Intelligent sustainable supplier selection using multi-agent technology: Theory and application for Industry 4.0 supply chains
Intelligent sustainable supplier selection using multi-agent technology: Theory and application for Industry 4.0 supply chains
复制标题
使用多代理技术的智能可持续供应商选择：工业4.0供应链的理论与应用
DOI：
10.1016/j.cie.2018.10.050
复制
发表时间：
2019-01-01
影响因子：
7.9
通讯作者：
Heavey, Cathal
中科院分区：
文献类型：
--
作者：
Ghadimi, Pezhman;Wang, Chao;Heavey, Cathal
信息和通信系统的进步为供应链的智能化和自主性提供了巨大的机会，为工业4.0供应链(SC)奠定了基石。作为供应链的一项重要决策，可持续供应商评价和选择过程在以往的文献中已经有了大量的论述。然而，这一过程尚未在工业4.0供应链中实现，其中互联、实时信息透明、技术援助和物理系统成员(即供应链成员)的分散化被视为主要设计原则。为了缩小这种差距，提出了一种多智能体系统(MASS)方法来解决供应商的可持续评估和选择过程，以便在供应商和制造商之间提供适当的沟通渠道、结构化的信息交换和可见性。此外，详细研究了MASS在这一过程中的应用，以及它们作为向工业4.0供应链过渡的使能技术之一的自然适用性。结果表明，该方法可以帮助制造企业内部的决策者在较少的人与人交互的情况下快速做出决策。通过在一家医疗器械制造商上的实际实施，展示了所开发的MAS的优点。最后，给出了该方法的局限性和优点，并对今后的工作提出了一些看法。
Advancements in information and communication systems offer immense opportunities for supply chain intelligence and autonomy establishing stepping stones for Industry 4.0 supply chains (SCs). As a crucial SC decision, sustainable supplier evaluation and selection process have been addressed abundantly in the previous literature. However, this process has not yet been realized within Industry 4.0 SCs where interconnection, real-time information transparency, technical assistance and decentralization of members of a physical system (i.e., supply chain members) are regarded as the main design principles. To narrow the identified gap, a Multi-Agent Systems (MASs) approach is proposed for addressing sustainable supplier evaluation and selection process to provide a proper communication channel, structured information exchange and visibility among suppliers and manufacturers. Furthermore, the application of MASs in this process and their natural applicability as one of the enabling technologies in moving towards Industry 4.0 SCs are investigated in detail. It is found that the proposed approach can help decision-makers inside manufacturing firms to make prompt decisions with less human interactions. The merit of the developed MAS is demonstrated through a real-world implementation on a medical device manufacturer. Finally, the limitations and advantages of the proposed approach are presented together with some remarks for future work.