Big Data-driven Airport Resource Optimization and Management
Meeting the large and increasing throughput demand by managing limited resources is a critical problem in airport operation. We develop BigARM that is an airport resource management engine equipped with a series of data-driven models and algorithms to accurately predict the dynamics and make intelligent allocation decisions for optimizing airport resource utilization. Particularly in managing the reclaim belt while handling inbound baggage, BigARM first makes an initial allocation plan based on the upcoming flight schedules. Then, it performs real-time predictions of the flight arrival time, the terminating bag count, and the baggage reclaim profile. When adjustments to the current plans are needed based on the predictions, the engine will produce a new allocation plan and make adjustment recommendations by maximizing the balanced use of the reclaim belts. The predictions and recommendations are continuously made until the next daily routine. BigARM is the first time to fully adopt data-driven approaches for airport resource optimization and management. It well adapts to the high dynamics and uncertainty in the real airport operation environment when optimizing the resource allocation, thus leading to better performance.
- A gold medal at 2021 China (Shanghai) International Innovation and Invention Exhibition
- Jia Wang, Jiannong Cao, Senzhang Wang, Zhongyu Yao, and Wengen Li: IRDA: Incremental Reinforcement Learning for Dynamic Resource Allocation. IEEE Transactions on Big Data 2020
- Ka Ho Wong, Jiannong Cao, Yu Yang, Wengen Li, Jia Wang, Zhongyu Yao, Suyan Xu, Esther Ahn Chian Ku, Chun On Wong, and David Leung: BigARM: A Big-data-driven Airport Resource Management Engine and Application Tools. DASFAA 2020: 741-744
- Zhuo Li, Jiannong Cao, Zhongyu Yao, Wengen Li, Yu Yang, and Jia Wang: Recursive Balanced k-Subset Sum Partition for Rule-constrained Resource Allocation. ACM CIKM 2020: 2121-2124
Dr. Wengen Li, Yu Yang and Esther Ahn Chian Ku.
Previous members include Dr. Jia Wang, Dr. Zhuo Li, Ka Ho Wong, Zhongyu Yao, Suyan Xu, Ho Kuen Lee (Gavin), Cheuk Lun Lee (Alan), Jiandong Li, Sha Li, and Patrick Yau. Thanks for their contributions.