Purpose: In the cold supply chain (SC), effective risk management is regarded as an essential component to address the risky and uncertain SC environment in handling time- and temperature-sensitive products. However, existing multi-criteria decision-making (MCDM) approaches greatly rely on expert opinions for pairwise comparisons. Despite the fact that machine learning models can be customised to conduct pairwise comparisons, it is difficult for small and medium enterprises (SMEs) to intelligently measure the ratings between risk criteria without sufficiently large datasets. Therefore, this paper aims at developing an enterprise-wide solution to identify and assess cold chain risks.Design/methodology/approach: A novel federated learning (FL)-enabled multi-criteria risk evaluation system (FMRES) is proposed, which integrates FL and the best–worst method (BWM) to measure firm-level cold chain risks under the suggested risk hierarchical structure. The factors of technologies and equipment, operations, external environment, and personnel and organisation are considered. Furthermore, a case analysis of an e-grocery SC in Australia is conducted to examine the feasibility of the proposed approach.Findings: Throughout this study, it is found that embedding the FL mechanism into the MCDM process is effective in acquiring knowledge of pairwise comparisons from experts. A trusted federation in a cold chain network is therefore formulated to identify and assess cold SC risks in a systematic manner.Originality/value: A novel hybridisation between horizontal FL and MCDM process is explored, which enhances the autonomy of the MCDM approaches to evaluate cold chain risks under the structured hierarchy.
1 July 2003 in Journal of Materials Processing Technology
In view of increasing supply chain disruption events, for example the China–United States trade war, the COVID-19 pandemic, and the Russia–Ukraine war, the complexity and dynamicity of global freight management keeps increasing. To build a resilient and sustainable supply chain, industrial practitioners are eager to systematically revamp the freight management decision process related to the selection of carriers, shipping lanes, and third-party logistics service providers. Therefore, this study aims at strengthening decision-making capabilities for global freight management, in which an industrial blockchain-based global freight decision framework (IB-GFDF) is proposed to incorporate consortium blockchain technology with the Bayesian best-worst method. Through the blockchain technology, pairwise comparisons can be conducted over the international freight network in a decentralized and immutable manner, and thus, a secure and commonly agreed-on pairwise comparison dataset is acquired. Subsequently, the pairwise comparison dataset with multi-stakeholder opinions is analyzed using the Bayesian best-worst method in order to prioritize the selection decision criteria related to carriers, shipping lanes, and 3PL service providers for global freight management. To verify the methodological feasibility, a case study of an Australian agricultural supply chain firm was conducted to support the development end-to-end (E2E) supply chain solutions originated from Australia. It was found that port infrastructure, ports of call and communication effectiveness were the major criteria for the selection decision, which can be emphasized in future global freight collaboration. In addition, an immutable and append-only record of pairwise comparisons can be established to support the visibility of time-varying stakeholders’ preferences.
The fashion industry operates in a fast moving and dynamic environment which requires fashion designers to respond to market trends continuously. This paper investigates potential for application of Internet of Things (IoT) and the related state-of-the-art research efforts in fashion retail. Non-purchasing behaviors are seldom recorded in retail. What customers are interested in may not reflect totally in products they actually bought and are recorded in Point of Sales (POS). However, some of in-store behaviors may reflect their hidden preferences. These data were not easy to record in the past. This study is relevant to use of sensing devices that can capture in-store customer behaviors and transmit the same through wireless network to cloud. A data analytics model with fuzzy logic approach is developed to generate data of purchasing intentions. The paper fills the literature gap on the use of specific non-purchase data in retail stores. With the aid of fuzzy logic, behavioral intentions can be used by salespersons for recommending products to customers and guide supply chain planning. It also enriches the application of IoT in fashion retail environment. The findings from this research can be applied to other products in the retail industry.
1 July 2001 in Artificial Intelligence in Engineering