Grey linguistic term sets for decision-making

In the era of Big Data, decision-making has become more complex and more uncertain. Faced with this situation, fuzzy linguistic approach may be an information representation model that is closer to natural language and people's cognition habits than exact numerical models. Although Big Data has a large amount of data, the useful information is incomplete, scattered and poor. Thus, an expert may use a more uncertain linguistic expression, i.e., there exists one term, several non-consecutive terms and linguistic intervals at the same time. In view of grey system theory for presenting objective uncertainty, we propose a new concept of grey linguistic term set (GLTS). GLTS has realized a unified description of uncertain linguistic term, hesitant fuzzy linguistic term set, extended hesitant fuzzy linguistic term set, and can provide an effective tool for data generation, processing and fusion of Big Data. Firstly, inspired by generalised grey numbers, we propose the basic representation model, kernel, and degree of greyness for GLTS. Secondly, we study some basic operations for GLTSs, e.g., whitenisation, subset, complement, union, intersection, merge, meet, and two aggregation operators. Finally, we develop a novel multicriteria group decision making method for quality function deployment with grey linguistic information. An illustrative example about electric vehicle manufacturing company is used to demonstrate the application and effectiveness of the developed method.

Automated summary

In the era of Big Data, decision-making has become more complex and uncertain. The fuzzy linguistic approach provides a closer representation to natural language and people's cognitive habits. The concept of grey linguistic term set (GLTS) is proposed to describe uncertain linguistic terms and can effectively support data generation, processing, and fusion in Big Data.