Supervised learning for maritime search operations: An artificial intelligence approach to search efficiency evaluation

作者：

Highlights：

• We develop metamodels to compute a search and rescue mission success probability.

• Our metamodels, based on supervised learning, can partly replace costly simulations.

• We combine the metamodels with a heuristic to improve an operational DSS.

• The approach, evaluated on real-life data, can recommend high quality plans faster.

• Our approach can contribute to saving lives.

摘要

•We develop metamodels to compute a search and rescue mission success probability.•Our metamodels, based on supervised learning, can partly replace costly simulations.•We combine the metamodels with a heuristic to improve an operational DSS.•The approach, evaluated on real-life data, can recommend high quality plans faster.•Our approach can contribute to saving lives.

论文关键词：Search and rescue,Search theory,Supervised learning,Decision support system,Metamodeling,Metasimulation

论文评审过程：Received 20 November 2021, Revised 28 April 2022, Accepted 10 June 2022, Available online 13 June 2022, Version of Record 17 June 2022.

论文官网地址：https://doi.org/10.1016/j.eswa.2022.117857