Data-driven operational risk analysis in E-Commerce Logistics. (April 2019)

• Record Type: Journal Article
• Authors: Xu, Gangyan; Qiu, Xuan; Fang, Meng; Kou, Xiaofei; Yu, Ying
• Is Part Of: Advanced engineering informatics. Volume 40(2019)
• Page Start: 29
• Page End: 35
• Publication Date: 2019-04
• Subjects: E-Commerce Logistics -- Operational risks -- Data analytics -- Risk analysis -- Gaussian mixture model; Computer-aided engineering -- Periodicals; Engineering -- Data processing -- Periodicals; 620.00285
• DOI: 10.1016/j.aei.2019.03.001
• Languages: English
• ISSNs: 1474-0346

Abstract

Abstract: The efficiency of E-Commerce Logistics (ECL) has become a major success factor for e-commerce companies in the competitive marketplace nowadays. However, the operation of ECL is complex and vulnerable to many risks, which would severely threaten its performance. A clear understanding of these risks would benefit a lot for conducting targeted measures to effectively mitigate their adverse effects. Therefore, this paper proposes a quantitatively analysis approach for operational risks in ECL based on extensive historical e-commerce transaction data. More specifically, the typical operation process of ECL is extracted through sequential analysis of key activities. After that, taking operation time as the key performance indicator, the performance patterns of different operation phases are analyzed. Then, considering the diverse distributions of operation time in different phases, especially the multimodal distribution of transportation time, a Gaussian Mixture Model (GMM) based risk analysis approach is proposed. Finally, an experimental case study is provided to measure the operational risks using real-life ECL data, and several managerial implications are also discussed based on the results.