Final published version
Research output: Contribution to conference - Without ISBN/ISSN › Abstract › peer-review
Research output: Contribution to conference - Without ISBN/ISSN › Abstract › peer-review
}
TY - CONF
T1 - An evolutionary heuristic for solving the robust single airport slot allocation problem
AU - Pirogov, Aleksandr
AU - Zografos, K. G.
N1 - Conference code: 4
PY - 2023/4/19
Y1 - 2023/4/19
N2 - We propose an evolutionary heuristic method for solving the robust single airport slot allocation problem. The proposed method aims to discover a new solution by applying one or several mutation procedures to the current best solution. The best solution is defined by a fitness value that covers constraints violation parameters and allocation properties. Constraints’ violation parameters control the feasibility of the solution. Such structural aspects allow us to move towards acceptable results after each evolutionary iteration. Allocation properties work in a similar manner but defining a solution quality via indicators measuring slot allocation objectives used in the literature, such as, total displacement, maximum displacement, number of rejections, and number of displaced requests.The method’s implementation has a flexible design providing the range of parameters that can be set to change the algorithm’s way of work. The benchmarking is done for different number of iterations, different weights of fitness function components, list of allowed mutations (for both equivalent and different mutation probabilities), and number of mutations per iteration. The efficiency of the method is analyzed from both computational complexity and direct performance comparison to the exact method. The results show that heuristic methods adjusted for the problem structure could be used as a general replacement for exact approaches which require an enormous amount of time to solve even some small size instances.
AB - We propose an evolutionary heuristic method for solving the robust single airport slot allocation problem. The proposed method aims to discover a new solution by applying one or several mutation procedures to the current best solution. The best solution is defined by a fitness value that covers constraints violation parameters and allocation properties. Constraints’ violation parameters control the feasibility of the solution. Such structural aspects allow us to move towards acceptable results after each evolutionary iteration. Allocation properties work in a similar manner but defining a solution quality via indicators measuring slot allocation objectives used in the literature, such as, total displacement, maximum displacement, number of rejections, and number of displaced requests.The method’s implementation has a flexible design providing the range of parameters that can be set to change the algorithm’s way of work. The benchmarking is done for different number of iterations, different weights of fitness function components, list of allowed mutations (for both equivalent and different mutation probabilities), and number of mutations per iteration. The efficiency of the method is analyzed from both computational complexity and direct performance comparison to the exact method. The results show that heuristic methods adjusted for the problem structure could be used as a general replacement for exact approaches which require an enormous amount of time to solve even some small size instances.
M3 - Abstract
T2 - 4th IMA and OR Society Conference on Mathematics of Operational Research
Y2 - 27 April 2023 through 28 April 2023
ER -