Purpose: Social media data from financial websites contain information related to enterprise credit risk. Mining valuable new features in social media data helps to improve prediction performance. This paper proposes a credit risk prediction framework that integrates social media information to improve listed enterprise credit risk prediction in the supply chain.Design/methodology/approach: The prediction framework includes four stages. First, social media information is obtained through web crawler technology. Second, text sentiment in social media information is mined through natural language processing. Third, text sentiment features are constructed. Finally, the new features are integrated with traditional features as input for models for credit risk prediction. This paper takes Chinese pharmaceutical enterprises as an example to test the prediction framework and obtain relevant management enlightenment.Findings: The prediction framework can improve enterprise credit risk prediction performance. The prediction performance of text sentiment features in social media data is better than that of most traditional features. The time-weighted text sentiment feature has the best prediction performance in mining social media information.Practical implications: The prediction framework is helpful for the credit decision-making of credit departments and the policy regulation of regulatory departments and is conducive to the sustainable development of enterprises.Originality/value: The prediction framework can effectively mine social media information and obtain an excellent prediction effect of listed enterprise credit risk in the supply chain.

The customer order decoupling point (CODP) is an important decision objective for decision makers. As customers' customized requirements increase, the multi-CODP problem has become a hot topic in the field of logistics customization service (CS) industry. A multiobjective mixed-integer nonlinear team decision model is proposed for the multi-stage CODP of selecting suppliers and assigning corresponding orders, taking into account logistics service integrators (LSI), logistics service providers (LSP), and customers, including minimization of LSI costs and maximization of synthetic satisfaction of LSPs and customers. Subsequently, the nondominated sorting genetic algorithm (NSGA-II), where the key modules such as crossover operations, chromosome structure and mutation operations are reconstructed, is developed to solve the programming model, and the Pareto-optimal fronts of the two objectives are investigated. In order to demonstrate the validity of NSGA-II as well as the reliability of mathematical model, cases from a specific dataset are generated and solved. Moreover, the classic genetic algorithm (GA) is also applied to solve the two objectives separately. Finally, the solution results of the two approaches are compared. To our knowledge, this paper is the first to integrate multi-CODP for selecting suppliers and assigning corresponding orders problem using NSGA-II solution. In summary, the results of this paper provide decision makers easy access to different supplier selection and order allocation schemes with multi-CODP, which is essential to the diverse customization needs of many challenging real-world logistics issues.