快递业务系统货件派送异常自动侦测流程的研究

发布时间：2018-03-24 10:26

本文选题：物流派件异常侦测　切入点：异常预警　出处：《浙江工业大学》2014年硕士论文

【摘要】：当前,物流派件异常侦测已经成为各大企业愈加关注的领域,尤其是最近几年在企业电子商务领域中十分流行。不同于传统行业,随着电子商务的迅猛发展,物流作为近年来新兴的行业,有着自身的业务特色。异常侦测对于企业尤其是电商来说,货件的快速派送与货件派送的优质服务是提升物流效率和强化企业形象等方面的关键决胜点,但是当前物流中存在大量的丢包延误而被客户投诉的事件,可见物流派件跟踪的异常侦测需要改善。本课题根据目前的物流异常侦测系统或软件的特色和业务流程,通过对派件时间范围的设定加入异常预警、异常检测、数据分析和数据可视化等新功能,并利用Python Web技术,再开发了一个异常侦测系统。本文的工作主要包括以下几个方面： (1)详细介绍了物流派件异常侦测的发展过程和技术特点,总结了物流派件异常侦测相关功能,利用Python Web技术完善物流派件过程中出现异常可实现的侦测与告警。对目前市场上已有的物流ERP系统异常侦测系统或软件,进行了调研,综合总结它们各自的功能特点和侧重方向,作为本文开发的参考。 (2)在传统功能实现的基础上加入异常预警和数据分析模块,通过计算各信息节点的时间间隔,来预测是否可能存在异常,分析收集的异常数据,反馈异常预警结果,调节预警阀值,增加异常预警的准确性,从而实现对异常的预警,提升物流企业处理异常的及时性。 (3)学习和研究Python Web和其他Web开发的相关技术,包括：Python(服务器端编程语言),JavaScript(客户端编程语言),jQuery (JavaScript框架),Django (Python集成框架)。利用上述技术和框架,完成了物流异常侦测系统的开发,并对该系统进行了测试和评估。
[Abstract]:At present, the abnormal detection of material distribution has become an increasingly concerned area for large enterprises, especially in recent years, it is very popular in the field of enterprise e-commerce. Different from the traditional industry, with the rapid development of e-commerce, Logistics, as an emerging industry in recent years, has its own business characteristics. The quick delivery of goods and the high quality service of delivery of goods are the key decisive points in improving the efficiency of logistics and strengthening the image of the enterprise. However, there are a large number of incidents in the current logistics of losing packets and being complained by customers. It can be seen that the anomaly detection of logistics dispatch tracking needs to be improved. According to the characteristics and business process of the current logistics anomaly detection system or software, this subject adds anomaly warning and anomaly detection through setting the time range of the distribution. Based on the new functions of data analysis and data visualization, an anomaly detection system is developed using Python Web technology. The work of this paper mainly includes the following aspects:. In this paper, the development process and technical characteristics of abnormal detection of material distribution are introduced in detail, and the related functions of abnormal detection of logistics distribution are summarized. This paper makes use of Python Web technology to perfect the detection and alarm of abnormal in the process of logistics distribution. The existing abnormal detection system or software of logistics ERP system in the market is investigated. Summarize their respective functional characteristics and focus on the direction, as a reference for the development of this paper. 2) adding anomaly warning and data analysis module on the basis of traditional function realization, calculating the time interval of each information node to predict the possible existence of anomaly, analyzing the collected abnormal data, and feedback the abnormal warning result. Adjust the warning threshold, increase the accuracy of abnormal warning, so as to achieve the abnormal warning, improve the timeliness of logistics enterprises to deal with anomalies. 3) to study and study the relevant technologies of Python Web and other Web development, including: Python (client programming language / Java / Java / Java / Java / Java / Java / Java / Java / Java / Java / Java / Java / Java / Java / Java / Java / Java / Java / Java / Java / Java / Java / Java / Java / Java / Java). The system is tested and evaluated.
【学位授予单位】：浙江工业大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TP393.09

【参考文献】