基于机器学习的分拨中心安全评估与风险预警研究

Research on Safety Assessment and Risk Warning of Distribution Centers Based on Machine Learning

分拨中心的分拣作业是快递企业成本最高的环节,分拨中心的安全运营和管理也是影响服务质量的关键因素。本文探讨了分类中心的安全控制问题。仿真结果表明,在分类性能指标方面,LSTM-CNN-Attention模型的Recall、Accuracy和F1分数均优于LSTM和CNN模型。此外,与LSTM和CNN模型的分类性能指标相比,LSTM-CNN-Attention模型的Recall、Accuracy和F1分数都有显著提高,证明LSTM-CNN-Attention模型可用于配送中心的数据分类处理。该模型在处理快递分拣环节数据时表现更佳,证实了LSTM-CNN-Attention模型的优越性。The sorting operation of the distribution center is the most costly part of the courier enterprise, and the safe operation and management of the distribution center is also a key factor affecting the quality of service. This paper discusses the security control problem of the sorting center. The simulation results show that the Recall, Accuracy and F1 scores of the LSTM-CNN-Attention model outperform those of the LSTM and CNN models in terms of classification performance metrics. In addition, the Recall, Accuracy and F1 scores of the LSTM-CNN-Attention model are significantly improved compared with the classification performance metrics of the LSTM and CNN models, proving that the LSTM-CNN-Attention model can be used for data classification processing in distribution centers. The model performs better when processing data from courier sorting sessions, confirming the superiority of the LSTM-CNN-Attention model.