Customer Perceived Value and Social Media Analytics : How Supplier Evaluation Can Benefit from Aspect Based Sentiment Analysis and Fuzzy Inference

Capturing customer value perception is crucial for driving supply chain performance improvements. In fact, the core of supplier management resides in its impact on value generation and on end-customer satisfaction. Social media analytics on customer sentiment enables large scale group decision making (LSGDM) to support supplier value generation. However, studies in the literature do not elaborate on the linkage between the supply management process and customer value drivers, neither how the customer value perception relates to the value delivered by suppliers. State-of-the-art natural language processing approaches, such as aspect-based sentiment analysis (ABSA), have the potential to address this issue. Nevertheless, such a proposition is not found in the literature. Furthermore, the inherent subjectivity of the mined sentiments and uncertainty of managerial judgments should be accounted for. To overcome this, computing with words with the fuzzy set theory has been applied to support supply chain decisional processes. Therefore, the present study proposes an LSGDM model that combines ABSA with fuzzy inference systems (FIS) to support supplier evaluation driven by customer value perception. The central idea of the proposed model is to track customer perceived value quantitatively over time with ABSA, from social media, and use it as an input to a multiple-FIS based supplier evaluation approach. An application considering the case of a large manufacturer of smartphones illustrates the model’s application. The main results show the model capability of capturing customer perceptions and internalizing them as a driver for continuous improvement initiatives in supplier development