1. TÜBİTAK An Optimization Approach for Airport Ground Operations with A Shortest Path Algorithm 12 November 2015 Orhan Eroglu - TUBITAK BILGEM, Turkey Zafer Altug Sayar - TUBITAK BILGEM, Turkey Guray Yilmaz – TurAFA ASTIN, Turkey

2. TÜBİTAK Overview Introduction Floyd-Warshall Shortest Path Algorithm The FLOW-GRO System Conclusion Future Work

3. TÜBİTAK Introduction

4. Introduction TUBITAK, BILGEM: The Scientific and Technological Research Council of Turkey, Informatics and Information Security Research Center Department of Aeronautics  Air Traffic Control Studies Tower Simulator 3D IG Systems Ground Radar Control Ground Optimization Radar Simulator Approach/Enroute Radar Control

5. TÜBİTAK Introduction Ground Radar

6. TÜBİTAK Introduction Air traffic seems to going to have twice or three times the current density, in 2020’s [1] Ground route conflicts everywhere!  Develope new decision support systems [2] usable for controllers to manage ground traffic  Route conflicts and routing delays may occur more often when taxiing of vehicles is not performed in the shortest possible way [3]  Support human controllers by computer systems  Revise and optimize traffic control procedures and standards, improving monitoring abilities by development of new traffic surveillance and communication technologies  Harden the practices of air traffic controllers

7. TÜBİTAK Introduction In this study, a computer system is proposed, which generates the shortest path between two points during the taxi procedures of ground traffics

8. TÜBİTAK Floyd-Warshall Shortest Path Algorithm

9. TÜBİTAK Floyd-Warshall Shortest Path Alg. Finds shortest paths between all node pairs in a weighted graph. Easy to implement Not much challenging but proposes an appropriate application for the domain.

10. TÜBİTAK Floyd-Warshall Shortest Path Alg. Input:  The graph array: Holds the edge weights between directly connected nodes. Outputs:  Distance Matrix (D): Holds the distance of the shortest path between each node-pair, if exists.  Sequentiality Matrix (S): Holds the sequential relationship between nodes, i.e. it holds the last node before the terminal one, if it exists, in the shortest path between two nodes

11. TÜBİTAK Floyd-Warshall Shortest Path Alg. procedure [array] FloydWarshall(D, S) for k in 1 to n do for i in 1 to n do for j in 1 to n do if D[i][j] > D[i][k] + D[k][j] then D[i][j] = D[i][k] + D[k][j] S[i][j] = S[k][j] Return S

12. TÜBİTAK Floyd-Warshall Shortest Path Alg. Pro’s:  Easy to implement  Very suitable for the problem space (ground traffic control) since airports can be represented as weighted graphs  Can be applied for the whole airport at first rather than time of ground operations Con’s:  O( ) complexity  Number of iterations may drastically increase for huge airports with thousands of nodes

13. TÜBİTAK The FLOW-GRO System

14. TÜBİTAK The FLOW-GRO System Airports as a node-based, weighted graph [6]:  Parking positions  Aprons  Taxiway segments  Lineup areas  Runway points Detailed view of the airport graph

15. TÜBİTAK The FLOW-GRO System Performing the Floyd-Warshall algorithm on an airport graph:  To guarantee to find the shortest taxiway between any two airport nodes, mathematically The FLOW-GRO System has been developed  The most interesting proposal: Deciding the edge weights  Euclidean distance is the very first option  Taxiway speeds may vary Inappropriate  Our proposed weight: The minimum taxiing-time of each edge, which can be calculated by dividing the edge distance by the maximum allowed taxi speed

16. TÜBİTAK The FLOW-GRO System The FLOW-GRO system has been examined several times in the ATC simulator system developed by TUBITAK-BILGEM  Includes database models of 4 different airports in Turkey  Each consists of hundreds of nodes, i.e. varying speed taxiways.

17. TÜBİTAK Conclusion

18. Conclusion The study do not cope with the hardest challenges  Besides, it proposes the usage of a well-known method for a common problem in air traffic control:  Optimization of taxiway routings A simple application work of a node-based algorithm in a domain of more complex computational problems

19. TÜBİTAK Future Work

20. TÜBİTAK Future Work Inevitably, the system must perform dynamic conflict resolution in order to become more interesting  In addition to static computation of the shortest taxiways, the system may also determine whether conflict exists between different traffics dynamically It can be exploited for the sake of evaluating the ability of traffic controllers to generate the most appropriate taxiing routes over the ground movements  It may also propose the next possible shortest path for the conflicting ground traffics

21. TÜBİTAK References

22. Future Work 1.H. Erzberger, "Transforming the NAS: The next generation air traffic control system." 24th International Congress of the Aeronautical Sciences, Yokohama, Japan. 2004. 2.A. P. Sage, Decision support systems. John Wiley & Sons, Inc., 1991. 3.J. B. Gotteland et al. "Aircraft ground traffic optimization." ATM 2001, 4th USA/Europe Air Traffic Management Research and Development Seminar. 2001. 4.E. W. Weisstein, "Floyd-Warshall Algorithm.", 2008. 5.R W. Floyd, "Algorithm 97: shortest path." Communications of the ACM5.6: 345, 1962. 6.A. G. Marin, "Airport management: taxi planning." Annals of Operations Research 143.1: 191-202, 2006.

23. TÜBİTAK Thank you…