Assessing the adoption barriers for the AI in food supply chain finance applying a hybrid interval-valued Fermatean fuzzy CRITIC-ARAS model

Abstract

Abstract The identification and evaluation of barriers to artificial intelligence (AI) adoption in food supply chain finance (FSCF) can be addressed as a multiattribute decision-making problem. However, only a few studies have reported the application of decision models for evaluating barriers to the implementation of AI in FSCF, especially within an uncertain context. Hence, this work explores the evaluation issue of implementation barriers via an integrated decision model. In this model, the conventional additive ratio assessment (ARAS) model integrated with the Choquet integral and criteria importance through intercriteria correlation (CRITIC) is extended into the interval-valued Fermatean fuzzy (IVFF) setting for ranking the barriers. The IVFF weighted average operator based on the Choquet integral is introduced to form a group decision matrix. Then, the developed ARAS model with the IVFF-CRITIC method is proposed to evaluate the implementation barriers for AI in FSCF, which can depict the interactions between the barriers. Finally, a case of an FSCF, including four participants, is presented to illustrate the application of the reported model and demonstrate its reliability. The result shows that “Data privacy” ( $${c_{10}}$$ ) is the main barrier impeding AI adoption in FSCF, and the participant “small and medium-sized processing enterprises” ( $${a_3}$$ ) has the highest barrier level to AI adoption.

Keywords

• Food supply chain
• Interval-valued Fermatean fuzzy set
• ARAS
• Barrier analysis
• AI