Reverse logistics cost prediction Based on BP neural network

Jianchang Liu*, Wenjing Wei, Xiaojie Gu

*Corresponding author for this work

    Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

    1 Citation (Scopus)

    Abstract

    The reverse logistics cost forecast is the basis of reverse logistics cost control, and the key link related to the success of reverse logistics management. The factors that influence reverse logistics cost are complicated and numerous. The paper begins with analysis of the composition of reverse logistics, introduces the relations between reverse logistics cost and its influence factors, builds the reverse logistics cost prediction Based on BP neural network. An empirical study was carried out on a product of a company. Then simulates the model by Matlab and acquires good forecasting results. It is expected to provide some reference for enterprises to implement reverse logistics cost management.

    Original languageEnglish
    Title of host publication2011 2nd International Conference on Artificial Intelligence, Management Science and Electronic Commerce, AIMSEC 2011 - Proceedings
    Pages5139-5141
    Number of pages3
    DOIs
    Publication statusPublished - 2011
    Event2011 2nd International Conference on Artificial Intelligence, Management Science and Electronic Commerce, AIMSEC 2011 - Zhengzhou, China
    Duration: 8 Aug 201110 Aug 2011

    Publication series

    Name2011 2nd International Conference on Artificial Intelligence, Management Science and Electronic Commerce, AIMSEC 2011 - Proceedings

    Conference

    Conference2011 2nd International Conference on Artificial Intelligence, Management Science and Electronic Commerce, AIMSEC 2011
    Country/TerritoryChina
    CityZhengzhou
    Period8/08/1110/08/11

    Keywords

    • BP neural network
    • Cost forecast
    • Reverse logistics

    Fingerprint

    Dive into the research topics of 'Reverse logistics cost prediction Based on BP neural network'. Together they form a unique fingerprint.

    Cite this