Predicting product information diffusion for sustainable quality management in industry 4.0: an improved Bass model approach

Abstract

Abstract In the era of Industry 4.0, accurately predicting the popularity of product information is vital for optimizing marketing strategies and supporting sustainable business growth. Although the classical Bass model has been widely used in diffusion research, it overlooks two critical aspects of modern digital environments: the heterogeneity in user decision-making and the temporal decay of user interest. To address these limitations, we propose an enhanced Bass model that integrates a two-phase diffusion framework. This model accounts for user conversion across heterogeneous information sources and incorporates an interest decay mechanism by relaxing the assumption of constant influence coefficients. We validate the model using movie trailer data from the Weibo platform. The results show that our model significantly outperforms the classical Bass model as well as its power-function and exponential-function variants in predicting information popularity. Further analysis reveals that both internal and external influences decay over time, with internal diffusion also shaped by social pressure. These findings offer valuable insights for improving data-driven diffusion models and enhancing digital marketing effectiveness. Future research may explore model applicability across different product categories and social media platforms.