【摘要】 本文聚焦船员不同阶段的指标关联性，构建了基于船员岗前、出岗不同阶段多模态数据的适岗状态监测预警指标体系，建立了船员适岗状态监测预警的随机森林模型.结果表明：考虑船员岗前、出岗不同阶段多模态数据关联的适岗状态监测预警模型准确率达到94.4%,仅考虑船员单一阶段适岗状态监测预警模型准确率为80.6%;船员适岗状态影响程度较大的指标为疲劳程度(0.182 7)、工作压力(0.136 8)、水上服务资历(0.117 8)、血压(0.085 4)、工作状态(0.076 3)、岗前(0.070 8)和出岗心率(0.065 9).更多还原

【Abstract】 Focusing on the correlation of indicators in different stages of crew, a monitoring and early warning index system of crew’s suitability for work based on multimodal data in different stages of crew’s pre-job and post-job was constructed, and a random forest model for monitoring and early warning of crew’s suitability for work was established. The results show that the accuracy of the monitoring and early warning model of the crew’s suitability for work is 94.4% considering the multimodal data association of the crew in different stages before and after their posts, and the accuracy of the monitoring and early warning model of the crew’s suitability for work in a single stage is 80.6%. The indicators that have great influence on the crew’s suitability for duty are fatigue(0.182 7), working pressure(0.136 8), water service qualification(0.117 8), blood pressure(0.085 4), working status(0.076 3), pre-job(0.070 8) and off-job heart rate(0.065 9).更多还原