Harnessing Structural Equation Modelling Based Synthetic Data with Artificial Intelligence on Employee Performance Prediction Model in Multinational Organisations

Abstract

As organizations evolve, their employee performance capabilities must also advance to maintain competitive advantage. Employees are integral to organizational success, with their performance significantly influencing overall business outcomes. The rapid advancement of information technology, particularly Artificial Intelligence (AI), has permeated various organizational departments, including Human Resource Management (HRM). Despite extensive research in AI, there remains a deficiency in investigating the practical applicability and reliability of AI/ML tools in business contexts. This paper introduces a novel method termed Harnessing Structural Equation Modeling Based on Synthetic Data with Artificial Intelligence on Employee Performance Prediction Model (HSEMSD-AIEPPM). The HSEMSD-AIEPPM model offers a cutting-edge approach for forecasting employee performance in multinational corporations utilizing AI methodologies. Initially, the model applies z-score normalization to preprocess data, enhancing input quality. Subsequently, Structural Equation Modeling (SEM) is employed to generate synthetic data, ensuring a comprehensive dataset that encapsulates varied employee performance scenarios. For the predictive analysis, the model harnesses the long short-term memory (LSTM) technique. Ultimately, LSTM parameters are optimized using the Improved Pelican Optimization Algorithm (IPOA) to enhance model efficacy. To ascertain the model's improved performance, extensive simulations are conducted, and results are evaluated across various metrics. Comparative analysis demonstrates the superiority of the HSEMSD-AIEPPM method over existing techniques.