To address the limitations of traditional product and service models and meet consumers’ demands for personalized, intelligent, and seamless experiences, the Smart Product Service Ecosystem (SPSE) has recently garnered increasing attention as a new business model. Ensuring its stable operation is crucial to maintain competitiveness and achieve long-term sustainability. The stability of the SPSE is conceptualized as the outcome of the interconnectedness and co-evolution of multiple factors. Therefore, identifying key factors affecting stability and implementing targeted measures are key to maintaining stable system operation and continuous optimization. In this study, through a comprehensive literature review and analysis of the system’s operating logic, 36 factors affecting the stability of the SPSE are identified based on the dimensions of structure, function, mechanism and benefit. Given the strengths of the ORESTE method and its traditional model’s limitations in fully leveraging decision-making information, this paper proposes a CN-PF-ORESTE model to rank these stability factors. Complex network (CN) captures the correlation among factors and objectively measures the relative importance of factors through the centrality index. Pythagorean fuzzy sets (PF) can deal with the ambiguity of attribute evaluation information. The improved ORESTE model takes into account both the subjective analysis of the expert’s empirical judgment and the objective analysis of the information structure. Finally, the implementation process of the method is illustrated with an example of Smart Home Product Service Ecosystem (SHSE), and the comparative analysis and discussions are conducted to demonstrate its rationality and flexibility.

The proliferation of multimodal data provides a valuable repository of information for financial distress prediction. However, the use of multimodal data faces critical challenges, such as heterogeneity within and among modalities and difficulties in discriminating complementary and redundant information among modalities. To this end, we propose an attentive and regularized deep learning method for predicting financial distress using multimodal data, including financial indicators, current reports, and interfirm networks. Specifically, considering heterogeneity within and among modalities, we design three modality-specific attentions, i.e., ratio-aware, report-aware, and neighbor-aware attentions, for adaptively extracting key information from financial indicators, current reports, and interfirm networks, respectively. Considering difficulties in discriminating complementary and redundant information among modalities, we design a conditional entropy-based regularization to guide the method focusing on complementary information while discarding redundant information during modality fusion. We also propose the use of focal loss to address the class imbalance problem. Empirical evaluation shows that the proposed method significantly outperformed all benchmarked methods in terms of predictive and representation performance. We also provide key findings and implications for stakeholders.