# Statistics and Machine Learning in Aviation Environmental Impact Analysis: A Survey of Recent Progress

^{*}

^{†}

## Abstract

**:**

## 1. Introduction

**Summary of methodology**: On the summary of methods from statistics and machine learning, the emphasis is to present a comprehensive landscape and development process for each field. There are many excellent textbooks and review papers in the literature which introduce the mathematical foundation, detailed algorithms, and experimental analysis of of the these methods. The summary of methodology in this paper is not an attempt to replicate or elevate those existing methodology-oriented review papers. Instead, from an engineering researcher’s perspective, the aim of this part is to clearly convey the basic ideas in the methods and the differences between them such that applied researchers can have a clearer big picture.**Organization of representative works**: Most similar survey papers focusing on other application areas group relevant works in the literature by the type of method used. In that manner, the existing literature on aviation environmental impact analysis would be divided into, for example, the applications of regression analysis, clustering, dimensionality reduction, feature selection, neural networks, etc. In our approach, representative works in the literature are grouped by the purpose of applying statistical and machine learning methods, i.e., for what reasons were these method used to tackle different problems. Thus, the main section of this paper is organized into seven themes:**data reduction**,**efficient computation**,**predictive modeling**,**uncertainty quantification**,**pattern discovery**,**verification and validation**, and**infrastructure and tools**. For each theme, we present both the necessary background and the representative papers.**Diversity**: This paper is by no means an exhaustive list of every relevant work in this area. The overarching objective is to summarize the overall research trends through representative works/projects. Under the premise that each selected work has enough quality and correct scope, we hope to present a diverse research portfolio which covers different methods, different application directions, and even different regions in the world (although with lower priority than the previous two aspects). For example, on the methodology side we cover from basic statistical analysis and regression models to unsupervised learning approaches such as clustering and dimensionality reduction, different types of neural networks (ordinary, convolutional, recurrent), and graphical model. On the application side we cover the modeling of fuel burn, emissions, and noise, for fixed-wing aircraft, helicopter, airport, and air transportation system. The selection range also covers works from different entities and regions to reflect the fact that sustainable aviation is a global effort.

**Optimization**: There are three facets of analytics: descriptive, predictive, and prescriptive analytics. On the methodology side we only cover the former two facets of analytics. Optimization is at the kernel of prescriptive analytics. Although it also belongs to data-driven approaches, it is not covered here simply because it is a rich area that is worthy of an independent survey paper. Optimization methods have been used to design aircraft operations that reduce environmental impacts. A sample of such works include [9,10,11,12]. Reference [13] is a recent survey paper on climate optimal aircraft trajectory planning.**Aircraft design**: Since sustainable aviation became a major research area in the aerospace community, novel methods in aircraft design and Multidisciplinary Design Optimization (MDO) have started to incorporate environmental considerations into aircraft conceptual and preliminary design phases. An early work of this type [14] dates back to almost two decades ago. Examples of some more recent works include [15,16,17].**Physics-based methods**: Under the category of efficient and accurate modeling of aviation environmental impacts, some recent progresses/capabilities are physics-based which do not involve much data-driven components discussed in this paper. This type of approaches is also a crucial and indispensable part of aviation environmental impact modeling. Interested readers can refer to [18,19,20,21] as starting point.

## 2. A Brief Overview of Methods from Statistics and Machine Learning

#### 2.1. Statistical Methods

#### 2.2. Machine Learning Methods

**Representation**: In the first step, a learner must be represented in a format for computer to handle. Selecting a set of representations for a learner forms the hypothesis space of the learner. A learner cannot be learned if it is not in the hypothesis space.**Evaluation**: An evaluation function, also referred to as the objective function, is needed to distinguish good learners from bad ones. The construction of the objective function must consider issues in optimization such that it may differ from the direct objective one wants to optimize.**Optimization**: An optimization method searches through the space of possible hypotheses for one with the best performance. The choice of optimization method is key to both the efficiency and efficacy of the learner. Table 1 includes typical examples of each of the three components.

**Table 1.**Examples of the three components of learning algorithm (Original structure from [30]).

Component | Examples |
---|---|

Representation | Instance-based: k-nearest Neighbor, Support Vector Machines |

Hyperplanes: Naive Bayes, Logistic Regression | |

Trees-based: Classification and Regression Trees, Boosted Trees | |

Rule-based: Association Rules | |

Neural Networks: Artificial Neural Networks | |

Graphical Models: Bayesian Networks, Markov Random Fields | |

Evaluation | Mean Squared Error, Likelihood, ${R}^{2}$ |

Accuracy, Precision, Recall | |

Mutual Information, Homogeneity | |

Posterior Probability, K-L Divergence, Cost/Utility | |

Optimization | Discrete: Greedy Search, Branch-and-bound, Beam Search |

Continuous (Unconstrained): Gradient Descent, Newton’s Method | |

Continuous (Constrained): Linear Programming, Augmented Lagrangian |

## 3. The Main Application Themes

#### 3.1. Data Reduction

#### 3.2. Efficient Computation

#### 3.3. Predictive Modeling

#### 3.4. Uncertainty Quantification

**Inputs uncertainty**: The inputs of a model/system may have inherent uncertainty and substantial variation around a deterministic value.**Model uncertainty**: All models are “wrong” because they inevitably include assumptions, approximations, and errors and are therefore not exact representations of reality. Two aspects of uncertainties related to model are model-form uncertainty and uncertainty about parameters within the model.**Computational and numerical uncertainty**: Normally numerical errors from running simulations or solving mathematical models, including simplified equations, convergence error, truncation, etc.**Physical testing uncertainty**: A result of uncontrolled or unknown inputs, measurement errors, and limitations in the design and implementation of tests.

#### 3.5. Pattern Discovery

**Descriptive patterns**: The identification of these patterns usually do not involve advanced algorithms. They are obtained through descriptive statistics or sometimes the direct results of data collection.**Associative patterns**: These patterns are mainly about co-occurring phenomena. A typical statement of associative pattern is: “If A happens, then B is also likely to happen”.**Periodic patterns**: These patterns repeat themselves with a specific period, which can be found in time series data, sequence data, and spatiotemporal data.**Structural patterns**: These patterns are extracted summary information represented in terms of a structure that can be reasoned about. There are different structural forms such as graphs, trees, sets, clusters, etc.**Abnormal patterns**: A substantial divergence from normal behaviour is considered abnormal. These abnormalities could be signals of risk or opportunities for novel discoveries.

#### 3.6. Verification and Validation

#### 3.7. Infrastructure and Tools

## 4. Future Opportunities

**advanced statistical modeling and data mining**,

**physics-informed learning**,

**explainable/interpretable models**,

**Bayesian methods**, and

**data-driven optimization**. Some of these methodologies are already mature in their respective fields or in other application domains, yet their applications in aviation environmental impact analysis have been limited so far. For some research opportunities, one can see the appearance of progress in some latest advances; for other research opportunities, they are proposed in a more speculative manner.

#### 4.1. Future Opportunity 1: Advanced Statistical Modeling and Data Mining

**High-dimensional data analysis**: Dataset size n and dimension p are two primary indicators to choose among data analysis frameworks. Many real-world aviation datasets are high-dimensional in nature. For such datasets with a large number of attributes, traditional statistical theories and methodologies are inadequate and can break down in unexpected ways. A main challenge here is the Curse of Dimensionality (CoD) [107], which refers to a set of phenomena and challenges that do not normally occur in low-dimensional spaces yet arise when the data has too many attributes/features. Modern advances in high-dimensional data analysis can perform statistical inference and prediction in high-dimensional settings. A key assumption behind most such analyses is that high-dimensional data typically concentrates on low-dimensional, sparse, or degenerate structures. Dimensionality reduction is a common way to transform high-dimensional data into a lower dimensional representation while preserving the intrinsic properties of the data. The other two categories of methods that can find applications in aviation environmental impact analysis are Functional Data Analysis (FDA) and tensor data analysis. FDA [108,109] deals with the analysis and theory of data that are in the form of functions or curves. FDA can also be thought of as the statistical analysis of samples of curves and surfaces. With the deployment of big data technologies, more and more aviation data are being recorded continuously during a time interval or intermittently at discrete time points. Section 3 highlights the use of flight operation and performance data, a typical example of functional data, for accurate environmental impact analysis. Some popular FDA techniques include Functional Principal Component Analysis (FPCA), functional regression, and clustering/classification of functional data. Tensor data in the form of multi-dimensional array can be found in the analysis of for example image streams, or aircraft noise or emissions data measured at different locations in a two-dimensional plane (two-dimensional data) sampled over different times (the third dimension, leading to three-dimensional data). Tensor decomposition [110] techniques can be applied to process and analyze tensor data.**Spatio-temporal data analysis**: Some representative works in the literature have started to explore the spatial, temporal, and spatio-temporal patterns of aviation emissions and noise. Because aviation environmental impacts have inherently spatial or temporal context, the modeling process must take into account the space and/or time component to better understand and interpret the data. Spatio-temporal data differ from relational data in that both spatial and temporal attributes are available in addition to the actual measurements/attributes, which introduces additional challenges and requires novel formulations to analyze. Of note, References [111,112] are two good references for the statistics and data mining for spatio-temporal data. Temporal data analysis applies to events ordered by one or more dimensions of time [113]. Within temporal data analysis, the discovery of similar patterns within the same time sequence or among different time sequences relies on time series analysis—an active research field of statistics. On the other hand, spatial statistics [114] provides techniques and tools to analyze data that has a spatial characteristic to it. Since the future of aviation is likely to incorporate emerging components such as Urban Air Mobility (UAM) and Unmanned Aerial Vehicles (UAV), research topics at the intersection of aeronautics and urban and regional studies can contribute to the integration of such disruptive concepts into the existing transportation system. This includes the design of UAM and UAV operations with maximum efficiency and minimum societal (including environmental) impacts. Methods from spatial statistics can play a pivotal role in this interdisciplinary area.

#### 4.2. Future Opportunity 2: Physics-Informed Learning

**Greater physical consistency**: Purely data-driven models may fit training data very well, but predictions may be physically inconsistent or implausible. Through integrating governing physical laws in the learning process, the model produce predictions that respect the underlying physical principles.**Improved trainability**: Physics-informed learning can find meaningful solutions even when the problem is not perfectly well posed—with incomplete models and incomplete data. Specific to physics-informed learning, there are also effective ways to accelerate training.**Better generalization**: Normal deep learning methods require big data for training, which may not be available for problems in science and engineering. Physics-informed learning performs well in the small data regime and has strong generalization capability from small data.**Uncertainty quantification**: There are multiple ways of quantifying the uncertainties due to physics, data, and learning models. One such example is Bayesian PINNs (B-PINNs), which integrates the Bayesian approach with physics-informed learning for uncertainty quantification.

#### 4.3. Future Opportunity 3: Explainable/Interpretable Models

#### 4.4. Future Opportunity 4: Bayesian Methods

**Combination of expert knowledge and data**: in some problems, people seek approaches which can combine both SME knowledge/opinions and collected data to make better decisions. Bayesian methods incorporate background information, knowledge, or beliefs into the modeling process through prior elicitation—the translation of background information into a suitable prior distribution. Common strategies for prior elicitation include asking an expert or a panel of experts for judgements, or analyzing historical data. The result from Bayesian modelling (posterior) can also be regarded as a compromise or balancing between the prior knowledge (prior) and the observed data (likelihood).**Uncertainty quantification**: Bayesian methods are a natural fit for uncertainty quantification. When a Bayesian framework is used for model fitting, probability distributions are assigned to the model parameters to describe the associated uncertainties. The uncertainty in the resulting posterior is jointly determined by the the informativeness (or variance) of the prior, and the sample size of the observed data. For a weakly informative prior, the posterior result is weighted more by the observed data. When the sample size is small, Bayesian methods often require more informative priors to output appropriate results.

#### 4.5. Future Opportunity 5: Data-Driven Optimization

## 5. Conclusions

## Author Contributions

## Funding

## Institutional Review Board Statement

## Informed Consent Statement

## Data Availability Statement

## Conflicts of Interest

## References

- Waitz, I.; Townsend, J.; Cutcher-Gershenfeld, J.; Greitzer, E.; Kerrebrock, J. Aviation and the Environment, A National Vision Statement, Framework for Goals and Recommended Actions; United States Congress: Washington, DC, USA, 2014.
- FAA (Federal Aviation Administration). Aviation Emissions, Impacts and Mitigation: A Primer; Federal Aviation Administration: Washington, DC, USA, 2015.
- Lee, D.; Fahey, D.; Skowron, A.; Allen, M.; Burkhardt, U.; Chen, Q.; Doherty, S.; Freeman, S.; Forster, P.; Fuglestvedt, J.; et al. The contribution of global aviation to anthropogenic climate forcing for 2000 to 2018. Atmos. Environ.
**2021**, 244, 117834. [Google Scholar] [CrossRef] [PubMed] - Brasseur, G.P.; Gupta, M.; Anderson, B.E.; Balasubramanian, S.; Barrett, S.; Duda, D.; Fleming, G.; Forster, P.M.; Fuglestvedt, J.; Gettelman, A.; et al. Impact of Aviation on Climate: FAA’s Aviation Climate Change Research Initiative (ACCRI) Phase II. Bull. Am. Meteorol. Soc.
**2016**, 97, 561–583. [Google Scholar] [CrossRef][Green Version] - Basner, M.; Clark, C.; Hansell, A.; Hileman, J.; Janssen, S.; Shepherd, K.; Sparrow, V. Aviation Noise Impacts: State of the Science. Noise Health
**2017**, 19, 41–50. [Google Scholar] [PubMed] - Brunton, S.L.; Nathan Kutz, J.; Manohar, K.; Aravkin, A.Y.; Morgansen, K.; Klemisch, J.; Goebel, N.; Buttrick, J.; Poskin, J.; Blom-Schieber, A.W.; et al. Data-Driven Aerospace Engineering: Reframing the Industry with Machine Learning. AIAA J.
**2021**, 59, 2820–2847. [Google Scholar] [CrossRef] - Mangortey, E.; Monteiro, D.; Ackley, J.; Gao, Z.; Puranik, T.G.; Kirby, M.; Pinon-Fischer, O.J.; Mavris, D.N. Application of Machine Learning Techniques to Parameter Selection for Flight Risk Identification. In Proceedings of the AIAA Scitech 2020 Forum, Orlando, FL, USA, 6–10 January 2020; American Institute of Aeronautics and Astronautics: Reston, VA, USA, 2020; p. 1850. [Google Scholar]
- Gao, Z. Representative Data and Models for Complex Aerospace Systems Analysis. Ph.D. Thesis, Georgia Institute of Technology, Atlanta, GA, USA, 2022. [Google Scholar]
- Sölveling, G.; Solak, S.; Clarke, J.P.B.; Johnson, E.L. Scheduling of runway operations for reduced environmental impact. Transp. Res. Part D Transp. Environ.
**2011**, 16, 110–120. [Google Scholar] [CrossRef] - Park, S.G.; Clarke, J.P. Vertical trajectory optimization to minimize environmental impact in the presence of wind. J. Aircr.
**2016**, 53, 725–737. [Google Scholar] [CrossRef][Green Version] - Matthes, S.; Grewe, V.; Lee, D.; Linke, F.; Shine, K.; Stromatas, S. ATM4E: A concept for environmentally-optimized aircraft trajectories. In Proceeding of the Greener Aviation 2016 Conference, Brussels, Belgium, 11–13 October 2016. [Google Scholar]
- Tian, Y.; Wan, L.; Han, K.; Ye, B. Optimization of terminal airspace operation with environmental considerations. Transp. Res. Part D Transp. Environ.
**2018**, 63, 872–889. [Google Scholar] [CrossRef] - Simorgh, A.; Soler, M.; González-Arribas, D.; Matthes, S.; Grewe, V.; Dietmüller, S.; Baumann, S.; Yamashita, H.; Yin, F.; Castino, F.; et al. A Comprehensive Survey on Climate Optimal Aircraft Trajectory Planning. Aerospace
**2022**, 9, 146. [Google Scholar] [CrossRef] - Antoine, N.E.; Kroo, I.M. Aircraft optimization for minimal environmental impact. J. Aircr.
**2004**, 41, 790–797. [Google Scholar] [CrossRef][Green Version] - Henderson, R.P.; Martins, J.R.; Perez, R.E. Aircraft conceptual design for optimal environmental performance. Aeronaut. J.
**2012**, 116, 1–22. [Google Scholar] [CrossRef] - Ilario da Silva, C.R.; Orra, T.H.; Alonso, J.J. Multi-objective aircraft design optimization for low external noise and fuel burn. In Proceedings of the 58th AIAA/ASCE/AHS/ASC Structures, Structural Dynamics, and Materials Conference, Grapevine, TX, USA, 9–13 January 2017; p. 1755. [Google Scholar]
- Proesmans, P.J.; Vos, R. Airplane design optimization for minimal global warming impact. J. Aircr.
**2022**, 59, 1363–1381. [Google Scholar] [CrossRef] - Zaporozhets, O.; Tokarev, V.; Attenborough, K. Aircraft Noise: Assessment, Prediction and Control; CRC Press: Boca Raton, FL, USA, 2011. [Google Scholar]
- Filippone, A. Aircraft noise prediction. Prog. Aerosp. Sci.
**2014**, 68, 27–63. [Google Scholar] [CrossRef] - Torija, A.J.; Self, R.H.; Flindell, I.H. A model for the rapid assessment of the impact of aviation noise near airports. J. Acoust. Soc. Am.
**2017**, 141, 981–995. [Google Scholar] [CrossRef] [PubMed][Green Version] - Torija, A.J.; Self, R.H.; Flindell, I.H. Airport noise modelling for strategic environmental impact assessment of aviation. Appl. Acoust.
**2018**, 132, 49–57. [Google Scholar] [CrossRef][Green Version] - Efron, B.; Hastie, T. Computer Age Statistical Inference: Algorithms, Evidence, and Data Science; Institute of Mathematical Statistics Monographs; Cambridge University Press: Cambridge, UK, 2016. [Google Scholar]
- Chatfield, C. Statistics for Technology: A Course in Applied Statistics; CRC Press: Boca Raton, FL, USA, 1983; Volume 3. [Google Scholar]
- Mallows, C. The Zeroth Problem. Am. Stat.
**1998**, 52, 1–9. [Google Scholar] - Brown, E.N.; Kass, R.E. What Is Statistics? Am. Stat.
**2009**, 63, 105–110. [Google Scholar] [CrossRef] - Romeijn, J.W. Statistics as Inductive Inference. In Philosophy of Statistics; North-Holland: Amsterdam, The Netherlands, 2011; Volume 7, pp. 751–774. [Google Scholar]
- Spiegelhalter, D. The Art of Statistics: Learning from Data; Penguin UK: London, UK, 2019. [Google Scholar]
- Pearl, J. Causal inference in statistics: An overview. Stat. Surv.
**2009**, 3, 96–146. [Google Scholar] [CrossRef] - Holland, P.W. Statistics and Causal Inference. J. Am. Stat. Assoc.
**1986**, 81, 945–960. [Google Scholar] [CrossRef] - Domingos, P. A Few Useful Things to Know about Machine Learning. Commun. ACM
**2012**, 55, 78–87. [Google Scholar] [CrossRef][Green Version] - Sutton, R.S.; Barto, A.G. Reinforcement Learning: An Introduction; MIT Press: Cambridge, MA, USA, 2018. [Google Scholar]
- Zhu, X.; Goldberg, A.B. Introduction to Semi-Supervised Learning Synthesis Lectures on Artificial Intelligence and Machine Learning; University of Wisconsin: Madison, WI, USA, 2009; Volume 3, pp. 1–130. [Google Scholar]
- Liu, X.; Zhang, F.; Hou, Z.; Mian, L.; Wang, Z.; Zhang, J.; Tang, J. Self-supervised Learning: Generative or Contrastive. IEEE Trans. Knowl. Data Eng.
**2021**, 14, 1–20. [Google Scholar] [CrossRef] - Goodfellow, I.; Pouget-Abadie, J.; Mirza, M.; Xu, B.; Warde-Farley, D.; Ozair, S.; Courville, A.; Bengio, Y. Generative Adversarial Nets. In Proceedings of the Advances in Neural Information Processing Systems, Montreal, QC, Canada, 5–10 December 2014; Curran Associates, Inc.: Red Hook, NY, USA, 2014; Volume 27. [Google Scholar]
- Settles, B. Active Learning Literature Survey; University of Wisconsin-Madison Department of Computer Sciences: Madison, WI, USA, 2009. [Google Scholar]
- Murphy, K.P. Machine Learning: A Probabilistic Perspective; MIT Press: Cambridge, MA, USA, 2012. [Google Scholar]
- Han, J.; Pei, J.; Kamber, M. Data Mining: Concepts and Techniques; Elsevier: Amsteradm, The Netherlands, 2011. [Google Scholar]
- Gao, Z.; Puranik, T.G.; Mavris, D.N. Probabilistic REpresentatives Mining (PREM): A Clustering Method for Distributional Data Reduction. AIAA J.
**2022**, 60, 2580–2596. [Google Scholar] [CrossRef] - Gao, Z.; Li, Y.; Puranik, T.G.; Mavris, D.N. Minimax and Multi-Criteria Selection of Representative Model Portfolios for Complex Systems Analysis. AIAA J.
**2022**, 60, 1505–1521. [Google Scholar] [CrossRef] - Jensen, L.; Thomas, J.; Brooks, C.; Brenner, M.; Hansman, R.J. Development of Rapid Fleet-Wide Environmental Assessment Capability. In Proceedings of the AIAA Modeling and Simulation Technologies Conference, London, UK, 23–27 September 2017; p. 3339. [Google Scholar]
- Torija, A.J.; Self, R.H. Aircraft classification for efficient modelling of environmental noise impact of aviation. J. Air Transp. Manag.
**2018**, 67, 157–168. [Google Scholar] [CrossRef][Green Version] - Gao, Z.; Kampezidou, S.I.; Behere, A.; Puranik, T.G.; Rajaram, D.; Mavris, D.N. Multi-level aircraft feature representation and selection for aviation environmental impact analysis. Transp. Res. Part C Emerg. Technol.
**2022**, 143, 103824. [Google Scholar] [CrossRef] - Pagoni, I.; Psaraki-Kalouptsidi, V. Calculation of aircraft fuel consumption and CO
_{2}emissions based on path profile estimation by clustering and registration. Transp. Res. Part D Transp. Environ.**2017**, 54, 172–190. [Google Scholar] [CrossRef] - Sun, J.; Ellerbroek, J.; Hoekstra, J. Flight extraction and phase identification for large automatic dependent surveillance–broadcast datasets. J. Aerosp. Inf. Syst.
**2017**, 14, 566–572. [Google Scholar] [CrossRef][Green Version] - Murca, M.C.R.; Hansman, R.J. Identification, characterization, and prediction of traffic flow patterns in multi-airport systems. IEEE Trans. Intell. Transp. Syst.
**2018**, 20, 1683–1696. [Google Scholar] [CrossRef] - Kadyk, T.; Schenkendorf, R.; Hawner, S.; Yildiz, B.; Römer, U. Design of fuel cell systems for aviation: Representative mission profiles and sensitivity analyses. Front. Energy Res.
**2019**, 7, 35. [Google Scholar] [CrossRef][Green Version] - Gao, Z.; Behere, A.; Li, Y.; Lim, D.; Kirby, M.; Mavris, D.N. Development and Analysis of Improved Departure Modeling for Aviation Environmental Impact Assessment. J. Aircr.
**2021**, 58, 847–857. [Google Scholar] [CrossRef] - Queipo, N.V.; Haftka, R.T.; Shyy, W.; Goel, T.; Vaidyanathan, R.; Kevin Tucker, P. Surrogate-based analysis and optimization. Prog. Aerosp. Sci.
**2005**, 41, 1–28. [Google Scholar] [CrossRef][Green Version] - Bernardo, J.E.; Kirby, M.; Mavris, D. Development of a Rapid Fleet-Level Noise Computation Model. J. Aircr.
**2015**, 52, 721–733. [Google Scholar] [CrossRef] - Lee, C.; Thrasher, T.; Hwang, S.; Shumway, M.; Zubrow, A.; Hansen, A.; Koopmann, J.; Solman, G. Aviation Environmental Design Tool (AEDT) User Manual Version 3c; Technical Report; Federal Aviation Administration: Washington, DC, USA, 2020.
- Kim, S.; Lim, D.; Lee, K. Reduced-order modeling applied to the aviation environmental design tool for rapid noise prediction. J. Aerosp. Eng.
**2018**, 31, 04018056. [Google Scholar] [CrossRef] - Ashok, A.; Lee, I.H.; Arunachalam, S.; Waitz, I.A.; Yim, S.H.; Barrett, S.R. Development of a response surface model of aviation’s air quality impacts in the United States. Atmos. Environ.
**2013**, 77, 445–452. [Google Scholar] [CrossRef] - LeVine, M.J.; Bernardo, J.E.; Kirby, M.; Mavris, D.N. Average generic vehicle method for fleet-level analysis of noise and emission tradeoffs. J. Aircr.
**2018**, 55, 929–946. [Google Scholar] [CrossRef] - Monteiro, D.J.; Prem, S.; Kirby, M.; Mavris, D.N. React: A rapid environmental impact on airport community tradeoff environment. In Proceedings of the 2018 AIAA Aerospace Sciences Meeting, Kissimmee, FL, USA, 8 January 2018; p. 263. [Google Scholar]
- Greenwood, E.; Schmitz, F.H.; Sickenberger, R.D. A semiempirical noise modeling method for helicopter maneuvering flight operations. J. Am. Helicopter Soc.
**2015**, 60, 1–13. [Google Scholar] [CrossRef][Green Version] - Yanto, J.; Liem, R.P. Aircraft fuel burn performance study: A data-enhanced modeling approach. Transp. Res. Part D Transp. Environ.
**2018**, 65, 574–595. [Google Scholar] [CrossRef] - Seymour, K.; Held, M.; Georges, G.; Boulouchos, K. Fuel Estimation in Air Transportation: Modeling global fuel consumption for commercial aviation. Transp. Res. Part D Transp. Environ.
**2020**, 88, 102528. [Google Scholar] [CrossRef] - Kuhn, M.; Johnson, K. Applied Predictive Modeling; Springer: New York, NY, USA, 2013; Volume 26. [Google Scholar]
- Kapoor, A.; Horvitz, Z.; Laube, S.; Horvitz, E. Airplanes aloft as a sensor network for wind forecasting. In Proceedings of the IPSN-14 13th International Symposium on Information Processing in Sensor Networks, Berlin, Germany, 8–11 April 2014; IEEE: Piscatway, NJ, USA, 2014; pp. 25–33. [Google Scholar]
- Kang, L.; Hansen, M. Improving airline fuel efficiency via fuel burn prediction and uncertainty estimation. Transp. Res. Part C Emerg. Technol.
**2018**, 97, 128–146. [Google Scholar] [CrossRef] - Kang, L.; Hansen, M. Quantile Regression–Based Estimation of Dynamic Statistical Contingency Fuel. Transp. Sci.
**2021**, 55, 257–273. [Google Scholar] [CrossRef] - Baklacioglu, T. Modeling the fuel flow-rate of transport aircraft during flight phases using genetic algorithm-optimized neural networks. Aerosp. Sci. Technol.
**2016**, 49, 52–62. [Google Scholar] [CrossRef] - Khan, W.A.; Ma, H.L.; Ouyang, X.; Mo, D.Y. Prediction of aircraft trajectory and the associated fuel consumption using covariance bidirectional extreme learning machines. Transp. Res. Part E Logist. Transp. Rev.
**2021**, 145, 102189. [Google Scholar] [CrossRef] - Jarry, G.; Delahaye, D.; Feron, E. Approach and landing aircraft on-board parameters estimation with lstm networks. In Proceedings of the 2020 International Conference on Artificial Intelligence and Data Analytics for Air Transportation (AIDA-AT), Singapore, 3–4 February 2020; IEEE: Piscataway, NJ, USA, 2020; pp. 1–6. [Google Scholar]
- Vela, A.E.; Oleyaei-Motlagh, Y. Ground level aviation noise prediction: A sequence to sequence modeling approach using LSTM recurrent neural networks. In Proceedings of the 2020 AIAA/IEEE 39th Digital Avionics Systems Conference (DASC), San Antonio, TX, USA, 11–16 October 2020; IEEE: Piscataway, NJ, USA, 2020; pp. 1–8. [Google Scholar]
- Wan, J.; Zhang, H.; Lyu, W.; Zhou, J. A Novel Combined Model for Short-Term Emission Prediction of Airspace Flights Based on Machine Learning: A Case Study of China. Sustainability
**2022**, 14, 4017. [Google Scholar] [CrossRef] - Uzun, M.; Demirezen, M.U.; Inalhan, G. Physics Guided Deep Learning for Data-Driven Aircraft Fuel Consumption Modeling. Aerospace
**2021**, 8, 44. [Google Scholar] [CrossRef] - Wiedemann, A.; Fuller, C.; Pascioni, K. Constructing a physics-guided machine learning neural network to predict tonal noise emitted by a propeller. In Proceedings of the INTER-NOISE and NOISE-CON Congress and Conference Proceedings, Seoul, Republic of Korea, 23–26 August 2022; Institute of Noise Control Engineering: Washington, DC, USA, 2022; Volume 264, pp. 151–162. [Google Scholar]
- Kiureghian, A.D.; Ditlevsen, O. Aleatory or Epistemic? Does It Matter? Struct. Saf.
**2009**, 31, 105–112. [Google Scholar] [CrossRef] - Guo, J.; Du, X. Sensitivity Analysis with Mixture of Epistemic and Aleatory Uncertainties. AIAA J.
**2007**, 45, 2337–2349. [Google Scholar] [CrossRef] - Roy, C.J.; Oberkampf, W.L. A Comprehensive Framework for Verification, Validation, and Uncertainty Quantification in Scientific Computing. Comput. Methods Appl. Mech. Eng.
**2011**, 200, 2131–2144. [Google Scholar] [CrossRef] - Allaire, D.; Willcox, K. Surrogate modeling for uncertainty assessment with application to aviation environmental system models. AIAA J.
**2010**, 48, 1791–1803. [Google Scholar] [CrossRef][Green Version] - Allaire, D.; Noel, G.; Willcox, K.; Cointin, R. Uncertainty quantification of an aviation environmental toolsuite. Reliab. Eng. Syst. Saf.
**2014**, 126, 14–24. [Google Scholar] [CrossRef] - Lim, D.; Li, Y.; LeVine, M.J.; Kirby, M.; Mavris, D.N. Parametric uncertainty quantification of aviation environmental design tool. In Proceedings of the 2018 Multidisciplinary Analysis and Optimization Conference, Atlanta, GA, USA, 25–29 June 2018; p. 3101. [Google Scholar]
- Behere, A.; Lim, D.; Li, Y.; Jin, Y.C.D.; Gao, Z.; Kirby, M.; Mavris, D.N. Sensitivity Analysis of Airport level Environmental Impacts to Aircraft thrust, weight, and departure procedures. In Proceedings of the AIAA Scitech 2020 Forum, Orlando, FL, USA, 23–27 January 2020; p. 1731. [Google Scholar]
- Simone, N.W.; Stettler, M.E.; Barrett, S.R. Rapid estimation of global civil aviation emissions with uncertainty quantification. Transp. Res. Part D Transp. Environ.
**2013**, 25, 33–41. [Google Scholar] [CrossRef] - Graas, R.; Sun, J.; Hoekstra, J. Quantifying accuracy and uncertainty in data-driven flight trajectory predictions with gaussian process regression. In Proceedings of the 11th SESAR Innovation Days, Online Conference, 7–9 December 2021. [Google Scholar]
- Amaral, S.; Allaire, D.; Blanco, E.D.L.R.; Willcox, K.E. A decomposition-based uncertainty quantification approach for environmental impacts of aviation technology and operation. AI EDAM
**2017**, 31, 251–264. [Google Scholar] [CrossRef][Green Version] - Gao, Z.; Lim, D.; Schwartz, K.G.; Mavris, D.N. A nonparametric-based approach for the characterization and propagation of epistemic uncertainty due to small datasets. In Proceedings of the AIAA Scitech 2019 Forum, San Diego, CA, USA, 7 January 2019; p. 1490. [Google Scholar]
- June, J.C.; Thomas, R.H.; Guo, Y. System Noise Prediction Uncertainty Quantification for a Hybrid Wing–Body Transport Concept. AIAA J.
**2020**, 58, 1157–1170. [Google Scholar] [CrossRef] - Akatsuka, J.; Ishii, T. System Noise Assessment and Uncertainty Analysis of a Conceptual Supersonic Aircraft. Aerospace
**2022**, 9, 212. [Google Scholar] [CrossRef] - Van Pham, V.; Tang, J.; Alam, S.; Lokan, C.; Abbass, H.A. Aviation emission inventory development and analysis. Environ. Model. Softw.
**2010**, 25, 1738–1753. [Google Scholar] [CrossRef] - Jansson, M. Development of a Fast Method to Analyze Patterns in Airport Noise. Master’s Thesis, Massachusetts Institute of Technology, Cambridge, MA, USA, 2021. [Google Scholar]
- Sun, J.; Dedoussi, I. Evaluation of aviation emissions and environmental costs in Europe using OpenSky and OpenAP. Eng. Proc.
**2021**, 13, 5. [Google Scholar] - Filippone, A.; Parkes, B.; Bojdo, N.; Kelly, T. Prediction of aircraft engine emissions using ADS-B flight data. Aeronaut. J.
**2021**, 125, 988–1012. [Google Scholar] [CrossRef] - Filippone, A.; Parkes, B. Evaluation of commuter airplane emissions: A European case study. Transp. Res. Part D Transp. Environ.
**2021**, 98, 102979. [Google Scholar] [CrossRef] - Maruhashi, J.; Grewe, V.; Frömming, C.; Jöckel, P.; Dedoussi, I.C. Transport Patterns of Global Aviation NOx and their Short-term O3 Radiative Forcing–A Machine Learning Approach. Atmos. Chem. Phys. Discuss.
**2022**, 22, 14253–14282. [Google Scholar] [CrossRef] - Quadros, F.D.; Snellen, M.; Sun, J.; Dedoussi, I.C. Global Civil Aviation Emissions Estimates for 2017–2020 Using ADS-B Data. J. Aircr.
**2022**, 35, 1–11. [Google Scholar] [CrossRef] - Kulik, L. Satellite-Based Detection of Contrails Using Deep Learning. Ph.D. Thesis, Massachusetts Institute of Technology, Cambridge, MA, USA, 2019. [Google Scholar]
- Fan, C. Formal Methods for Safe Autonomy: Data-Driven Verification, Synthesis, and Applications. Ph.D. Thesis, University of Illinois at Urbana-Champaign, Champaign, IL, USA, 2019. [Google Scholar]
- Li, Y.; Lim, D.; Kirby, M.; Mavris, D.N.; Noel, G. Uncertainty Quantification Analysis of the Aviation Environmental Design Tool in Emission Inventory and Air Quality Modeling. In Proceedings of the 2018 Aviation Technology, Integration, and Operations Conference, Reston, VA, USA, 25–29 June 2018; p. 3050. [Google Scholar]
- Gabrielian, A.B.; Puranik, T.G.; Bendarkar, M.V.; Kirby, M.; Mavris, D.; Monteiro, D. Noise Model Validation using Real World Operations Data. In Proceedings of the AIAA Aviation 2021 Forum, Online Conference, 2–6 August 2021; p. 2136. [Google Scholar]
- Meister, J.; Schalcher, S.; Wunderli, J.M.; Jäger, D.; Zellmann, C.; Schäffer, B. Comparison of the aircraft noise calculation programs sonAIR, FLULA2 and AEDT with noise measurements of single flights. Aerospace
**2021**, 8, 388. [Google Scholar] [CrossRef] - Botre, M.; Brentner, K.S.; Horn, J.F.; Wachspress, D. Validation of helicopter noise prediction system with flight data. In Proceedings of the Vertical Flight Society 75th Annual Forum & Technology Display, Held, Ukraine, 13–16 May 2019; Volume 13. [Google Scholar]
- Filippone, A.; Zhang, M.; Bojdo, N. Validation of an integrated simulation model for aircraft noise and engine emissions. Aerosp. Sci. Technol.
**2019**, 89, 370–381. [Google Scholar] [CrossRef] - Simons, D.G.; Besnea, I.; Mohammadloo, T.H.; Melkert, J.A.; Snellen, M. Comparative assessment of measured and modelled aircraft noise around Amsterdam Airport Schiphol. Transp. Res. Part D Transp. Environ.
**2022**, 105, 103216. [Google Scholar] [CrossRef] - Vieira, A.; von den Hoff, B.; Snellen, M.; Simons, D.G. Comparison of Semi-Empirical Noise Models with Flyover Measurements of Operating Aircraft. J. Aircr.
**2022**, 1–14. [Google Scholar] [CrossRef] - Huynh, J.L.; Mahseredjian, A.; John Hansman, R. Delayed Deceleration Approach Noise Impact and Modeling Validation. J. Aircr.
**2022**, 59, 1–13. [Google Scholar] [CrossRef] - Jackson, D.C.; Rindfleisch, T.C.; Alonso, J.J. A System for Measurement and Analysis of Aircraft Noise Impacts. Eng. Proc.
**2021**, 13, 6. [Google Scholar] - Omidvar-Tehrani, B.; Nandi, A.; Meyer, N.; Flanagan, D.; Young, S. Dv8: Interactive analysis of aviation data. In Proceedings of the 2017 IEEE 33rd International Conference on Data Engineering (ICDE), San Diego, CA, USA, 19–22 April 2017; IEEE: Piscatway, NJ, USA, 2017; pp. 1411–1412. [Google Scholar]
- Eckstein, A.; Kurcz, C.; Silva, M. Threaded Track: Geospatial Data Fusion for Aircraft Flight Trajectories; Technical Report; MITRE Corporation: McLean, VA, USA, 2012. [Google Scholar]
- Sun, J.; Ellerbroek, J.; Hoekstra, J.M. WRAP: An open-source kinematic aircraft performance model. Transp. Res. Part C Emerg. Technol.
**2019**, 98, 118–138. [Google Scholar] [CrossRef] - NREL (National Renewable Energy Laboratory). Flight DNA: An Anonymized Aviation Data Tool and Repository; NREL: Golden, CO, USA, 2022.
- Olive, X. Traffic, a toolbox for processing and analysing air traffic data. J. Open Source Softw.
**2019**, 4, 1518. [Google Scholar] [CrossRef][Green Version] - Sun, J.; Vû, H.; Ellerbroek, J.; Hoekstra, J.M. pymodes: Decoding mode-s surveillance data for open air transportation research. IEEE Trans. Intell. Transp. Syst.
**2019**, 21, 2777–2786. [Google Scholar] [CrossRef] - Ayala, R.; Ayala, D.; Vidal, L.S.; Ruiz, D. openSkies-Integration of Aviation Data into the R Ecosystem. R J.
**2021**, 13, 590–599. [Google Scholar] [CrossRef] - Verleysen, M.; François, D. The curse of dimensionality in data mining and time series prediction. In Proceedings of the International Work-Conference on Artificial Neural Networks, Athens, Greece, 10–14 September 2005; Springer: New York, NY, USA, 2005; pp. 758–770. [Google Scholar]
- Wang, J.L.; Chiou, J.M.; Müller, H.G. Functional data analysis. Annu. Rev. Stat. Its Appl.
**2016**, 3, 257–295. [Google Scholar] [CrossRef][Green Version] - Kokoszka, P.; Reimherr, M. Introduction to Functional Data Analysis; Chapman and Hall/CRC: Boca Raton, FL, USA, 2017. [Google Scholar]
- Kolda, T.G.; Bader, B.W. Tensor decompositions and applications. SIAM Rev.
**2009**, 51, 455–500. [Google Scholar] [CrossRef] - Cressie, N.; Wikle, C.K. Statistics for Spatio-Temporal Data; John Wiley & Sons: Hoboken, NJ, USA, 2015. [Google Scholar]
- Atluri, G.; Karpatne, A.; Kumar, V. Spatio-temporal data mining: A survey of problems and methods. ACM Comput. Surv. (CSUR)
**2018**, 51, 1–41. [Google Scholar] [CrossRef] - Roddick, J.F.; Spiliopoulou, M. A bibliography of temporal, spatial and spatio-temporal data mining research. ACM SIGKDD Explor. Newsl.
**1999**, 1, 34–38. [Google Scholar] [CrossRef] - Ripley, B.D. Spatial Statistics; John Wiley & Sons: Hoboken, NJ, USA, 2005. [Google Scholar]
- Kutz, J.N. Deep learning in fluid dynamics. J. Fluid Mech.
**2017**, 814, 1–4. [Google Scholar] [CrossRef] - Kashinath, K.; Mustafa, M.; Albert, A.; Wu, J.; Jiang, C.; Esmaeilzadeh, S.; Azizzadenesheli, K.; Wang, R.; Chattopadhyay, A.; Singh, A.; et al. Physics-informed machine learning: Case studies for weather and climate modelling. Philos. Trans. R. Soc. A
**2021**, 379, 20200093. [Google Scholar] [CrossRef] [PubMed] - Karniadakis, G.E.; Kevrekidis, I.G.; Lu, L.; Perdikaris, P.; Wang, S.; Yang, L. Physics-informed machine learning. Nat. Rev. Phys.
**2021**, 3, 422–440. [Google Scholar] [CrossRef] - Raissi, M.; Perdikaris, P.; Karniadakis, G.E. Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations. J. Comput. Phys.
**2019**, 378, 686–707. [Google Scholar] [CrossRef] - Mao, Z.; Jagtap, A.D.; Karniadakis, G.E. Physics-informed neural networks for high-speed flows. Comput. Methods Appl. Mech. Eng.
**2020**, 360, 112789. [Google Scholar] [CrossRef] - Cai, S.; Mao, Z.; Wang, Z.; Yin, M.; Karniadakis, G.E. Physics-informed neural networks (PINNs) for fluid mechanics: A review. Acta Mech. Sin.
**2022**, 37, 1727–1738. [Google Scholar] [CrossRef] - Haghighat, E.; Raissi, M.; Moure, A.; Gomez, H.; Juanes, R. A physics-informed deep learning framework for inversion and surrogate modeling in solid mechanics. Comput. Methods Appl. Mech. Eng.
**2021**, 379, 113741. [Google Scholar] [CrossRef] - Murdoch, W.J.; Singh, C.; Kumbier, K.; Abbasi-Asl, R.; Yu, B. Definitions, methods, and applications in interpretable machine learning. Proc. Natl. Acad. Sci. USA
**2019**, 116, 22071–22080. [Google Scholar] [CrossRef][Green Version] - Linardatos, P.; Papastefanopoulos, V.; Kotsiantis, S. Explainable AI: A Review of Machine Learning Interpretability Methods. Entropy
**2021**, 23, 18. [Google Scholar] [CrossRef] [PubMed] - Shukla, B.; Fan, I.S.; Jennions, I. Opportunities for Explainable Artificial Intelligence in Aerospace Predictive Maintenance. In Proceedings of the PHM Society European Conference, Sanya, China, 12–14 November 2020; Volume 5, p. 11. [Google Scholar]
- Grushin, A.; Nanda, J.; Tyagi, A.; Miller, D.; Gluck, J.; Oza, N.C.; Maheshwari, A. Decoding the Black Box: Extracting Explainable Decision Boundary Approximations from Machine Learning Models for Real Time Safety Assurance of the National Airspace. In Proceedings of the AIAA Scitech 2019 Forum, San Diego, CA, USA, 7–11 January 2019. Number AIAA 2019-0136. [Google Scholar]
- Memarzadeh, M.; Matthews, B.; Templin, T. Multi-Class Anomaly Detection in Flight Data Using Semi-Supervised Explainable Deep Learning Model. In Proceedings of the AIAA Scitech 2021 Forum, Online Conference, 19–21 January 2021. Number AIAA 2021-0774. [Google Scholar]
- van de Schoot, R.; Depaoli, S.; King, R.; Kramer, B.; Märtens, K.; Tadesse, M.G.; Vannucci, M.; Gelman, A.; Veen, D.; Willemsen, J.; et al. Bayesian statistics and modelling. Nat. Rev. Methods Prim.
**2021**, 1, 1–26. [Google Scholar] [CrossRef] - Gelman, A.; Shalizi, C.R. Philosophy and the practice of Bayesian statistics. Br. J. Math. Stat. Psychol.
**2013**, 66, 8–38. [Google Scholar] [CrossRef] [PubMed] - Hoff, P.D. A First Course in Bayesian Statistical Methods; Springer: New York, NY, USA, 2009; Volume 580. [Google Scholar]
- Ning, C.; You, F. Optimization under uncertainty in the era of big data and deep learning: When machine learning meets mathematical programming. Comput. Chem. Eng.
**2019**, 125, 434–448. [Google Scholar] [CrossRef][Green Version] - Yang, J.; Dong, J.; Hu, L. A data-driven optimization-based approach for siting and sizing of electric taxi charging stations. Transp. Res. Part C Emerg. Technol.
**2017**, 77, 462–477. [Google Scholar] [CrossRef][Green Version] - Van Nguyen, T.; Zhang, J.; Zhou, L.; Meng, M.; He, Y. A data-driven optimization of large-scale dry port location using the hybrid approach of data mining and complex network theory. Transp. Res. Part E Logist. Transp. Rev.
**2020**, 134, 101816. [Google Scholar] [CrossRef] - Kim, J.; Justin, C.; Mavris, D.; Briceno, S. Data-Driven Approach Using Machine Learning for Real-Time Flight Path Optimization. J. Aerosp. Inf. Syst.
**2022**, 19, 3–21. [Google Scholar] [CrossRef] - Elmachtoub, A.N.; Grigas, P. Smart “predict, then optimize”. Manag. Sci.
**2022**, 68, 9–26. [Google Scholar] [CrossRef]

Year | Paper | Topic | Key Contributions |
---|---|---|---|

2017 | [40] | Development of Rapid Fleet-Wide Environmental Assessment Capability | Develops a methodology for rapid analysis of fleet-level noise and emissions. The method extracts representative flight trajectories from large operations (ASDE-X) data and uses a subset of representative aircraft types to reduce the computational cost. |

2017 | [43] | Calculation of Aircraft Fuel Consumption and CO${}_{2}$ Emissions based on Path Profile Estimation by Clustering and Registration | Calculates typical aircraft fuel burn and CO${}_{2}$ emissions on the Climb-Cruise-Descent (CCD) cycle using representative flight paths and aircraft performance model. The method applies clustering on large dataset to extract flight characteristics and converts them into representative flight profiles. |

2017 | [44] | Flight Extraction and Phase Identification for Large Automatic Dependent Surveillance–Broadcast Datasets | Performs flight extraction and phase identification on the large ADS-B datasets. The flight extraction part utilizes clustering to extract continuous flights. The flight phase identification part then applies fuzzy logic to segment flight data into different phases. |

2018 | [41] | Aircraft Classification for Efficient Modelling of Environmental Noise Impact of Aviation | Conducts rapid fleet-level noise contour computation through aircraft classification. The method reduces the UK commercial aircraft fleet to four representative-in-class aircraft using aircraft physical, emissions, and noise features. |

2018 | [45] | Identification, Characterization, and Prediction of Traffic Flow Patterns in Multi-Airport Systems | Presents a data-driven framework to identify, characterize, and predict traffic flow patterns in complex airspace. The methods applies machine learning methods, mainly multi-layer clustering and multi-way classification, on historical flight and weather data. |

2019 | [46] | Design of Fuel Cell Systems for Aviation: Representative Mission Profiles and Sensitivity Analyses | Specifies requirements for the design of fuel cell systems for passenger aircraft. The work extracts representative mission profiles through statistical analysis on flight data and construction of probabilistic model for the mission profile. |

2021 | [47] | Development and Analysis of Improved Departure Modeling for Aviation Environmental Impact Assessment | Develops representative aircraft departure procedures from real-world flight operations data for simulating aircraft takeoff environmental impacts. The simulation results are then evaluated through statistical analysis and statistical learning to uncover patterns. |

2022 | [38] | Probabilistic REpresentatives Mining (PREM): A Clustering Method for Distributional Data Reduction | Develops a methodology for distributional data reduction which is effective and consistent at small sample sizes. The method enables the use of only a very small subset of aircraft operations data for efficient uncertainty propagation in environmental impact modeling. |

2022 | [39] | Minimax and Multi-Criteria Selection of Representative Model Portfolios for Complex Systems Analysis | Develops a methodology which utilizes minimax and multi-criteria considerations to select a small subset of representative aircraft types to cover the richness and complexity in entire population for building the costly aircraft noise and performance models. Multiple machine learning and data visualization techniques are involved. |

2022 | [42] | Multi-level Aircraft Feature Representation and Selection for Aviation Environmental Impact Analysis | Conducts a comprehensive aircraft feature selection study on aviation environmental impacts. The result provides improved (and reduced) aircraft representation for the aircraft grouping problem and generates insights on aircraft features that are influential on different levels of environmental impacts. |

Year | Paper | Topic | Key Contributions |
---|---|---|---|

2013 | [52] | Development of a Response Surface Model of Aviation’s Air Quality Impacts in the United States | Constructs a response surface model (RSM) to evaluate the air quality impacts of aviation in the U.S. for present-day and future scenarios. The surrogate model is a rapid version which approximates the computationally expensive Community Multiscale Air Quality (CMAQ) modeling system. |

2015 | [49] | Development of a Rapid Fleet-level Noise Computation Model | Presents a rapid fleet-level noise computation model that leverages the fidelity of detailed models. The simplified method performs generic aircraft operations upfront and recombines events later to evaluate the impact of new technologies and perform trades of different noise mitigating strategies. |

2015 | [55] | A Semiempirical Noise Modeling Method for Helicopter Maneuvering Flight Operations | Develops a semi-empirical noise model for helicopter blade–vortex interaction (BVI) noise during maneuvering flight. The methods uses performance and acoustic data from both flight and wind tunnel tests to build a computationally efficient analytical model for acoustic mission planning. |

2018 | [53] | Average Generic Vehicle Method for Fleet-level Analysis of Noise and Emission Tradeoffs | Proposes a method named generating emissions and noise, evaluating residuals, and using inverse methods for choosing the best alternatives (GENERICA). The method uses surrogate models to model average generic vehicles for fleet-level analysis of technology impacts on environmental metrics. |

2018 | [54] | REACT: A Rapid Environmental Impact on Airport Community Tradeoff Environment | Proposes a rapid computational environment named Rapid Environmental Impact on Airport Community Tradeoff (REACT). The environment has a user interface and can rapidly tradeoff various noise mitigation strategies to manage airport community noise exposure in current and future airport scenarios. |

2018 | [51] | Reduced-Order Modeling Applied to the Aviation Environmental Design Tool for Rapid Noise Prediction | Develops a rapid approximation of the aviation environmental design tool (AEDT) noise model via reduced-order modeling (ROM). The method uses proper orthogonal decomposition (POD) for orthonormal basis extraction and kriging for basis coefficient prediction. |

2018 | [56] | Aircraft Fuel Burn Performance Study: A Data-enhanced Modeling Approach | Develops a data-enhanced surrogate model for aircraft fuel burn. The method improves the efficiency and accuracy of fuel burn modeling by combining a low-fidelity physics-based model with aircraft operation and performance data. A sample-based linear regression model is built for each aircraft type. |

2020 | [57] | Fuel Estimation in Air Transportation: Modeling Global Fuel Consumption for Commercial Aviation | Develops a framework named Fuel Estimation in Air Transportation (FEAT). It is a rapid analysis capability which consists of (1) a high fidelity flight profile simulator based on EUROCONTROL’s aircraft performance model, and (2) a reduced order fuel burn model with airport pair and aircraft type as inputs. |

Year | Paper | Topic | Key Contributions |
---|---|---|---|

2014 | [59] | Airplanes Aloft as a Sensor Network for Wind Forecasting | Applies machine learning on aircraft air and ground speeds data to develop a wind forecasting model for reducing the environmental impact of aviation. The method involves the use of Probabilistic Graphical Model (PGM) and Gaussian Process Regression (GPR) for wind prediction. |

2016 | [62] | Modeling the Fuel Flow-rate of Transport Aircraft During Flight Phases using Genetic Algorithm-optimized Neural Networks | Develops a deep learning model to predict the fuel consumption of transport aircraft for minimizing emissions and saving fuel. The method develops from real flight data a genetic algorithm-optimized neural network topology that is specifically designed for the fuel flow rate problem. |

2018 | [60] | Improving Airline Fuel Efficiency via Fuel Burn Prediction and Uncertainty Estimation | Proposes a discretionary fuel prediction method for reducing the discretionary fuel loading by dispatchers while maintaining the same safety level and saving fuel. The method applies ensemble learning to improve the prediction of fuel burn and construct uncertainty intervals for the model predictions. |

2020 | [64] | Approach and Landing Aircraft On-Board Parameters Estimation with LSTM Networks | Develops a model to estimate aircraft on-board parameters such as the fuel flow rate for enhancing the system’s safety and efficiency. The method applies Long Short Term Memory (LSTM) neural network on Flight Data Monitoring (FDM) data records to estimate target parameters. |

2020 | [65] | Ground Level Aviation Noise Prediction: A Sequence to Sequence Modeling Approach Using LSTM Recurrent Neural Networks | Develops a deep learning model to predict ground level aviation noise. The method applies Sequence-to-sequence Long Short Term Memory (LSTM) Recurrent Neural Network (RNN) on large radar and noise datasets to predict aviation noise at a ground location near Washington National Airport. |

2021 | [61] | Quantile Regression–Based Estimation of Dynamic Statistical Contingency Fuel | Applies machine learning to estimate the Statistical Contingency Fuel (SCF) for reducing fuel consumption. The method employs quantile regression on a large fuel burn dataset from a major U.S.-based airline to estimate the SCF and account for uncertainties. |

2021 | [67] | Physics Guided Deep Learning for Data-Driven Aircraft Fuel Consumption Modeling | Presents a framework which uses physics-guided deep learning to model aircraft fuel burn. The method guides the neural network with fuel flow dynamics equations and embeds physical knowledge as extra losses in the model training to outperform other model-based and supervised learning approaches. |

2021 | [63] | Prediction of Aircraft Trajectory and the Associated Fuel Consumption using Covariance Bidirectional Extreme Learning Machines | Applies deep learning to predict aircraft trajectory and the associated fuel consumption. The method uses covariance bidirectional extreme learning machine (CovB-ELM) to achieve a more accurate and robust performance than the existing methods. |

2022 | [66] | A Novel Combined Model for Short-Term Emission Prediction of Airspace Flights Based on Machine Learning: A Case Study of China | Applies machine learning to predict short-term flight emissions within enroute airspace. The method uses an adaptive weighting approach on results from a Long Short Term Memory (LSTM) prediction model and an extreme gradient boosting (XGBoost) prediction model to improve the performance. |

2022 | [68] | Constructing a Physics-guided Machine Learning Neural Network to Predict Tonal Noise Emitted by a Propeller | Applies deep learning to predict propeller tonal noise in the time domain over a broad range of flight conditions. The method uses physics-guided neural networks to improve the prediction performance while alleviating the dataset size requirement for experimental data. |

Year | Paper | Topic | Key Contributions |
---|---|---|---|

2010 | [72] | Surrogate Modeling for Uncertainty Assessment with Application to Aviation Environmental System Models | Proposes a surrogate modeling methodology designed specifically for uncertainty propagation and sensitivity analysis. The method is demonstrated on a large-scale aviation environmental model and can provide fast predictions with confidence intervals to support environmental policy-making. |

2013 | [76] | Rapid Estimation of Global Civil Aviation Emissions with Uncertainty Quantification | Develops a methodology and open source code for rapidly computing global aviation emissions with uncertainty quantification. The method enables global fleet-wide simulations for rapid policy analyses and quantification of uncertainties from operational factors, scientific knowledge, and model fidelity. |

2014 | [73] | Uncertainty Quantification of an Aviation Environmental Toolsuite | Describes uncertainty quantification of a complex computational tool for aviation environmental impact. The method consists of surrogate modeling to overcome the complexities of long run times and sensitivity analysis to identifying high priority areas for future research. |

2017 | [78] | A Decomposition-based Uncertainty Quantification Approach for Environmental Impacts of Aviation Technology and Operation | Proposes a divide-and-conquer approach, similar to the decomposition-based approaches in multidisciplinary analysis and optimization, to quantify uncertainty in multicomponent systems. Performs uncertainty analysis and global sensitivity analysis for environmental impacts of enhanced aviation technologies and operations. |

2018 | [74] | Parametric Uncertainty Quantification of Aviation Environmental Design Tool | Conducts parametric uncertainty quantification at the vehicle level for Aviation Environmental Design Tool (AEDT). The study to identifies the main contributors to AEDT output uncertainties and gains better insights on the areas of future AEDT improvements. |

2019 | [79] | A Nonparametric-based Approach for the Characterization and Propagation of Epistemic Uncertainty due to Small Datasets | Proposes a nonparametric framework to characterize and propagate uncertainty when only small datasets are available. The approach requires less assumption on the type of probability distribution of the uncertainty sources and brings greater flexibility into the UQ process. |

2020 | [75] | Sensitivity Analysis of Airport Level Environmental Impacts to Aircraft Thrust, Weight, and Departure Procedures | Conducts sensitivity analysis for fleet-level fuel burn, noise, and emissions to changes in uncertain factors such as aircraft takeoff weight, thrust, and departure profiles. The result underlines the importance of these factors when optimizing aircraft departure operations for environmental impact mitigation. |

2020 | [80] | System Noise Prediction Uncertainty Quantification for a Hybrid Wing–Body Transport Concept | Performs uncertainty quantification on the noise of a hybrid wing–body aircraft configuration. The method propagates element-level uncertainties through Monte Carlo simulation to the system level for noise predictions at the three certification locations and provides future research directions. |

2021 | [77] | Quantifying Accuracy and Uncertainty in Data-Driven Flight Trajectory Predictions with Gaussian Process Regression | Performs uncertainty quantification on data-driven 4D flight trajectory predictions using a two-stage Gaussian Process Regression (GPR). The study also evaluates and quantifies how flight-plan and meteorological information can help reducing the prediction error and uncertainty. |

2022 | [81] | System Noise Assessment and Uncertainty Analysis of a Conceptual Supersonic Aircraft | Performs system noise assessment, uncertainty analysis, and validation tests for a conceptual supersonic aircraft using Monte Carlo simulation. The result also identifies the noise factors that have significant impact on the landing and takeoff noise (LTO) noise. |

Year | Paper | Topic | Key Contributions |
---|---|---|---|

2010 | [82] | Aviation Emission Inventory development and analysis | Develops a 4D aviation emission inventory using air traffic trajectory data from Australian Airspace for spatial and temporal emission analysis. The result shows the disparity of CO${}_{2}$ concentration in different parts of Australia and the impact of NO${}_{x}$ emission on different layers of the atmosphere. |

2019 | [89] | Satellite-based Detection of Contrails using Deep Learning | Trains a Convolutional Neural Network (CNN) on satellite images for the automated detection of aircraft contrails, a major source of climate warming effect by aviation emissions. The result estimates that contrails cover an average of 0.55% of the contiguous U.S. and discovers the relationship between contrail coverage and air traffic as a function of time and location. |

2021 | [83] | Development of a Fast Method to Analyze Patterns in Airport Noise | Uses large quantity of flight track data and a fast noise approximation model on airport noise modeling. The result highlights the variability in noise patterns depending on evolving airport runway configuration at Boston Logan International Airport (KBOS). |

2021 | [84] | Evaluation of Aviation Emissions and Environmental Costs in Europe Using OpenSky and OpenAP | Proposes a data-driven approach for rapid estimations of cruise-level flight emissions over Europe using open data (ADS-B data) and open models (OpenAP emission models). The result shows cruise-level flight emissions by different airlines, geographic regions, altitudes, and timeframe. |

2021 | [85] | Prediction of Aircraft Engine Emissions using ADS-B Flight Data | Combines real-time flight data from ADS-B and flight performance model to predict aviation emissions at altitude – greater than 3,000 ft and exclude takeoff and landing. The result shows that NO${}_{x}$ and water vapour emissions concentrate around tropospheric altitudes only for long-range flights. |

2021 | [86] | Evaluation of Commuter Airplane Emissions: A European Case Study Author Links Open Overlay Panel | Simulates flights using ADS-B/Mode-S data to evaluate commuter airplane emissions in Europe. It studies a network of short-haul commuter flights (less than 300 n-miles) and analyzes fuel burn and emissions as function of distance, altitude, city pairs. It finds out that flight range is the biggest clear discriminator in emissions. |

2022 | [87] | Transport Patterns of Global Aviation NO${}_{x}$ and their Short-term O${}_{3}$ Radiative Forcing – A Machine Learning Approach | Uses global-scale simulations and the unsupervised QuickBundles clustering approach to study the transport patterns of emitted NO${}_{x}$ and their associated climate impacts in different regions and seasons. The result highlights the spatially and temporally heterogeneous nature of the NO${}_{x}$–O${}_{3}$ chemistry from a global perspective. |

2022 | [88] | Global Civil Aviation Emissions Estimates for 2017–2020 Using ADS-B Data | Uses ADS-B data, Base of Aircraft Data (BADA) aircraft performance model, and ICAO’s Engine Emissions Databank to estimate global emissions from aircraft operations for the years 2017–2020. The result quantifies global aviation emissions and the evolution of the fleet average emission indices over time, including impact from COVID-19. |

Year | Paper | Topic | Key Contributions |
---|---|---|---|

2018 | [91] | Uncertainty Quantification Analysis of the Aviation Environmental Design Tool in Emission Inventory and Air Quality Modeling | Conducts an uncertainty quantification analysis on AEDT provide verification and validation of AEDT’s emission inventory and air quality modeling. It investigates causes that lead to the differences between AEDT and the legacy tool Emissions and Dispersion Modeling System (EDMS). |

2019 | [94] | Validation of Helicopter Noise Prediction System with Flight Data | Conducts a validation exercise for a helicopter noise prediction system to understand its limitations. It compares the Sound Exposure Level (SEL) noise contours between the model predictions and the acoustic flight test data for a range of flight conditions and concludes the predictions are overall satisfactory. |

2019 | [95] | Validation of an Integrated Simulation Model for Aircraft Noise and Engine Emissions | Conducts a validation exercise for an integrated aircraft environmental simulation software’s acoustic and engine exhaust emissions modules. It compares between the microphone field measurements at Manchester airport and numerical predictions for 12 common commercial airplanes. |

2021 | [93] | Comparison of the Aircraft Noise Calculation Programs sonAIR, FLULA2 and AEDT with Noise Measurements of Single Flights | Compares the actual noise exposure measurements with calculations of several thousand single flights using three noise calculation programs: sonAIR, FLULA2, and AEDT. It mentions that all three programs show good result, yet sonAIR can perform better in modeling single flights. |

2021 | [92] | Noise Model Validation using Real World Operations Data | Provides a structured and repeatable framework for noise model validation using real-world operations data. The validation utilizes multiple types of real-world data including detailed airline flight data records, noise monitoring data from stations around airport, and historical weather data. |

2022 | [98] | Delayed Deceleration Approach Noise Impact and Modeling Validation | Presents a validation methodology for delayed deceleration approach using noise measurements and radar data for several aircraft types. The method is demonstrated through comparing modeled sound exposure levels of these new procedures with available ground-noise-monitor data at two major airports. |

2022 | [96] | Comparative Assessment of Measured and Modelled Aircraft Noise around Amsterdam Airport Schiphol | Compares the “Dutch aircraft noise model” predictions to measured values from the NOise MOnitoring System (NOMOS) around Amsterdam Airport Schiphol between 2012 and 2018. It finds out that the model prediction improved throughout the years due several factors. |

2022 | [97] | Comparison of Semi-Empirical Noise Models with Flyover Measurements of Operating Aircraft | Investigates the sensitivity of semi-empirical models of engine and airframe noise to uncertainties in geometrical parameters and aircraft operating conditions, and compares the predictions to measurements of A320, A330, and B777. It identifies reasons behind the mismatch and improves the model. |

Year | Paper | Topic | Key Contributions |
---|---|---|---|

2012 | [101] | Threaded Track: Geospatial Data Fusion for Aircraft Flight Trajectories | Presents the threaded track repository: a robust and efficient capability of fusing radar trajectories from a variety of surveillance sources based on their temporal and spatial proximity to produce a synthetic track with the best possible coverage and fidelity. The Threaded Track represents the optimal representation of an aircraft’s end to end trajectory to support a wide range of safety, security, and efficiency analyses. |

2017 | [100] | DV8: Interactive Analysis of Aviation Data | Proposes DV8: an interactive data visualization framework for providing visualized aviation-oriented insights, with a focus on evaluating the deviations among flights by route, type, airport, and aircraft performance. DV8 can be utilized in areas such as capacity planning, flight route prediction, and fuel consumption. |

2019 | [104] | traffic: A Toolbox for Processing and Analyzing Air Traffic Data | Presents traffic: a Python toolbox for preprocessing and analyzing trajectories data evolving in airspaces. The tool can prepare data for aviation researchers and data scientists needing to compute statistics, performance indicators and building datasets for common machine learning tasks. |

2019 | [102] | WRAP: An Open-source Kinematic Aircraft Performance Model | Presents WRAP: a comprehensive set of methods for extracting different aircraft performance parameters from large scale open ADS-B data. This open-source data includes a set of more than 30 parameters from 7 distinct flight phases for 17 common commercial aircraft types and the fitted parametric models. |

2020 | [105] | pyModeS: Decoding Mode-S Surveillance Data for Open Air Transportation Research | Proposes pyModeS: an open-source library and new heuristic-probabilistic method to decode the Mode-S Comm-B replies and to check the correctness of the messages. It fills the gap of handling interrogation-based surveillance data and gives researchers broader access to accurate aircraft state updates that are transmitted through Enhanced Mode-S. |

2021 | [99] | A System for Measurement and Analysis of Aircraft Noise Impacts | Presents the system architecture, design, and current set of capabilities of the Metroplex Overflight Noise Analysis (MONA) system. The MONA project seeks to measure, analyze, and archive the ground noise data from aircraft overflights for a variety of purposes, such as an openly-available database for V&V of improved noise prediction methods. |

2021 | [106] | openSkies: Integration of Aviation Data into the R Ecosystem | Present openSkies: the first R package for processing public air traffic data. The package provides an interface to resources in the OpenSky Network, standardized data structures to represent the different entities involved in air traffic data, and functionalities to analyze and visualize such data. |

2022 | [103] | Flight DNA: An Anonymized Aviation Data Tool and Repository | Introduces Flight DNA: a common database with anonymized data on aviation components, systems, technologies, and operations. It includes planning and analysis tools, and repository for aviation emissions, energy consumption, and performance profiles. |

Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. |

© 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).

## Share and Cite

**MDPI and ACS Style**

Gao, Z.; Mavris, D.N. Statistics and Machine Learning in Aviation Environmental Impact Analysis: A Survey of Recent Progress. *Aerospace* **2022**, *9*, 750.
https://doi.org/10.3390/aerospace9120750

**AMA Style**

Gao Z, Mavris DN. Statistics and Machine Learning in Aviation Environmental Impact Analysis: A Survey of Recent Progress. *Aerospace*. 2022; 9(12):750.
https://doi.org/10.3390/aerospace9120750

**Chicago/Turabian Style**

Gao, Zhenyu, and Dimitri N. Mavris. 2022. "Statistics and Machine Learning in Aviation Environmental Impact Analysis: A Survey of Recent Progress" *Aerospace* 9, no. 12: 750.
https://doi.org/10.3390/aerospace9120750