Preference-based evolutionary algorithm for airport surface operations. (June 2018)

• Record Type: Journal Article
• Authors: Weiszer, Michal; Chen, Jun; Stewart, Paul; Zhang, Xuejun
• Is Part Of: Transportation research. Volume 91(2018)
• Page Start: 296
• Page End: 316
• Publication Date: 2018-06
• Subjects: Airport ground operations -- Runway scheduling -- Multiobjective optimisation -- Preference search; Transportation -- Periodicals; Transportation -- Technological innovations -- Periodicals; 388.011
• DOI: 10.1016/j.trc.2018.04.008
• Languages: English
• ISSNs: 0968-090X

Abstract

Highlights: An EMO framework with approximate preferences is proposed. Interval preference information defines the extent of the region of interest. A filtering procedure finds a uniformly-distributed subset of solutions. The algorithm is applied to the ground movement and runway scheduling problem. Abstract: In addition to time efficiency, minimisation of fuel consumption and related emissions has started to be considered by research on optimisation of airport surface operations as more airports face severe congestion and tightening environmental regulations. Objectives are related to economic cost which can be used as preferences to search for a region of cost efficient and Pareto optimal solutions. A multi-objective evolutionary optimisation framework with preferences is proposed in this paper to solve a complex optimisation problem integrating runway scheduling and airport ground movement problem. The evolutionary search algorithm uses modified crowding distance in the replacement procedure to take into account cost of delay and fuel price. Furthermore, uncertainty inherent in prices is reflected by expressing preferences as an interval. Preference information is used to control the extent of region of interest, which has a beneficial effect on algorithm performance. As a result, the search algorithm can achieve faster convergence and potentially better solutions. A filtering procedure is further proposed to select an evenly distributed subset of Pareto optimal solutions in order to reduce its size and help the decision maker. The computational results with data from major international hub airports show the efficiency of the proposed approach.