Generating decision support for alarm processing in cold supply chains using a hybrid k-NN algorithm
作者：
Highlights：
• The problem of decision support generation for alarm processing is addressed.
• A hybrid algorithm is proposed based on k-NN, decision trees, and nearest neighbor.
• A way of representing chronological shipment location is proposed in a fuzzy set.
• 16,525 temperature alarms with associated contextual data are used in tests.
• The proposed algorithm consistently outperforms k-NN in our evaluations.
摘要
•The problem of decision support generation for alarm processing is addressed.•A hybrid algorithm is proposed based on k-NN, decision trees, and nearest neighbor.•A way of representing chronological shipment location is proposed in a fuzzy set.•16,525 temperature alarms with associated contextual data are used in tests.•The proposed algorithm consistently outperforms k-NN in our evaluations.
论文关键词：k-nearest neighbors,Fuzzy set,Recommendation,Decision Support,Pharmaceutical supply chain,Temperature deviation
论文评审过程：Received 7 September 2020, Revised 21 December 2020, Accepted 6 November 2021, Available online 17 November 2021, Version of Record 22 November 2021.
论文官网地址：https://doi.org/10.1016/j.eswa.2021.116208
