# Optimising Airport Ground Resource Allocation for Multiple Aircraft Using Machine Learning-Based Arrival Time Prediction

^{1}

^{2}

^{3}

^{*}

## Abstract

**:**

## 1. Introduction

## 2. Methodology

- 1
- Decision variables:

- 2
- Result variables:

- 3
- Uncontrollable variables:

- 4
- Optimisation function:

- 5
- Total resources required at time slot j:

- 6
- Revised number of aircraft due to delay:

- 7
- Aircraft resource allocation:

- 8
- Resource capacity constraint:

- 9
- Arrival time variation constraint:

- 10
- Binary variables:

#### 2.1. Arrival Time Prediction—Model Development Process

#### 2.2. Exploratory Data Analysis

#### 2.3. The Feature Engineering

#### 2.4. Multiple Linear Regression (MLR)

_{Delay}is the dependent variable or the output variable that we are trying to predict or explain, which represents the arrival delay in minutes for a flight.

_{0}is the intercept or constant term of the model, which represents the expected value of ARR

_{Delay}when all the independent variables are zero. β

_{1}–β

_{11}are the regression coefficients for the respective independent variables, which indicate the strength and direction of the relationship between each variable and the ARR

_{Delay}.

_{DAY}is a categorical variable representing the day of the week, which is typically encoded as a set of binary variables (e.g., 0/1 for Monday/Tuesday). Flight ID

_{2}, Origin 2, and Destination 2 are categorical variables representing the flight ID, origin airport, and destination airport, respectively, which are typically encoded using dummy variables. SFT, ATD_1, DEP_DEL_CAT_1, DEP_DEL_1, STA_2, AIR_CRAFT_2, and EMA_FT are continuous variables representing the scheduled flight time, actual time of departure from origin airport 1, departure delay category, departure delay, scheduled time of arrival, aircraft type, and estimated maximum altitude, respectively. ε is the error term or residual, which represents the variability or uncertainty in the model that is not explained by the independent variables.

_{Delay}. This enables us to identify the most important independent variables that are associated with ARR

_{Delay}and make predictions of ARR

_{Delay}for new flights based on the values of the independent variables.

#### 2.5. Multilayer Perceptron (MLP)

^{th}hidden layer, denoted as ${H}^{l}={h}^{{l}_{1}},{h}^{{l}_{2}}\dots ,{h}^{{l}_{N-1}}$, can be obtained using the following equation:

^{th}layer, and g is the activation function applied element-wise to the linear combination of the input features or the outputs of the previous layer.

^{th}unit in the l

^{th}layer.

## 3. Results

#### 3.1. Scenario 1 (One Origin–One Destination—1 Round-trip)

#### 3.2. Scenario 2 (One Origin–Multiple Destination—Multiple Round-trip)

#### 3.3. Classification of Arrival Category

## 4. Discussion

## 5. Conclusions

## Author Contributions

## Funding

## Data Availability Statement

## Acknowledgments

## Conflicts of Interest

## References

- AllThingsOnTime. Solutions for Improving Ground Operations Efficiency—All Things On Time, 20 October 2022. Available online: https://www.allthingsontimeperformance.com/9-solutions-for-improving-ground-operations-efficiency/ (accessed on 15 October 2022).
- Evler, J.; Asadi, E.; Preis, H.; Fricke, H. Airline ground operations: Schedule recovery optimization approach with constrained resources. Transp. Res. Part C Emerg. Technol.
**2021**, 128, 103129. [Google Scholar] [CrossRef] - Jan, E.; Ehsan, A.; Henning, P.; Hartmut, F. Airline ground operations: Optimal schedule recovery with uncertain arrival times. J. Air Transp. Manag.
**2021**, 92, 969–6997. [Google Scholar] - Wu, C.; Law, K. Modelling the delay propagation effects of multiple resource connections in an airline network using a Bayesian network model. Transp. Res. Part E: Logist. Transp.
**2019**, 122, 62–77. [Google Scholar] [CrossRef] - Wu, C.-L. Airline Operations and Delay Management: Insights from Airline Economics, Networks and Strategic Schedule Planning; Routledge: Oxfordshire, UK, 2016. [Google Scholar]
- Gök, Y.S.; Padrón, S.; Tomasella, M.; Guimarans, D.; Ozturk, C. Constraint-based robust planning and scheduling of airport apron operations through simheuristics. Ann. Oper. Res.
**2023**, 320, 795–830. [Google Scholar] [CrossRef] - Deepudev, S.; Palanisamy, P.; Varun, P.G.; Manjunath, K.N.; Asok, K. Prediction of gate in time of scheduled flights and schedule conformance using machine learning-based algorithms. Int. J. Aviat. Aeronaut. Aerosp.
**2020**, 7, 9. [Google Scholar] - Guleria, Y.; Cai, Q.; Sameer, A.; Li, L. A multi-agent approach for reactionary delay prediction of flights. IEEE Access
**2019**, 7, 181565–181579. [Google Scholar] [CrossRef] - Codina, D.; Ramon, S.; Belkoura; Herbert, N.; Franck, B.; Wagnick, S. Improving the predictability of take-off times with Machine Learning: A case study for the Maastricht upper area control centre area of responsibility. In Proceedings of the 9th SESAR Innovation Days, Athens, Greece, 2–5 December 2019; pp. 1–8. [Google Scholar]
- Deepudev, S.; Palanisamy, P.; Gopi, V.; Guruswami, S.; Adithya, K.K. A machine learning-based approach to predict random variation in the landing time of scheduled flights. Int. J. Sustain. Aviat.
**2021**, 7, 293–318. [Google Scholar] - Jiang, Y.; Liu, Y.; Liu, D.; Song, H. Applying Machine Learning to Aviation Big Data for Flight Delay Prediction; IEEE: Piscataway, NJ, USA, 2020. [Google Scholar]
- Ding, Y. Predicting flight delay based on multiple linear regression. IOP Conf. Ser. Earth Environ. Sci.
**2017**, 81, 012198. [Google Scholar] [CrossRef] - Catur, S.; Rafrastara, F.A.; Astuti, Y.P.; Kencana, L.I. Clustered Logistic Regression Algorithm for Flight Delay Prediction. Int. J. Comput. Sci. Inf. Secur. IJCSIS
**2021**, 2, 19. [Google Scholar] - Hajar, A.; Lahcen, M.; Youssef, B. A multilayer perceptron neural network with selective-data training for flight arrival delay prediction. Sci. Program. Hindawi
**2021**, 2021, 12. [Google Scholar] - Deepudev, S.; Palanisamy, P.; Varun, P.G.; Manjunath, K.N. Ground-based 4d trajectory prediction using bi-directional LSTM networks. Appl. Intell.
**2022**, 52, 16417–16434. [Google Scholar] - Kim, Y.J.; Choi, S.; Simon, B.; Dimitri, M. A deep learning approach to flight delay prediction. In Proceedings of the 2016 IEEE/AIAA 35th Digital Avionics Systems Conference (DASC), Sacramento, CA, USA, 25–29 September 2016. [Google Scholar]
- Zhang, J.; Peng, Z.; Yang, C.; Wang, B. Data-driven flight time prediction for arrival aircraft within the terminal area. IET Intell. Transp. Syst.
**2022**, 16, 263–275. [Google Scholar] [CrossRef] - Lim, Z.J.; Sim, K.G.; Imen, D.; Sameer, A. Causal effects of landing parameters on runway occupancy time using causal machine learning models. In Proceedings of the 2020 IEEE Symposium Series on Computational Intelligence (SSCI), Canberra, ACT, Australia, 1–4 December 2020; pp. 2713–2722. [Google Scholar]
- Chow, H.W.; Jun, L.Z.; Sameer, A. Data-driven Runway Occupancy Time Prediction using Decision Trees. In Proceedings of the IEEE/AIAA 40th Digital Avionics Systems Conference (DASC), San Antonio, TX, USA, 3–7 October 2021; pp. 1–9. [Google Scholar]
- Shai, S.-S.; Shai, B.-D. Understanding Machine Learning: From Theory to Algorithms; Cambridge University Press: Cambridge, UK, 2014. [Google Scholar]
- Bojia, Y.; Liu, B.; Yong, T.; Wan, L. A methodology for predicting aggregate flight departure delays in airports based on supervised learning. Sustainability
**2020**, 12, 7. [Google Scholar] - Airport Technology. Airport Technology: Ground Handling Suite (GHS) for Improved Efficiency of Airport Operations. 2020. Available online: www.airport-technology.com (accessed on 15 October 2022).
- Offord, R. Study on Airport Ownership and Management and the Ground Handling Market in Selected Non-EU Countries; No. MOVE/E1/SER/2015-247-3; Steer Davies Gleave: London, UK, 2016. [Google Scholar]
- Kafle, N.; Zou, B. Modeling flight delay propagation: A new analytical-econometric approach. Transp. Res. Part B Methodol.
**2016**, 93, 520–542. [Google Scholar] [CrossRef] - Boysen, N.; Malte, F. Scheduling aircraft landings to balance workload of ground staff. Comput. Ind. Eng.
**2011**, 60, 206–217. [Google Scholar] [CrossRef] - Maragno, D.; Wiberg, H.; Bertsimas, D.; Birbil, S.I.; den Hertog, D.; Fajemisin, A. Mixed-integer optimization with constraint learning. arXiv
**2021**, arXiv:2111.04469. [Google Scholar] - Mitchell, S.; OSullivan, M.; Dunning, I. PuLP: A linear programming toolkit for python. Univ. Auckl. Auckl. N. Z.
**2011**, 65. [Google Scholar] - Hart, W.E.; Watson, J.-P.; Woodruff, D.L. Pyomo: Modeling and solving mathematical programs in Python. Math. Program. Comput.
**2011**, 3, 219–260. [Google Scholar] [CrossRef] - Cheong, Y.; Koo, C. Analysis of airline on-time performance with multivariate adaptive regression splines. Expert Syst. Appl.
**2009**, 36, 1919–1926. [Google Scholar] - Aoki, M.; Tsuruta, K.; Wakabayashi, Y.; Sawada, H. Predicting flight arrival times with machine learning methods. In Proceedings of the IEEE Symposium Series on Computational Intelligence (SSCI), Athens, Greece, 6–9 December 2016; pp. 1–8. [Google Scholar]
- International Airports Council. 11 April 2022. Available online: https://aci.aero/2022/04/11/the-top-10-busiest-airports-in-the-world-revealed/ (accessed on 4 January 2023).
- Flightradar24. 15 December 2022. Available online: https://www.flightradar24.com/ (accessed on 25 December 2022).
- Herbers, J.; Juraj, H. Models and algorithms for ground staff scheduling on airports. Fakultät für Mathematik, Informatik und Naturwissenschaften No. pp. RWTH–CONV-121335, RWTH Aachen Technical University, 52062 Aachen, 2005. Available online: http://publications.rwth-aachen.de/record/59558/files/Herbers_Joerg.pdf (accessed on 18 May 2022).
- Wheeler, C.F. Strategies for maximizing the profitability of airline hub-and-spoke networks. Transp. Res. Rec.
**1989**, 1214, 1–9. [Google Scholar] - Tranmer, M.; Mark, E. Multiple linear regression. Cathie Marsh Cent. Census Surv. Res. CCSR
**2008**, 5, 1–5. [Google Scholar] - Suzuki, K. Artificial Neural Networks: Methodological Advances and Biomedical Applications; BoD–Books on Demand: Norderstedt, Germany, 2011. [Google Scholar]
- Syarif, I.; Prugel-Bennett, A.; Wills, G. SVM parameter optimization using grid search and genetic algorithm to improve classification performance. TELKOMNIKA Telecommun. Comput. Electron. Control.
**2016**, 14, 1502–1509. [Google Scholar] [CrossRef] - Holt, C.C. Forecasting seasonals and trends by exponentially weighted moving averages. Int. J. Forecast.
**2004**, 20, 5–10. [Google Scholar] [CrossRef] - Qu, J.; Ting, Z.; Meng, Y.; Li, J.; Liu, C. Flight delay prediction using deep convolutional neural network based on fusion of meteorological data. Neural Process. Lett.
**2020**, 52, 1461–1484. [Google Scholar] [CrossRef] - Thiagarajan, B.; Lakshminarasimhan, S.; Sharma, A.V.; Dinesh, S.; Vineeth, V. A machine learning approach for prediction of on-time performance of flights. In Proceedings of the 2017 IEEE/AIAA 36th Digital Avionics Systems Conference (DASC), St. Petersburg, FL, USA, 17–21 September 2017. [Google Scholar]

**Figure 1.**Schematic overall Gantt chart conceptual view created for turnarounds with on time, delayed, and early arrivals with its impacts on turnaround time and outbound flight.

**Figure 2.**A sample comparison between number of scheduled flight departures and actual flight departures during the hours of a day at Dubai International Airport (based on data from Flightradar24).

**Figure 3.**A sample comparison between number of scheduled flight arrival and actual arrival during the hours of a day at Dubai International Airport (prepared based on actual data collected).

**Figure 4.**A comparison between number of flights scheduled as per STA and actual number of flights landed as per ATA between 00:00 and 08:00 h.

**Figure 6.**Residual plots: (

**a**) heteroscedasticity plot: The green dotted line represents the percentile boundary of the distribution, indicates the upper or lower limit within which the residuals are expected to fall, deviations suggest the presence of heteroscedasticity; (

**b**) Q-Q plot: The 45-degree red diagonal line represents the expected distribution based on theoretical assumptions. Deviations observed at the beginning and end points of the plot suggest departures from the expected distribution.

**Figure 7.**MLP architecture for arrival delay time prediction with 1 input layer, 3 hidden layers, and 1 output layer.

**Figure 8.**MLP architecture for arrival delay classification with 1 input layer, 3 hidden layers, and 1 output layer.

**Figure 9.**A graphical comparison highlighting the average turnaround time reliance on the early, on time, and delay arrival classes, with delayed arrival taking the shortest average turnaround time and early arrival having the longest average turnaround time for the majority of aircraft types.

**Table 1.**Comparative analysis of arrival time variation and its impact on corresponding departure of Dubai International Airport.

Arrival | Early | OnTime | Delay | ||||||
---|---|---|---|---|---|---|---|---|---|

Departure | Delay | Early | On Time | Delay | Early | On Time | Delay | Early | On Time |

Hours | |||||||||

12 a.m. | 11% | 0% | 89% | 24% | 0% | 76% | 78% | 0% | 22% |

01 a.m. | 16% | 0% | 84% | 27% | 1% | 72% | 87% | 1% | 12% |

02 a.m. | 14% | 4% | 82% | 26% | 1% | 72% | 70% | 0% | 30% |

03 a.m. | 13% | 0% | 87% | 17% | 0% | 83% | 63% | 0% | 37% |

04 a.m. | 6% | 3% | 91% | 16% | 1% | 83% | 60% | 0% | 40% |

05 a.m. | 33% | 0% | 67% | 26% | 0% | 74% | 78% | 0% | 22% |

06 a.m. | 13% | 0% | 87% | 21% | 0% | 79% | 65% | 0% | 35% |

07 a.m. | 10% | 0% | 90% | 16% | 0% | 84% | 72% | 0% | 28% |

08 a.m. | 20% | 0% | 80% | 27% | 0% | 73% | 88% | 0% | 12% |

09 a.m. | 9% | 3% | 88% | 16% | 1% | 83% | 70% | 0% | 30% |

10 a.m. | 9% | 0% | 90% | 20% | 0% | 80% | 63% | 0% | 37% |

11 a.m. | 8% | 2% | 91% | 25% | 1% | 74% | 80% | 0% | 20% |

12 p.m. | 8% | 1% | 91% | 20% | 0% | 80% | 69% | 0% | 31% |

01 p.m. | 11% | 1% | 88% | 21% | 0% | 79% | 84% | 0% | 16% |

02 p.m. | 18% | 0% | 82% | 22% | 0% | 78% | 79% | 0% | 21% |

03 p.m. | 15% | 0% | 85% | 23% | 1% | 76% | 82% | 0% | 18% |

04 p.m. | 6% | 1% | 93% | 19% | 0% | 80% | 81% | 0% | 19% |

05 p.m. | 10% | 1% | 89% | 35% | 0% | 65% | 88% | 0% | 12% |

06 p.m. | -% | 1% | 93% | 19% | 0% | 80% | 84% | 0% | 16% |

07 p.m. | 10% | 0% | 90% | 17% | 0% | 83% | 55% | 0% | 45% |

08 p.m. | 10% | 1% | 89% | 20% | 0% | 80% | 71% | 0% | 29% |

09 p.m. | 18% | 0% | 82% | 25% | 0% | 75% | 68% | 0% | 32% |

10 p.m. | 14% | 0% | 85% | 28% | 1% | 72% | 91% | 0% | 9% |

11 p.m. | 0% | 0% | 100% | 75% | 0% | 25% | 100% | 0% | 0% |

Day | 11% | 1% | 88% | 22% | 0% | 78% | 74% | 0% | 26% |

SN. | Feature | Description |
---|---|---|

1 | WEEK_DAY | weekday in which the flight was performed |

2 | Aircraft ID_2 | schedule ID from Origin 2 |

3 | Origin 2 | airport of second departure |

4 | Destination 2: | airport of arrival |

5 | SFT | scheduled flying time |

6 | ATD_1: Actual DEP | actual departure time |

7 | DEP_DEL_CAT_1 | departure delay category of Origin 1 |

8 | DEP_DEL_1 | departure delay in minutes of Origin 1 |

9 | STA_2 | scheduled arrival time at same departure airport |

10 | AICRAFT_2 | type of aircraft |

11 | EMA_FT | exponential moving average of flying time |

12 | ARR delay | arrival delay in minutes (one of our dependent variables) |

13 | ARV_DEL_CAT | classification arrival delay (one of our dependent variables) |

**Table 3.**Sample prediction results with relevant fields for scheduled aircraft arrival time predicted by proposed model (Predicted ATA_2) and actual arrival time (ATA_2) at Dubai airport on various days.

ID | DATE | Orgin | Dest | Aircraft | Flight Time | STD | ATD | Dep Delay_1 | ID_2 | Dest_2 | Flight Time_2 | STA_2 | ATA_2 | Predicted ATA_2 |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|

XX | 25/09/2022 | DXB | CGK | B77W | 08:02:00 | 00:10:00 | 00:37:00 | 27 | XY | DXB | 07:15:00 | 18:55:00 | 18:19:00 | 18:30:00 |

XX | 24/09/2022 | DXB | CGK | B77W | 08:04:00 | 00:10:00 | 00:26:00 | 16 | XY | DXB | 07:11:00 | 18:55:00 | 18:29:00 | 18:24:00 |

XX | 23/09/2022 | DXB | CGK | B77W | 07:58:00 | 00:10:00 | 00:26:00 | 16 | XY | DXB | 07:13:00 | 18:55:00 | 18:14:00 | 18:15:00 |

XX | 22/09/2022 | DXB | CGK | B77W | 08:02:00 | 00:10:00 | 00:39:00 | 29 | XY | DXB | 07:12:00 | 18:55:00 | 18:13:00 | 18:18:00 |

XX | 21/09/2022 | DXB | CGK | B77W | 08:06:00 | 00:10:00 | 00:22:00 | 12 | XY | DXB | 07:11:00 | 18:55:00 | 18:17:00 | 18:21:00 |

XX | 20/09/2022 | DXB | CGK | B77W | 07:54:00 | 00:10:00 | 00:24:00 | 14 | XY | DXB | 07:18:00 | 18:55:00 | 18:28:00 | 18:27:00 |

XX | 19/09/2022 | DXB | CGK | B77W | 07:49:00 | 00:10:00 | 00:21:00 | 11 | XY | DXB | 07:22:00 | 18:55:00 | 18:22:00 | 18:30:00 |

XX | 18/09/2022 | DXB | CGK | B77W | 07:55:00 | 00:10:00 | 00:28:00 | 18 | XY | DXB | 07:21:00 | 18:55:00 | 18:24:00 | 18:20:00 |

Variable | t-Value | p-Value |
---|---|---|

ATD_MIN_1 | 3.608071 | 3.41 × 10^{−4} |

SFT_MIN_DIFF_1 | −0.05753 | 9.54 × 10^{−1} |

DEP_DEL_1 | 5.810073 | 1.14 × 10^{−8} |

SFT_MIN_2 | −2.858086 | 4.45 × 10^{−3} |

RT_SFT | −2.252719 | 2.47 × 10^{−2} |

EMA_FT_0_5 | −2.231050 | 9.82 × 10^{−2} |

WEEK_DAY | 0.683288 | 4.95 × 10^{−1} |

AIRCRAFT_1_A388 | −2.050793 | 4.08 × 10^{−2} |

AIRCRAFT_1_B38M | −3.999751 | 7.35 × 10^{−5} |

AIRCRAFT_1_B39M | −3.984051 | 7.83 × 10^{−5} |

AIRCRAFT_1_B738 | −3.995765 | 7.47 × 10^{−5} |

AIRCRAFT_1_B77W | −2.114035 | 3.50 × 10^{−2} |

AIRCRAFT_2_A388 | −2.050793 | 4.08 × 10^{−2} |

AIRCRAFT_2_B38M | −3.999751 | 7.35 × 10^{−5} |

AIRCRAFT_2_B39M | −3.984051 | 7.83 × 10^{−5} |

AIRCRAFT_2_B738 | −3.995765 | 7.47 × 10^{−5} |

AIRCRAFT_2_B77W | −2.114035 | 3.50 × 10^{−2} |

DEP_DEL_CAT_1_DELAY | −2.953275 | 3.30 × 10^{−3} |

DEP_DEL_CAT_1_ONTIME | −2.906892 | 3.82 × 10^{−3} |

FLIGHT_ID_2_XXE2 | −1.416237 | 1.57 × 10^{−1} |

FLIGHT_ID_2_XXE6 | −2.93281 | 3.52 × 10^{−3} |

FLIGHT_ID_2_XXF6 | −4.036457 | 6.32 × 10^{−5} |

MLR | MLP | |
---|---|---|

Correlation coefficient (R2) | 0.698 | 0.6406 |

MAE | 7.7423 | 9.2085 |

RMSE | 10.0768 | 11.188 |

Study | Model | Number of Features | Number of Flight Legs Ahead | Accuracy (%) |
---|---|---|---|---|

Thiagarajan et al. [40] | Gradient Boosting with Binary Classification (Delay/No Delay) | 16 | 1 | 94.3 |

Qu et al. [39] | SE-DenseNet with Multiclass | 81 | 1 | 93.19 |

Proposed Model | MLP with Multiclass (Early, On Time, and Delay) | 11 | 2 | 93.57 |

**Table 7.**Comparison of variation of ground resource (manpower) requirements based on scheduled and actual flight movements.

As Per Scheduled Movement | As Per Actual Flight Movement | |||||
---|---|---|---|---|---|---|

Aircraft Type | Flight Count | Manpower/Flight | Total Manpower | Flight Count | Manpower/Flight | Total Manpower |

A20N | 4 | 13 | 52 | 4 | 13 | 52 |

A320 | 3 | 13 | 39 | 3 | 13 | 39 |

B38M | 3 | 13 | 39 | 4 | 13 | 52 |

B738 | 3 | 13 | 39 | 3 | 13 | 39 |

B772 | 1 | 26 | 26 | 1 | 26 | 26 |

B788 | 1 | 26 | 26 | 1 | 26 | 26 |

A388 | 0 | 1 | 27 | 27 | ||

SU95 | 0 | 1 | 8 | 8 | ||

Total Manpower (ground resource) for the hour | 221 | 269 | ||||

% of additional manpower required for the hour due to the actual arrival time variation | 22% |

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |

© 2023 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**

Sahadevan, D.; Al Ali, H.; Notman, D.; Mukandavire, Z.
Optimising Airport Ground Resource Allocation for Multiple Aircraft Using Machine Learning-Based Arrival Time Prediction. *Aerospace* **2023**, *10*, 509.
https://doi.org/10.3390/aerospace10060509

**AMA Style**

Sahadevan D, Al Ali H, Notman D, Mukandavire Z.
Optimising Airport Ground Resource Allocation for Multiple Aircraft Using Machine Learning-Based Arrival Time Prediction. *Aerospace*. 2023; 10(6):509.
https://doi.org/10.3390/aerospace10060509

**Chicago/Turabian Style**

Sahadevan, Deepudev, Hannah Al Ali, Dorian Notman, and Zindoga Mukandavire.
2023. "Optimising Airport Ground Resource Allocation for Multiple Aircraft Using Machine Learning-Based Arrival Time Prediction" *Aerospace* 10, no. 6: 509.
https://doi.org/10.3390/aerospace10060509