Revolutionizing reverse supply chain decision-making: Deep social media analysis in qualitative comparative analysis

Purpose
The aim of this study is to present an in-depth methodology known as deep qualitative comparative analysis to the domain of reverse logistics, offering a thorough set of instructions for supply chain academics on the effective application of this technique in the context of Industry 5.0.

Design/methodology/approach
The procedure of data gathering concentrated on evaluating the practices of managing the stream of returning products by analyzing social media data extracted using advanced deep learning techniques. To ensure the reliability of the data, business customers were involved in verifying the findings within a supplier-customer context. The analysis of the collected data followed a multi-step and hybrid approach in accordance with the deep qualitative comparative analysis method.

Findings
The research findings shed light on seven effective solutions that drive customer satisfaction to higher levels. These findings also highlight the concept of equifinality, in which different configurations have been identified as enough to result in the same level of customer satisfaction.

Research limitations/implications
This research makes a significant methodological contribution by integrating state-of-the-art machine learning techniques, specifically deep learning, into the QCA method. It also provides a detailed guideline for future supply chain researchers in the field of supply chain management.

Practical implications
The authors also highlight the practical implications by presenting seven innovative combinations of factors that can enhance customer satisfaction. By employing an automated framework based on deep learning analysis of user-generated data, this study enhances the core competency of mass-production businesses and aligns with the principles of Industry 5.0.

Originality/value
This research offers valuable insights to supply chain academics, enhancing their comprehension of the appropriate implementation of deep qualitative comparative analysis and its practical implications. In contrast to previous studies that primarily examined the “net effects” of antecedents, this research acknowledges the intricate nature of the relationships between different factors and their influence on customer satisfaction within the reverse supply chain domain. By harnessing the power of big data and employing deep learning analysis, this study underscores the benefits of automation and improved precision of results, aligning with the principles of Industry 5.0.