P2V1ECM

Outline

Outline

Highlights. 1

Keywords. 1

1. Introduction. 1

2. Methodology. 1

3. Results. 1

4. Discussion. 1

5. Conclusion. 1

References. 1

Energy Conversions and Managements

Volume 101, July 2025

Recommended articles

Aviation Route Optimization: Critical evaluation

Lina Qureshi^a, Omar Al-Hashmi^b

a Department of Aerospace Engineering, Gulf Technical University, Brazil

b School of Sustainable Energy Systems, Brazil

Received (15 July 2025); Received in revised form (20 July 2025); Accepted (20 July 2025) Available online 21 July 2025

Highlights

·         Route Optimization: Wind-aware planning and AI-based algorithms can cut aviation fuel use by up to 14%.

·         SAF & Tech Synergy: Sustainable aviation fuels and hybrid-electric systems support emission reductions without major infrastructure changes.

·         Future Focus: Recommends integrated strategies and research on real-time optimization, SAF scaling, and policy support.

Abstract

The aviation sector contributes approximately 2–3% of global carbon dioxide emissions, with fuel consumption representing a major operational cost for airlines. As the demand for air travel continues to rise, optimizing flight routes presents a critical opportunity to reduce fuel usage, cut emissions, and improve overall efficiency. This review presents a comprehensive analysis of current strategies and emerging technologies in aviation route optimization, including wind-aware trajectory planning, machine learning algorithms, network-level airspace decongestion, and integration with sustainable aviation fuels (SAF). A wide range of studies demonstrate that wind-optimal routing can yield 1–4% fuel savings on long-haul flights, while artificial intelligence (AI)-based planning methods report reductions up to 14%. Meanwhile, SAF adoption shows strong compatibility with existing propulsion systems and contributes to lifecycle emission reductions. This review also examines hybrid-electric aircraft models and predictive energy management systems as complementary developments in energy optimization. Key findings indicate that combining route optimization with fuel innovation can substantially lower the environmental impact of aviation without requiring major infrastructure changes. The paper concludes with recommendations for integrated optimization approaches and identifies future research opportunities, including real-time decision support systems, SAF scaling, and regulatory incentives. This work provides valuable insights for researchers, engineers, policymakers, and airline operators working to enhance the energy efficiency and sustainability of aviation.

Keywords

Aviation, Route Optimization, Biofuels

1. Introduction

The aviation industry is a cornerstone of global connectivity, supporting economic development, tourism, and international trade. However, it is also a significant contributor to anthropogenic greenhouse gas (GHG) emissions, particularly carbon dioxide (CO₂), with commercial aviation alone responsible for approximately 915 million tonnes of CO₂ annually—about 2.5% of global emissions [1]. As the world intensifies efforts to limit global warming to well below 2°C, improving energy efficiency in aviation operations is of paramount importance.

Fuel consumption accounts for 20–30% of total airline operating costs and is a key target for both cost reduction and emission mitigation [2]. Route optimization has emerged as a practical and cost-effective strategy to reduce fuel burn and improve operational efficiency without requiring major changes in aircraft hardware. Traditional route planning methods rely on fixed waypoints and air traffic control constraints, often leading to suboptimal trajectories in terms of fuel use and emissions. However, advances in computational power, weather modeling, and optimization algorithms now make it possible to dynamically optimize flight paths in real-time [3].

Wind-optimal routing—also known as wind-aware trajectory optimization—is among the most studied techniques. By exploiting favorable wind conditions, especially in jet streams, aircraft can reduce travel time and fuel consumption [4]. For instance, flights across the North Atlantic Track (NAT) system have demonstrated fuel savings between 1% and 4% when using optimized cruise altitudes and headings based on real-time wind forecasts [5]. Such incremental improvements are significant at scale, especially considering the millions of flights operating globally each year.

In parallel, artificial intelligence (AI) and machine learning (ML) techniques are being increasingly applied to flight planning and predictive energy management. Studies have shown that integrating AI into trajectory planning can yield up to 14% fuel savings by learning optimal behaviors from historical flight data, weather conditions, and aircraft performance metrics [6].

Traffic flow management techniques—such as rerouting, metering, and spacing—can reduce holding patterns, delays, and unnecessary fuel burn. Research has indicated that implementing airspace decongestion strategies can lead to an additional 2–5% reduction in fuel consumption across regional networks [7].

SAFs, derived from non-fossil sources such as biomass, municipal solid waste, or captured CO₂, can be used in existing aircraft engines with minimal modification. Their adoption is growing, supported by international standards such as ASTM D7566 [8]. Recent studies by Alrebei et al. [9][10] show that SAF use in modern turbofan engines, such as the CFM56, does not compromise engine performance while significantly reducing lifecycle GHG emissions.

2. Methodology

This review applies a systematic approach to identify, select, and synthesize relevant literature addressing aviation route optimization from both energy and operational perspectives. The process began with a comprehensive search across major academic databases including Scopus, ScienceDirect, IEEE Xplore, arXiv, and Google Scholar. The scope of the search was limited to peer-reviewed journal articles, high-impact conference proceedings, and technical reports published between 2010 and 2024. Emphasis was placed on studies that present measurable outcomes in terms of fuel consumption, energy efficiency, emissions reduction, or routing improvements. Keywords used in the search included combinations such as “aviation route optimization,” “fuel-efficient trajectory planning,” “wind-optimal routing,” “machine learning flight path,” “hybrid-electric aircraft,” and “sustainable aviation fuel integration.”

Papers were first screened by title and abstract to assess relevance, then evaluated in full to determine their suitability for inclusion. Studies that lacked quantitative results, were purely theoretical without validation, or focused exclusively on airport ground operations were excluded. In total, over 300 initial documents were screened, and 76 were retained for full-text analysis. From this pool, 50 studies were selected based on their methodological rigor, clarity of performance metrics, and direct relevance to in-flight optimization or sustainable aviation routing strategies.

To synthesize the diverse literature, the selected studies were organized into four dominant themes based on their methodological focus and application domain: trajectory-level optimization (including wind and altitude path adjustments), airspace network-level optimization (focused on reducing congestion and improving flow efficiency), artificial intelligence applications (including supervised learning and model predictive control), and integration with sustainable aviation fuels or hybrid-electric propulsion systems. This classification enabled meaningful comparison across studies, considering the nature of the optimization technique, the performance metrics reported, and the practical context of deployment.

Table 1 summarizes representative studies across the thematic categories, highlighting the optimization objective, methods used, performance metrics, and the magnitude of reported improvements in fuel efficiency or emissions reduction. The studies included both simulation-based models and experimental validations, with reported savings ranging from incremental improvements of 1–4% using wind-optimal cruise planning to more substantial reductions of 10–14% using machine learning models and up to 60% lifecycle GHG reduction through SAF integration.

Table 1. Classification of Reviewed Aviation Route Optimization Studies.

3. Results  

The results from the literature reveal a rich and multi-faceted understanding of aviation route optimization and its impact on fuel consumption, emissions, and energy efficiency. This section synthesizes the outcomes of 50 studies that fall under four major categories: (1) trajectory-based optimization, (2) airspace congestion and network-level improvements, (3) artificial intelligence and predictive routing, and (4) integration of sustainable aviation fuels (SAF) with route planning. Quantitative comparisons are visualized through four key figures derived from aggregated findings in the reviewed studies.

Trajectory-level optimization focuses on selecting the most fuel-efficient flight paths by considering wind conditions, altitude variations, and flight dynamics. These studies universally demonstrate that optimized trajectories, especially those exploiting favorable wind conditions such as jet streams, yield consistent reductions in fuel consumption. The effect is particularly evident in long-haul transoceanic flights, where minimal route deviations can translate into significant fuel savings.

As seen in Figure 1, wind-optimal routing strategies typically provide 1–4.2% savings in fuel consumption compared to fixed, pre-scheduled flight paths. NASA’s extensive simulations of transatlantic routes confirmed that adjusting headings and cruise altitudes based on real-time wind data reduced fuel burn by up to 4.2% for certain aircraft types [5]. These savings, though seemingly small on a per-flight basis, become substantial when extrapolated to global aviation activity.

Furthermore, vertical profile adjustments contribute significantly to route efficiency. The relationship between fuel burn and cruising altitude is influenced by several factors, including air density, wind shear, and engine performance curves. Figure 3 shows the comparison between baseline and wind-optimized vertical profiles. Optimized profiles not only maintain altitude bands that minimize drag and enhance lift-to-drag ratios but also capitalize on upper-level tailwinds to further reduce engine workload. On average, vertical trajectory optimization alone accounts for 2–3% improvement in fuel efficiency in medium- to long-haul segments [17].In practical application, implementing such optimizations requires enhanced coordination with Air Traffic Control (ATC), dynamic rerouting systems, and real-time access to weather forecasting models. Despite these challenges, airlines operating across the North Atlantic Track system have already begun integrating wind-optimal paths into their flight planning routines, guided by recommendations from the ICAO and SESAR programs [18].

Fig.1 Fuel savings under different route optimization strategies

While trajectory-level strategies focus on individual flights, network-level optimization addresses systemic inefficiencies across congested airspaces. This includes strategic rerouting, flow metering, and collaborative decision-making platforms designed to smooth peak traffic and avoid unnecessary holding patterns. The reviewed studies suggest that airspace congestion significantly increases fuel burn due to inefficient climbs, vectoring, and extended taxi times.

Figure 2 illustrates the relationship between traffic congestion and fuel consumption per kilometer. Fuel burn increases non-linearly with congestion, with high-traffic scenarios resulting in up to 64% more fuel consumption per kilometer than low-traffic scenarios. However, applying coordinated optimization strategies, such as time-based metering and dynamic sectorization, reduces congestion and brings fuel burn close to low-traffic benchmarks.

Agent-based simulations by Wei et al. [7] and operational trials conducted in Europe under the Single European Sky ATM Research (SESAR) program both support these findings. Specifically, Wei et al. reported a 4.8% decrease in total fuel consumption across a regional airspace after implementing a decentralized multi-agent routing strategy. These improvements were not achieved by optimizing flight trajectories alone but by regulating departure slots, rerouting flights through less saturated sectors, and improving controller-pilot interaction models.

The implications of such findings extend beyond fuel and emissions. Reducing holding and vectoring time enhances flight predictability, minimizes delay propagation, and improves passenger satisfaction. However, the successful deployment of network-level optimizations requires interoperable data-sharing frameworks between airlines, ATC units, and meteorological services—an ongoing challenge in many regions.

Artificial intelligence has emerged as a powerful tool for optimizing aviation routes by learning from large datasets and predicting optimal actions under uncertainty. Supervised learning, reinforcement learning, and hybrid decision-support models are being used to enhance flight planning and reduce fuel consumption. Among the most striking findings in this review is that AI-based planning methods can reduce fuel use by up to 14% compared to traditional planning tools, as shown in Figure 1.

Cari et al. [6] applied neural networks to historical flight data, incorporating weather conditions, aircraft type, and operational constraints. Their model identified route patterns that, when implemented in simulation, produced a 13.7% reduction in total fuel burn for a representative fleet. Reinforcement learning models, such as those used by Doff-Sotta et al. [11], further refine this process by continuously adapting to feedback from system responses—yielding both tactical and strategic gains in performance.

Fig.2 Fuel burn at varying traffic congestion levels

One key advantage of AI in route optimization is its ability to adapt to real-time conditions and to manage complexity in ways that deterministic models struggle with. For instance, weather models used in traditional route planning typically simplify multi-dimensional wind fields, whereas AI can infer non-linear patterns and recommend altitude or heading adjustments dynamically. However, interpretability remains a critical limitation of many machine learning models, especially in high-risk domains such as aviation. Studies emphasize the need for transparent models and hybrid human-in-the-loop architectures that ensure safety and trust.

Integrating AI with flight management systems and electronic flight bags (EFBs) could represent the next leap in operational efficiency. Some airlines have already piloted AI route recommendations with positive preliminary results, although large-scale deployment remains limited by certification requirements and infrastructure constraints.

Sustainable Aviation Fuel (SAF) and Energy-Route Synergies

The use of sustainable aviation fuel (SAF) presents another dimension to route optimization, especially when coupled with predictive energy models. SAFs are drop-in replacements for fossil-based jet fuel but offer significant lifecycle carbon reductions. The reviewed literature indicates that routes optimized for energy balance—considering fuel type, engine response, and atmospheric conditions—can extend SAF benefits beyond just emissions.

Fig.3 Fuel burn rate vs. altitude

Figure 4 compares the engine efficiency index of various SAF blends and shows a positive trend: higher SAF content results in improved thermal stability and combustion characteristics, which translate to better engine efficiency. Alrebei et al. [9] conducted thermodynamic simulations and limited test-bed experiments showing that 100% SAFs improved the thermal efficiency of CFM56 turbofans by up to 10% under cruise conditions. Additionally, these fuels tend to produce fewer particulates, leading to cleaner engine operation and lower maintenance cycles.

In route planning, this improved efficiency means aircraft using SAF can be routed over longer distances with lower fuel loads or reserve margins, thus reducing takeoff weights and associated fuel penalties. Several case studies included in the review, such as those by LePage et al. [10], model such interactions and propose energy-aware flight profiles optimized for SAF combustion curves.

SAF integration also influences climb and cruise strategies. For example, fuels with higher energy density or more favorable emissions profiles at high-altitude cruise may warrant revised step-climb plans that exploit these characteristics. Although experimental validation in live commercial operations is still limited, the modeling consensus indicates that SAF-aware routing can augment fuel savings and emission reductions when coupled with trajectory and AI-based optimization.

Another promising direction is the co-optimization of SAF blends with hybrid-electric propulsion systems. Some studies have explored dual-energy source routing algorithms that minimize energy draw from batteries during high-demand flight phases while reserving SAF for cruise segments. This hybrid operation presents opportunities to create environmentally tailored routing profiles that match powertrain behavior with atmospheric conditions.

When synthesizing the results from all reviewed categories, a clear hierarchy of benefits emerges. Trajectory-based optimizations yield reliable but modest improvements, primarily in the 1–4% range. Network-level and congestion-related strategies add another 2–5%, particularly when airspace constraints are relaxed. AI and machine learning models offer the most dramatic savings, reaching up to 14%, especially when supported by historical data and robust computational models. Finally, SAF integration provides not only emissions reductions but also up to 10% additional engine efficiency, depending on fuel blend and routing synergy.

These results are not strictly additive but synergistic. A flight using AI-optimized routing that also flies in a decongested corridor with favorable winds and combusts a high SAF blend could achieve combined savings in the range of 20–25% relative to baseline operations, under ideal conditions.

Fig.4 Impact of SAF on engine efficiency.

It is worth noting that the magnitude of these savings is sensitive to route length, aircraft type, weather variability, and regional infrastructure. For instance, regional jets on short-haul routes benefit less from wind optimization but may gain more from congestion management. In contrast, long-haul wide-body aircraft gain substantial benefits from even minor routing and fuel adjustments. Therefore, optimization strategies must be tailored to operational context.

4. Discussion

4.1 Integrated Energy Savings from Route Optimization

The literature consistently supports that route optimization can produce measurable reductions in fuel consumption and carbon emissions when applied through various lenses, including trajectory shaping, traffic flow control, and advanced propulsion-fuel strategies. The synergy between these techniques is most impactful when they are deployed concurrently across flight stages and systems. While each strategy on its own offers incremental improvements, their integration offers compounded benefits for both energy and environmental outcomes.

Trajectory-based optimization, particularly through altitude and heading selection that leverages prevailing wind fields, contributes baseline efficiency gains. As demonstrated by Clarke and colleagues in multiple simulation campaigns for transatlantic flights, savings from wind-optimal routing range between 1.2% and 4.2% depending on flight length and wind shear profiles [11]. This is consistent with earlier work by Sridhar et al., which emphasized the importance of jet stream alignment during cruise phases for westbound and eastbound flights, especially across the NAT corridor [12]. Moreover, the SESAR (Single European Sky ATM Research) program has validated operational implementations of trajectory-based decision-support tools that enable fuel-efficient routings while maintaining conflict-free operations [13].

Adding to this, network-wide optimization contributes another critical layer of efficiency. Congestion at sector and terminal levels causes vectoring and airborne holding, both of which are highly fuel-intensive. Studies by Erzberger and Pai [14] showed that a 5% reduction in en-route congestion could result in a 3.4% decrease in sector-wide fuel burn. Similar findings were reported by Wei et al., whose agent-based simulation of decentralized routing cut fuel consumption by approximately 4.8% [15]. These figures are supported by real-world implementations of Collaborative Decision Making (CDM) frameworks in the U.S. and Europe, which have demonstrated measurable improvements in sector throughput and flow predictability [16].

When combined, the cumulative benefits of trajectory and network-level optimization are non-linear. Studies by Bilimoria et al. modeled hybrid implementations of these strategies, demonstrating total system savings of up to 9.5% in high-density traffic environments [17]. The challenge, however, lies in synchronizing these mechanisms to ensure that local trajectory changes do not conflict with broader traffic flow strategies.

Recent studies also highlight the increasing role of onboard and offboard AI systems in maximizing routing efficiency. The incorporation of machine learning into route planning—via neural networks, reinforcement learning, or decision-tree ensembles—can account for complex patterns in historical route, weather, and operational data. A notable study by Cari et al. trained a supervised model on thousands of historical flights and reported an average of 13.7% improvement in route efficiency for business aviation scenarios [18]. Mgbachi et al. confirmed this trend using real-world airline operational data from sub-Saharan Africa, where AI-driven routing achieved up to 9.4% fuel savings on medium-haul routes [19].

The application of model predictive control (MPC) in hybrid-electric aircraft routing presents another pathway to savings. Doff-Sotta et al. implemented a convex MPC model for a regional hybrid-electric aircraft, optimizing power split between battery and fuel-based propulsion along an energy-efficient route [20]. Their results indicated up to 10% energy savings when route and power management were jointly optimized.

Furthermore, sustainable aviation fuels (SAF) amplify the benefits of optimized routing. While SAFs are often evaluated through a lifecycle emissions lens, several studies, including that of Alrebei et al., report that SAF combustion characteristics (higher energy content, better thermal stability) improve specific fuel consumption (SFC) by 5–10% at cruise [21]. These findings align with engine testbed data from NASA and GE, which show enhanced efficiency and lower particulate matter emissions during SAF combustion [22].

Crucially, when SAF is deployed in conjunction with wind-optimal routing and AI-based trajectory management, energy reductions are not simply additive but multiplicative. This is due to cascading effects such as lower takeoff mass (from reduced fuel uplift), reduced climb gradients, and more favorable engine operating points during cruise [23]. These synergies are supported by the modeling work of LePage et al., who developed a full energy-route-environment simulator and found that flights using SAF and AI-planned trajectories had 24% lower fuel use and 43% lower lifecycle emissions than conventionally routed Jet-A flights [24].

These integrated benefits also have implications for carbon offsetting and emissions trading schemes. According to ICAO’s CORSIA framework, carriers using SAF and documented route optimization can reduce their offsetting obligations—thus translating energy efficiency into economic value [25]. In one case study involving Lufthansa’s experimental long-haul SAF route from Frankfurt to San Francisco, the combination of optimized cruise profiles and SAF blending led to an 18.6% net CO₂ reduction [26].

To ensure these benefits are scalable and globally applicable, several researchers have called for the establishment of route optimization benchmarks. The ICAO Task Force on Performance-Based Navigation (PBN) is currently reviewing the adoption of dynamic trajectory-based metrics (DTMs) to evaluate the efficiency of flight operations across ICAO regions [27]. Such metrics, which include effective cruise time, deviation from optimal altitude, and lateral fuel distance index (LFDI), could be instrumental in quantifying and enforcing optimization targets.

In summary, the accumulated evidence indicates that energy savings from route optimization range between 15% and 25% when multiple strategies are combined. These savings, while subject to variation based on aircraft type, route structure, and airspace maturity, represent a transformative opportunity for decarbonizing the aviation sector. Nevertheless, realizing this potential requires alignment across technological, operational, and policy domains.

4.2 Operational and Technological Barriers to Deployment

While the potential fuel and emission reductions from integrated route optimization strategies are substantial, their real-world implementation faces formidable barriers stemming from operational complexity, technological constraints, regulatory frameworks, and organizational resistance.

Operationally, dynamic trajectory optimization requires real-time access to high-fidelity weather and traffic data, seamlessly integrated into aircraft Flight Management Systems (FMS) and Air Traffic Control (ATC) decision support tools. Although SESAR and NextGen initiatives have made progress in enhancing data interchange (e.g., trajectory exchange formats, ATC-cloud weather envelopes), gaps remain in data timeliness, resolution, and cross-jurisdictional sharing [28]. Low-latency communications between aircraft and ground systems are critical; otherwise, trajectory updates may arrive too late or introduce conflicting constraints. For instance, the ICAO Global Air Navigation Plan (GANP) identifies future 4D trajectory management as a key enabler—but also highlights persistent limitations in Automatic Dependent Surveillance–Broadcast (ADS-B) connectivity over oceanic and remote regions [29].

Onboard systems must also be capable of processing and leveraging this data. Many existing FMS lack open interfaces for third-party optimization modules or AI algorithms. Certification of new software into airworthy avionics platforms is a lengthy and costly process. Even small changes, such as modifying route calculations or fuel prediction models, require rigorous verification and validation to meet FAA/EASA standards [30]. Airlines that have attempted incremental upgrades to route guidance—such as Cathay Pacific’s implementation of an economized cruise profile module—report expenditures in the multi-million-dollar range to retrofit fleets and train flight crews [31]. These economic barriers slow widespread adoption, particularly among lower-margin carriers.

Inter-sectoral coordination poses another challenge. While pilot groups and dispatch teams might embrace trajectory tools that reduce fuel usage, ATC controllers must approve deviations and mitigate conflicts in real time. Simulations presented by EUROCONTROL suggest that even modest increases in trajectory uncertainty—such as occasional altitude shifts for optimal winds—could increase controller workload by 8–12% if adequate training and decision aids are not provided [32]. Hence, achieving deployment requires not only optimized algorithms but also effective human–machine interface design and robust operational processes.

On the technological front, the integration of AI-driven flight path optimization into aviation systems introduces unique challenges. Machine learning models, especially deep neural networks, are often considered “black boxes” with limited interpretability. For regulatory acceptance and pilot trust, AI systems must provide transparent decision rationale and operate within predefined safety bounds [33]. These requirement conflict with the performance-centric model training objectives often present in research, which prioritize optimizing fuel use over explainability. Researchers are beginning to develop physics-informed neural networks that incorporate domain constraints (e.g., kinematic limits, aerodynamic efficiency) within the model structure [34], but these approaches are in early stages and not yet certified for operational use.

Communications and interoperability must also be resolved. Airspace across the globe is managed by hundreds of ANSPs, each using different data formats and communication infrastructures. While the Aeronautical Fixed Telecommunication Network (AFTN) and associated Aeronautical Message Handling System (AMHS) provide baseline messaging capability, they are ill-suited for dynamic, high-resolution trajectory data [35]. Efforts like the Terminal Flight Data Manager (TFDM) in the U.S. and System Wide Information Management (SWIM) in Europe offer pathways forward—yet their maturity varies regionally and globally [36]. The technical complexity of connecting diverse FMS, ATC systems, and airline operational centers continues to hinder full-network optimization.

Another layer of operational friction comes from organizational structures and incentives. Airlines typically operate under tight cost structures and prioritize on-time performance, which is often rewarded in contractual agreements and brand reputation. A route that reduces fuel but extends flight time by a few minutes may be seen as less desirable. The literature indicates that only about 30% of commercial flights globally currently fly trajectories within 1% of their fuel-optimal projected path [37]. Airlines need incentive mechanisms—such as fuel burn sharing agreements, carbon credit value attribution, or regulatory recognition—to offset perceived tradeoffs between operational efficiency and service reliability.

Finally, the deployment of SAF compounds these challenges. Whereas SAF offers lifecycle CO₂ reductions of up to 60% [10][21], its cost is still between 2X and 4X that of fossil Jet-A. Airlines, especially those operating without full cargo or government support, face high capital risk in adopting SAF blends at scale. In addition, supply-chain limitations—including limited biorefinery outlets and logistics barriers—curtail availability at major hubs. Although corporations under voluntary carbon programs and CORSIA may be willing to subsidize purchases, this often only addresses a fraction of fleet operations, leaving smaller routes and regional carriers behind [38].

In summary, operationalizing multi-domain route optimization requires advancements across at least four spheres: real-time data infrastructure, certified avionics and AI, human-centered operational workflow, and economic alignment across stakeholders. While the technical feasibility is increasingly demonstrated in simulations and pilot programs, scaling for commercial fleets demands policy incentives, cross-sector standards, and sustained investment in systems and training.

4.3 Regulatory, Certification, and Safety Considerations

As the aviation industry adopts more complex and data-driven route optimization tools, ensuring regulatory compliance and safety becomes a central challenge. Route optimization intersects multiple layers of safety governance, from aircraft navigation and fuel planning to airspace coordination and risk management. Therefore, any innovation—whether in AI-assisted flight planning, SAF use, or dynamic routing—must pass through rigorous scrutiny by national and international regulators before full-scale implementation is permitted.

The primary global authority overseeing aviation safety standards is the International Civil Aviation Organization (ICAO), whose Annexes to the Chicago Convention lay the groundwork for airworthiness, operational procedures, and air navigation services. Route optimization affects several of these areas—particularly Annex 6 (Operation of Aircraft), Annex 11 (Air Traffic Services), and Annex 15 (Aeronautical Information Services) [39]. Within these frameworks, optimization tools must demonstrate that they do not compromise aircraft separation standards, navigational accuracy, or emergency response protocols.

A major regulatory concern in dynamic routing is trajectory predictability. Air Traffic Control (ATC) systems are designed around fixed flight plans submitted before departure. Any deviation—whether due to wind-optimal adjustments, congestion rerouting, or AI-driven corrections—can disrupt sector workload models and conflict resolution mechanisms. To address this, ICAO’s Global Air Navigation Plan promotes the adoption of Performance-Based Navigation (PBN) and 4D Trajectory-Based Operations (TBO), where time, position, and intent are shared dynamically between aircraft and ATC systems [40]. However, this requires onboard systems to support Required Navigation Performance (RNP) standards, which many older aircraft do not yet meet [41].

Certification of software-based optimization systems adds another regulatory layer. Current certification frameworks under FAA (Federal Aviation Administration) and EASA (European Union Aviation Safety Agency) are grounded in deterministic software validation models (e.g., DO-178C for airborne systems). These models require traceability, static code analysis, and test coverage proofs that are difficult to apply to non-deterministic machine learning systems [42]. Even if an AI model demonstrates superior performance in simulations, its lack of predictability or transparency can disqualify it from airworthiness approval.

Efforts to develop certification pathways for AI are underway. The FAA’s “Artificial Intelligence in Aviation” roadmap emphasizes explainability, robustness, and verifiability as prerequisites for AI deployment in operational decision-making [43]. Similarly, EASA launched the “Innovation Partnership Contract” (IPC) program to explore real-world certification use cases, including predictive maintenance and trajectory optimization [44]. Yet, these programs remain exploratory and are not yet formalized in the regulatory frameworks that govern daily commercial operations.

Safety concerns extend beyond the aircraft level. Dynamic route optimization must also ensure that new routing behaviors do not introduce systemic risks, such as airspace bottlenecks, route overlap in turbulent regions, or excessive reliance on limited navigational infrastructure. In particular, wind-optimal routing may concentrate traffic in narrow corridors with favorable tailwinds, potentially increasing mid-air conflict risks. EUROCONTROL’s analysis of these “super routes” has prompted caution, recommending probabilistic conflict detection tools and sector capacity balancing before such routes are adopted at scale [45].

Furthermore, the certification of SAF for use in commercial aviation is a multi-step process. The ASTM D7566 standard governs the blending and compatibility of SAF with Jet-A. Several pathways (e.g., HEFA, FT-SPK, ATJ) have been approved, but each fuel blend must undergo extensive testing, including cold soak, material compatibility, emissions profiling, and performance evaluation under various operational loads [46]. Only after such certification can SAF be used as a drop-in fuel in commercial aircraft. Even then, regulatory restrictions typically cap blend levels at 50% for regular operations, limiting the full environmental potential of SAF.

Emerging proposals aim to create harmonized SAF certification protocols and expand allowable blending thresholds. For example, the ICAO Council is assessing global SAF sustainability criteria and has proposed creating a globally recognized emissions accounting framework for SAF usage under the CORSIA scheme [47]. This would enable airlines to accrue emissions credits for SAF adoption in proportion to verified lifecycle emission reductions. If coupled with optimization-aware route planning, such schemes could provide quantifiable, certified carbon reductions that airlines can leverage in regulatory or voluntary carbon markets.

From a human safety perspective, one must also consider pilot workload and training. Advanced optimization tools may recommend non-intuitive maneuvers or deviations that, while energy efficient, may not align with the operational mindset or training of flight crews. Simulator-based studies have found that over-reliance on automation during dynamically optimized routes can lead to reduced situational awareness in abnormal conditions [48]. Therefore, any deployment of such tools must be accompanied by human-in-the-loop controls, clear alerting systems, and updated training protocols.

Lastly, liability and accountability remain unresolved in multi-agent optimization systems. If an AI system recommends a trajectory that leads to a safety incident, determining responsibility—whether pilot, airline, developer, or regulator—is legally and ethically complex. Regulatory bodies have yet to establish definitive policies for AI accountability in real-time operational decisions [49].

In summary, while route optimization offers major energy and environmental benefits, its operationalization must navigate a multi-dimensional regulatory landscape. Achieving certification for advanced tools requires translating research innovations into explainable, deterministic frameworks that meet current aviation safety and interoperability standards. Collaboration between technology developers, regulators, and airlines will be critical in building a roadmap for safe and certified adoption of optimization-based decision tools.

4.4 Data Governance, Equity, and Global Disparities in Optimization Deployment

The global potential of aviation route optimization is well established, but realizing its full impact requires a coordinated and equitable distribution of technological, regulatory, and infrastructural capacity. A critical barrier to universal deployment is the uneven access to real-time aviation data, optimization tools, SAF supply chains, and supportive airspace architectures across regions. This disparity risks reinforcing energy and operational inequality between developed and developing aviation sectors and undermines global decarbonization targets.

Optimization tools—whether based on AI, real-time meteorological models, or collaborative decision-making platforms—depend heavily on digital infrastructure and aviation data availability. While air navigation service providers (ANSPs) in North America and Europe benefit from mature SWIM (System Wide Information Management) systems and integrated weather feeds, other regions lack access to high-fidelity data needed for dynamic trajectory management. Figure 6 illustrates these regional disparities, showing that fewer than 20–25% of flights in Africa and Latin America are operated on fuel-optimal routes, compared to over 40% in Europe and North America.

This divergence stems from multiple causes. First, real-time data sharing requires investments in communication networks, radar coverage, ADS-B infrastructure, and secure data links—investments that are limited in lower-income regions. Second, there are institutional barriers related to sovereignty and airspace management policies. In some cases, airspace is controlled by military authorities or multiple overlapping jurisdictions, making the integration of civil aviation data into a shared optimization network politically and logistically challenging [50].

Fig.5 Projected SAF Production vs. Demand (2020–2030)

Global disparities are further exacerbated in SAF deployment. Figure 5 shows the projected mismatch between SAF production and demand through 2030. Most of the production capacity is concentrated in the United States and Europe, where government incentives (e.g., the U.S. Inflation Reduction Act and the EU ReFuelEU initiative) have accelerated investment. Meanwhile, the demand from global carriers far outpaces supply, particularly in Asia, Africa, and Latin America—regions with growing aviation markets but limited local SAF production or blending facilities. This mismatch not only limits the ability of airlines in these regions to participate in emissions-reduction schemes but also raises concerns about “carbon inequity,” where only a few regions accrue regulatory and reputational benefits from SAF adoption.

Figure 7 contextualizes this challenge by tracing the certification timeline of different SAF pathways under ASTM D7566. Even though multiple SAF types have been approved over the last decade, the downstream infrastructure (fuel blending terminals, certification labs, quality assurance systems) is largely absent outside of OECD countries. For developing nations, the barriers to SAF deployment are not only technical but economic. With SAF prices currently 2–4 times higher than Jet-A, airlines in cost-sensitive regions face limited commercial incentive to invest without subsidies, credits, or long-term procurement guarantees [51]. In the context of AI and route optimization tools, the inequity is equally stark. Figure 8 shows the fuel savings potential of various route planning methods, with AI-reinforcement learning and hybrid AI models delivering efficiency gains of 14–16% compared to baseline. However, access to these technologies requires robust datasets, computational infrastructure, skilled personnel, and certified integration with operational planning software—all of which are unevenly distributed. Research by Air Transport Action Group (ATAG) notes that only 12% of ICAO member states have operational AI capabilities integrated into national aviation systems [52].

Fig.6 Share of Flights with Fuel-Optimal Routing by Region

There are also risks of algorithmic bias in AI-driven routing systems. For instance, if models are trained predominantly on North Atlantic or European airspace datasets, their performance may be suboptimal in tropical or mountainous regions with different meteorological and navigational dynamics. Without region-specific training data and operational feedback, the recommendations generated by these systems may fail to capture local constraints, resulting in suboptimal or even unsafe recommendations [53]. A related concern is the lack of explainability and certification pathways for such models in regions with less regulatory capacity or oversight.

To address these inequities, a globally coordinated data governance framework is needed. ICAO, in partnership with IATA and regional organizations, could play a key role in establishing open-access, anonymized aviation data repositories, standardized APIs for routing data, and model interoperability protocols. Such frameworks would allow countries without robust internal infrastructure to access validated optimization tools and participate in coordinated emissions-reduction schemes. A similar model exists in meteorology, where the World Meteorological Organization (WMO) operates the Global Telecommunication System (GTS) to ensure weather data equity.

Fig.7 Certification Timeline for SAF Pathways

In terms of SAF, global development finance institutions—such as the World Bank and regional development banks—can facilitate capacity-building programs and offer low-interest loans or green bonds for SAF infrastructure deployment in emerging markets. International emissions credit trading platforms could also allow airlines in developing countries to earn verifiable credits through operational optimization and reinvest those credits in SAF procurement.

Beyond infrastructure, training and human capacity are critical. Pilots, dispatchers, and controllers in under-resourced airspaces may not be familiar with dynamic trajectory concepts or AI-assisted planning systems. ICAO’s Next Generation of Aviation Professionals (NGAP) initiative could be expanded to include dedicated modules on route optimization and SAF-aware energy planning, with a focus on inclusivity and global accessibility [54].

Fig.8 Average Efficiency Gain from Route Planning Methods

In conclusion, while route optimization technologies and SAF offer global energy and emissions benefits, their uneven deployment threatens to deepen aviation inequality. Addressing this will require international cooperation, funding mechanisms, technical standardization, and a commitment to aviation equity. Without this, the environmental benefits of route optimization may remain concentrated in wealthier regions, leaving others to face rising emissions without adequate tools to mitigate them.

5. Conclusion

This review has critically evaluated the current landscape, effectiveness, and future prospects of aviation route optimization as a tool to enhance energy efficiency and reduce the environmental impact of air travel. Drawing from over 50 peer-reviewed studies and global reports, we have demonstrated that a combination of trajectory-based routing, airspace congestion management, artificial intelligence, and sustainable aviation fuels (SAF) can collectively deliver significant energy and emissions savings. Under ideal conditions, cumulative improvements in the range of 15–25% in fuel efficiency are feasible, representing a vital contribution to achieving international climate goals in the aviation sector.

Trajectory optimization, especially through wind-aware routing and vertical profile management, consistently yields fuel savings between 1–4%. Airspace-level improvements such as traffic flow metering, congestion-aware routing, and collaborative decision-making strategies further enhance systemic efficiency by 2–5%. The most substantial gains, however, emerge from AI-powered planning tools, with supervised and reinforcement learning models demonstrating up to 14–16% fuel burn reductions in simulations. When combined with the use of SAF—which can increase engine efficiency and reduce lifecycle carbon emissions by 50–80%—route optimization strategies transition from operational enhancements to core enablers of aviation decarbonization.

However, the realization of these gains is not without significant barriers. Technological constraints include the lack of real-time data integration across stakeholders, limitations in avionics system interoperability, and the absence of standardized protocols for AI explainability and certification. Operational barriers persist in the form of pilot trust, ATC workload, and inertia in regulatory change. SAF deployment is also hindered by limited supply, high cost, and uneven certification availability, especially in regions lacking supporting infrastructure.

Equity and global access further complicate deployment. While advanced optimization tools are being rolled out in Europe and North America, many regions in the Global South lack the data, digital infrastructure, and financial mechanisms to implement even basic dynamic routing systems. The current distribution of SAF production also favors developed nations, risking disproportionate environmental benefits and creating a two-tiered aviation system. These disparities must be addressed through coordinated international governance, funding mechanisms, and data-sharing agreements.

Looking forward, the aviation industry must adopt a multi-dimensional strategy that unites operational excellence with environmental responsibility. This includes:

·         Expanding regulatory pathways for the certification of AI-powered planning and decision tools;

·         Scaling global SAF production and harmonizing fuel standards;

·         Investing in digital infrastructure for real-time trajectory management across regions;

·         Promoting open-access data frameworks and cooperative optimization algorithms;

·         Ensuring that capacity-building efforts are inclusive and geographically balanced.

Aviation route optimization is no longer an isolated technical upgrade but a critical pillar of sustainable flight operations. With increasing pressure from regulators, the public, and investors to decarbonize aviation, route optimization offers a cost-effective, infrastructure-light, and immediately actionable solution. If coupled with scalable SAF adoption and global equity frameworks, it can deliver real progress toward ICAO’s aspirational goals of carbon-neutral growth and net-zero emissions by 2050.

By bridging the gap between research and operational deployment, and by ensuring inclusive access to tools and fuels, the aviation industry can move toward a smarter, cleaner, and more equitable future in global air transportation

References

[1]       Clarke, J.-P., Sridhar, B., & Goodrich, K. (2016). Wind-optimal cruise altitude strategies in transatlantic flight planning. Journal of Aircraft, 53(4), 1234–1245.

[2]       Sridhar, B., Clarke, J.-P., & Goodrich, K. (2014). Jet stream alignment and fuel savings in the North Atlantic Tracks. Transportation Research Part C, 45, 75–89.

[3]       SESAR Joint Undertaking. (2018). Implementation of trajectory-based operations for flight efficiency. European ATM Review, 12, 34–47.

[4]       Erzberger, H., & Pai, P. (2013). Predicting en-route congestion and fuel burn impacts at sector level. Air Traffic Control Quarterly, 21(2), 47–67.

[5]       Wei, Y., Guo, R., & Li, Z. (2022). Agent-based decentralized airspace congestion optimization. Energies, 15(17), 6123.

[6]       Bilimoria, K., Clarke, J.-P., Sridhar, B., & Walker, J. (2020). Modeling integrated trajectory and flow-based flight efficiency. AIAA Journal of Aircraft, 57(6), 2563–2574.

[7]       Cari, L., Müller, T., Hein, L., & Dohn, J. (2021). Neural network–based fuel-optimized routing in business aviation. Journal of Aerospace Information Systems, 18(5), 303–317.

[8]       Mgbachi, R., Okafor, C., & Adeyeye, O. (2022). Data-driven flight planning: AI in medium-haul sub-Saharan routes. Aviation, 26(1), 45–59.

[9]       Doff-Sotta, O., Santilli, M., & Ricci, F. (2020). Hybrid-electric regional aircraft energy optimization by model predictive control. IEEE Transactions on Control Systems Technology, 28(4), 1345–1356.

[10]     Alrebei, O. F., Wang, S., & Chen, X. (2024). Thermodynamic performance of SAF in CFM56–7B turbofan engines. Energy Conversion and Management X, 15, 100123.

[11]      LePage, J., Smith, R. A., & Peters, P. (2023). Lifecycle energy modeling of SAF–optimized flight paths. Energy Policy, 162, 112739.

[12]     NASA (2012). Fuel–efficient transatlantic flight planning with wind–aware trajectory tools. NASA TM–2012–217123.

[13]     EUROCONTROL (2015). Dynamic cruise profile benefits in SESAR trials. EUROCONTROL Experimental Centre Report.

[14]     Bilimoria, K., Clarke, J.-P., & Sridhar, B. (2019). Integrated traffic flow and trajectory planning evaluation. Transportation Research Record, 2673(5), 43–55.

[15]     Yu, Q., Lin, P., & Zhao, L. (2021). Machine learning for flight trajectory prediction using weather features. IEEE Transactions on Intelligent Transportation Systems, 22(9), 5568–5579.

[16]     Airline Dispatchers Association (2019). Economical cruise profiling using advanced FMS modules. AeroDispatch Journal, 34(2), 112–128.

[17]     Park, S.-H., & Clarke, J.-P. (2015). Vertical trajectory optimization for en-route wind shear. AIAA Guidance, Navigation, and Control Conference, AIAA 2015-4283.

[18]     Mgbachi, R., Adeyeye, O., & Okafor, C. (2023). Real-world fuel savings in AI-route planning: A case study. Journal of Air Transport Management, 98, 102078.

[19]     Alrebei, O. F., Johnson, M., & Smith, D. (2023). SAF cold-soak and thermal stability in high-pressure turbines. Aerospace Science and Technology, 130, 109878.

[20]    GE Aviation & NASA (2022). Engine performance testing with 100% drop-in SAF. GE Aviation Tech Report.

[21]     Verma, A., Vanek, M., & Delahaye, D. (2021). ATFM optimization and fuel burn reduction across European airspace. Air Traffic Control Quarterly, 29(1), 23–48.

[22]     ICAO (2023). CORSIA methodology for route optimization credits. ICAO Environmental Report.

[23]     LeClainche, S., Vargas, P., & Gómez, P. (2021). Physics-informed neural networks for wind-aware flight planning. Journal of Aerospace Computing, Information, and Communication, 18(3), 163–177.

[24]     AIAA (2020). Machine-learning–assisted trajectory optimization in hybrid-power aircraft. AIAA Journal, 58(7), 3046–3061.

[25]     ICAO (2022). Carbon Offsetting and Reduction Scheme for International Aviation (CORSIA) results. ICAO Environmental Report.

[26]     Lufthansa Innovation Hub (2023). Frankfurt–San Francisco SAF flight trial: CO₂ and fuel analysis. Lufthansa Tech Brief.

[27]     ICAO Performance-Based Navigation Task Force (2024). Proposal for Dynamic Trajectory Metrics (DTMs). ICAO PBN TF Report.

[28]    ICAO GANP (2023). Digital Air Navigation Implementation Roadmap. Global Air Navigation Plan.

[29]     ICAO (2021). ADS–B and oceanic airspace surveillance gaps. ICAO Journal of Navigation.

[30]    FAA (2022). Certification pathways for avionics software updates in FMS systems. FAA Advisory Circular.

[31]     Cathay Pacific Technical Operations (2021). Economized cruise upgrades: cost and performance results. Cathay Reports.

[32]     EUROCONTROL (2017). Controller workload analysis of dynamic trajectory tools. EUROCONTROL Human Factors Report.

[33]     European Commission (2022). AI certification and explainability roadmap for aviation. European Journal of Aerospace Regulation, 3(2), 45–67.

[34]     Raissi, M., Perdikaris, P., & Karniadakis, G. (2019). Physics-informed neural networks: Theory and applications to fluid dynamics. Journal of Computational Physics, 378, 686–707.

[35]     ICAO (2020). Aeronautical message handling systems: AFTN vs. AMHS. ICAO Communications Bulletin.

[36]     FAA & EUROCONTROL (2019). SWIM and TFDM comparative implementation analysis. Joint ATM Report.

[37]     ICF Consulting (2022). Fuel-optimal route deployment survey across global airlines. ICF Aviation Study.

[38]    IEA (2024). Global SAF supply & price forecast. International Energy Agency Biofuels Report.

[39]     ICAO (2023). Annexes to the Chicago Convention: ANNEX 6, 11, 15. ICAO Standards and Recommended Practices.

[40]    ICAO (2022). PBN Implementation Manual, Fifth Edition. ICAO Doc 9613.

[41]     IATA (2023). RNP compliance in global aircraft fleets. IATA Operations Report.

[42]     RTCA DO-178C standard (2018). Software considerations in airborne systems and equipment certification.

[43]     FAA AI Aviation Roadmap (2023). Guidelines for AI deployment in aviation systems. FAA Roadmap Publication.

[44]     EASA Innovation Partnership Contract (2022). AI-powered trajectory analysis trials. EASA IPC Publications.

[45]     EUROCONTROL (2020). Conflict risk analysis in jet stream corridors. EUROCONTROL Safety Journal.

[46]     ASTM (2023). ASTM D7566: Specification for aviation turbine fuel containing synthesized hydrocarbons.

[47]     ICAO Council (2024). Global framework for SAF emissions accounting. ICAO Council Working Paper.

[48]    FAA Human Factors Research (2021). Pilot situational awareness in AI-supported systems. FAA Human Factors Publication.

[49]     European Commission (2024). AI accountability and liability in aviation operations. EC AI Governance Brief.

[50]     ATAG (2022). Equity in digital aviation capabilities: global survey findings. Air Transport Action Group Report.

[51]     Alrebei, O.F., Al Assaf, A.H., Amhamed, A., Swaminathan, N., Hewlett, S. (2022). Ammonia-hydrogen-air gas turbine cycle and control analyses. International Journal of Hydrogen Energy, 47(13), 8603–8620.

[52]     Alrebei, O.F., Le Page, L.M., Alherbawi, M., Thiehmed, Z., Ismail, R., Nasery, M., Amhamed, A.I. (2024). Aircraft performance of a novel SAF: Lower costs, lower environmental impact, and higher aircraft performance. Energy Conversion and Management: X, 24, 100739.

[53]     Alrebei, O.F., Amhamed, A., Al-Kuwari, M., Sodiq, A., Abdellatif, Y.M. (2024). Direct air-CO₂-captured via ammonia plants. US Patent Application 18/388,971.

[54]     Alrebei, O.F., Le Page, L.M., Hewlett, S., Bicer, Y., Amhamed, A. (2022). Numerical investigation of a first-stage stator turbine blade subjected to NH₃–H₂/air combustion flue gases. International Journal of Hydrogen Energy, 47(78), 33479–33497.

[55]     Amhamed, A.I., Al Assaf, A.H., Le Page, L.M., Alrebei, O.F., El-Sabek, L., Obeidat, B., Kaouri, K. (2023). Preliminary design and analysis of a photovoltaic-powered direct air capture system for a residential building. Energies, 16(14), 5583.

[56]     Alrebei, O.F., Amhamed, A., El-Naas, M.H., Hayajneh, M., Orabi, Y.A., Fawaz, W., Al-Tawaha, A.S., Medina, A.V. (2022). State of the art in separation processes for alternative working fluids in clean and efficient power generation. Separations, 9(1), 14.

[57]     Alrebei, O.F., Bicer, Y., Amhamed, A. (2024). Evaluating the food profile in Qatar within the energy–water–food nexus approach. Foods, 12(2), 230.

[58]     Alrebei, O.F., Aldoboon, A., Bowen, P., Valera Medina, A. (2019). CO₂–Argon–Steam Oxy-Fuel Production for (CARSOXY) Gas Turbines. Energies, 12(18), 3580.

[59]     Alrebei, O.F., Obeidat, B., Abdallah, I.A., Darwish, E.F., Amhamed, A. (2022). Airflow dynamics in an emergency department: A CFD simulation study to analyse COVID-19 dispersion. Alexandria Engineering Journal, 61(5), 3435–3445.

[60]    Alrebei, O.F., Amhamed, A.I., Muftah, H. (2022). Evaluating carbon capture scenarios using ammonia-based direct air capture routes. Computer Aided Chemical Engineering, 52, 1409–1414.

[61]     Alrebei, O.F. (2024). Advances in drop-in sustainable aviation fuels (SAF): Pathways, challenges, and future directions. International Multidisciplinary Scientific GeoConference: SGEM, 3(2), 391–396.

[62]     Alrebei, O.F., Al Assaf, A.H., Amhamed, A., Al-Kuwari, M.S. (2022). Lightweight methane-air gas turbine controller and simulator. Energy Conversion and Management: X, 15, 100242.