	222.95	3.72	46	2.20	15.57
	C	115.88	1.93	43	1.96	16.33
	D, E, F	226.52	3.78	12	0.32	28.32
2045	A, B	216.17	3.60	60	2.22	16.72
	C	117.60	1.96	51	1.99	15.87
	D, E, F	229.16	3.82	19	0.32	36.83
2045_H	A, B	229.70	3.83	60	2.43	14.82
	C	129.38	2.16	51	2.18	14.02
	D, E, F	246.04	4.10	19	0.34	33.20

constitute the input variables employed in Equation (1). A subset of these parameters, as the turnaround time, was extracted directly from the simulation outputs. Using simulation outputs ensures that the analytical model is parameterized with realistic data, granting methodological consistency and comparability between the analytical and simulation environment.

4.4. Comparison of the Results

The comparison of the outcomes from the analytical model and those from AirTOP requires the standardization of the operation-counting criterion. In fact, the Pareto charts and the simulation provide values expressed in total operations (the sum of arrivals and departures), whereas the Horonjeff model provides capacity in terms of the number of handled aircraft. The last one should theoretically be multiplied by 2 to account for the equivalence between arrivals and departures. However, a correction factor of 1.6 was applied in this study to adopt a more conservative criterion [42]. The adoption of a correction factor lower than the theoretical value of 2 reflects the asymmetry typically observed between arrival and departure flows under real operating conditions, as well as the presence of inefficiencies such as unbalanced traffic peaks, stand idle times, and operational constraints that prevent perfect one-to-one correspondence between on-block and off-block movements. In this context, the value of 1.6 is consistent with [42] and provides a more realistic and conservative estimation of apron performance by accounting for these practical limitations and avoiding overestimation of the effective capacity.

Table 7. Apron capacity values.

Scenario	Apron Capacity
	${\mathit{C}}_{\mathit{g}}$ (Aircraft/h)	${\mathit{C}}_{\mathit{g},\mathit{o}\mathit{p}}$ (Operation/h)	FTS (Operation/h)
2025	17	28	32
2035	17	28	36
2035_H	15	24	30
2045	16	25	38
2045_H	14	22	36

compares the apron capacity calculated through the analytical Horonjeff model with the capacity from FTS. Another key metric for comparison is the utilization ratio, which is a constant parameter representing the dynamics of apron usage. It was derived from the simulation data on stand occupancy, providing an “average” value that can be incorporated into Equation (1).

In all the examined scenarios, the values obtained from the simulation are higher than the theoretical ones, confirming that the Horonjeff model provides a conservative estimate of maximum apron capacity, as it does not account for the dynamic effects associated with operational traffic management. In the 2025 scenario, the difference between theoretical and simulated capacity is approximately 2%, demonstrating that the analytical model can reflect the real behavior of the airport system. The average difference between the two approaches ranges between 10% and 20%, with greater deviations in high-demand scenarios.

These findings, however, may not be directly transferable to all airport typologies. Airports characterized by high stand utilization, limited apron space, and short-haul traffic dominance—such as major European hubs and capacity-constrained regional airports—are more likely to experience significant reductions in apron capacity due to hydrogen-related operational constraints. In such contexts, where apron saturation often precedes runway limitations, the introduction of hydrogen-powered aircraft could exacerbate stand inflexibility, increase turnaround times, and require dedicated infrastructure, thereby amplifying capacity losses compared to airports where runway throughput remains the primary bottleneck.

5. Conclusions

Hydrogen represents a promising pathway for aviation decarbonization due to its high specific energy and zero carbon emissions at the point of use, but its low volumetric density requires complex storage solutions that significantly impact aircraft design and operations. Beyond technical challenges, the widespread adoption of hydrogen is constrained by high production costs, limited availability of low-carbon supply, stringent safety and certification requirements, and long development timelines typical of the aviation sector. Consequently, while hydrogen-powered aviation is technically feasible, its large-scale implementation will depend on the gradual resolution of these challenges and the parallel development of a global refueling infrastructure, implying a long-term and incremental transition. The introduction of hydrogen-powered aircraft in existing airports forces the analysis of airside infrastructures to assess the variation in capacity due to the different operational constraints. In this study, the apron capacity of a medium-size Italian airport was calculated by analyzing the evolution of traffic from 2025 to 2045, both without and with the introduction of hydrogen-powered aircraft and modifications of the apron layout. The analysis was carried out using a fast-time simulation (FTS) and the analytical model developed by Horonjeff.

The results show good consistency between theoretical and simulated capacity, with minimal discrepancies (about 2%) in the 2025 baseline scenario that plays a central role in validating the model. The discrepancies observed in future scenarios should not be interpreted as limitations of the models, but rather as consequences of the greater uncertainty of traffic forecasts, since flight schedules and fleet mix were not modified compared to the 2025 scenario, and of the increasing operational complexity due to the progressive introduction of hydrogen-powered aircraft. In the 2035 and 2045 scenarios, the increase in demand and the increased number of stands do not lead to a corresponding increase in throughput.

The progressive introduction of hydrogen-powered aircraft further amplifies this tendency. The 2035 and 2045 hydrogen scenarios show a reduction in effective apron capacity of 16% and 6% compared to the conventional scenarios for the same year. These effects are attributable to compatibility constraints between stands and hydrogen-powered aircraft and longer refueling times, which lead to an increase in average turnaround time. However, the overall throughput does not suffer a substantial reduction, confirming that the introduction of hydrogen does not change the maximum daily capacity but reduces operational efficiency and increases the system’s sensitivity to high-traffic conditions; the apron modification in 2045_H limits the reduction in capacity. The combined analysis of the Pareto charts confirms the findings from the comparison between analytical and FTS results. In all the scenarios analyzed, the Pareto frontiers show that the progressive introduction of hydrogen-powered aircraft shrinks the frontier to lower values. This indicates a simultaneous reduction in the maximum number of operations and in apron utilization efficiency, due to the constraints imposed by longer turnaround times and selective stand compatibility. Further increases in the number of stands or local modifications to the parking areas would not yield significant throughput improvements, because the main capacity constraint for the analyzed airport is not the apron but the runway and taxiway system.

Future steps in studying the impact of hydrogen-powered aircraft on airport infrastructure and capacity could focus on different hydrogen and conventional-fuel fleet compositions and extend the investigation to multiple airports with diverse layouts and traffic profiles. Additional developments could include testing alternative traffic samples, introducing variability in turnaround times to assess the operational impact of hydrogen refueling, and incorporating delay analysis to provide a more comprehensive evaluation of system performance. Increasing the number of simulations combined with stochastic variations in turnaround times could strengthen the robustness of the analysis by capturing a wider range of operational dynamics, reducing sensitivity to deterministic assumptions. Moreover, further research should address the technological and economic uncertainties associated with hydrogen propulsion, including fuel-cell power density and durability, system costs and maturity timelines, emissions to air, and the gravimetric and volumetric performance and aircraft integration of liquid-hydrogen storage, together with the implications for airframe redesign. These aspects fall outside the infrastructure-focused scope of the present study, but their resolution will be essential to understand the operational and capacity impacts of hydrogen-powered aviation and support the long-term planning of airport systems.