Smart Decision Support for Employee Satisfaction: A Hesitant Fuzzy Linguistic Regression Model Applied to Tech Startups

Real-world decision-making scenarios in modern organizations often involve uncertainty and linguistic ambiguity, particularly when evaluations are expressed in subjective terms rather than precise numerical values. Employee satisfaction assessment is one such domain, as feedback related to workload, salary, and promotions is typically linguistically conveyed. To address this challenge, fuzzy set theory (FST) offers a range of tools–among them, the hesitant fuzzy linguistic term set (HFLTS), which is particularly effective in handling hesitation in linguistic evaluations. This study introduces the Hesitant Fuzzy Linguistic Linear Regression Model (HFLLRM)—an innovative framework designed to integrate hesitant linguistic data into linear regression for enhanced multi-criteria decision-making (MCDM). The proposed model is applied to a real-world scenario involving employee satisfaction evaluation across twelve departments within a rapidly growing tech startup. Subjective assessments from multiple decision-makers on key satisfaction factors are captured using hesitant fuzzy linguistic elements (HFLEs), and model parameters are estimated using linear programming methods (LPM). To validate the model’s effectiveness, results are benchmarked against the established TOPSIS method, while rank correlation analysis is used to test the statistical consistency of outcomes. Findings demonstrate that HFLLRM not only provides more nuanced and reliable rankings but also enhances the practical applicability of decision-making models in complex, linguistically rich environments. This underscores the HFLLRM’s potential as a powerful decision-support tool for improving employee satisfaction analysis in dynamic, tech-driven organizations.