The accuracy of credit risk evaluation is crucial for the profitability of any financial institution. The factorization machine is a widely available model that can effectively be utilized for classification or regression through appropriate feature transformation. In this article, we apply the factorization machine model to the field of credit risk assessment. Since some features of the credit risk assessment data are not numerical, one-hot encoding is used, resulting in sparse training data. However, the computational complexity of the factorization machine is polynomial. To illustrate the effectiveness of the factorization machine credit risk assessment model and compare its performance with other classification approaches such as logical regression, support vector machine, k-nearest neighbors, and artificial neural network, we conduct numerical experiments on four real-world credit risk evaluation datasets. The experimental results demonstrate that the proposed factorization machine credit risk assessment model achieves higher accuracy compared to other machine-learning models on real-world datasets and is computationally more efficient. Therefore, the factorization machine model can be considered as a suitable candidate for credit risk assessment.

Proper supplier selection to meet production demand is a major aspect of all manufacturing and process industries. Green supplier selection has been one of the most critical factors for environmental protection on account of increasing consumption levels and for sustainable development as well. This paper aims at developing an applicable methodology for green supplier selection for the process industry. In this study, both economic and environmental criteria are considered and a comprehensive weighted grey incidence decision approach for green supplier evaluation and selection in a process industry is proposed. First, an overall green supplier selection index system for process industries is considered; then a weighted grey incidence decision‐making model with improved grey incidence coefficients and weighted degree of grey incidence is provided. Improved grey incidence coefficients are defined using transformation sequences of the initial data. To eliminate the ill effects from the use of equal weights, the maximum entropy method is used to determine the weights of the improved grey incidence coefficients. An application example is proposed with the data collected for the chemical processing industry, which provides acceptable results in determining the better supplier. In the end appendix, some theory regarding the weights for grey incidence coefficients is proposed. The empirical results indicate that the model is of great practical value for green supplier selection in the process industry.

As the digital economy experiences swift advancements, demand prediction in logistics holds a crucial significance for firms operating in the logistics sector. The primary aim of this research paper is to ascertain the optimal method for forecasting logistics demand based on the logistics demand data from the Chengdu-Chongqing Dual-City Economic Circle (CC-DEC). The importance and widespread application of machine learning technologies in intelligent forecasting are undeniable. Specifically in Logistics Demand Prediction, Support Vector Regression plays a pivotal role in enhancing accuracy. Precise prediction of logistics demand is essential for optimizing resource allocation efficiency, which is the core focus of this research endeavor. In this study, we conduct the Fuzzy Support Vector Regression Machine approach based on Adam optimization (FSVR-AD). Then we have developed a comprehensive Logistics Demand Prediction index system tailored for the CC-DEC in China, particularly focusing on the dimensions of carbon neutrality and carbon peaking. Three distinct forecasting models are constructed on historical data spanning from 2005 to 2021, aiming to accurately predict the logistics demand within the economic circle. Our analysis reveals that all three models exhibit high predictive accuracy. However, the FSVR-AD demonstrates superior performance, as its predictions align more closely with actual values, resulting in reduced error margins. Given its accuracy and precision, the FSVR-AD is an ideal choice for constructing logistic demand forecasts. Its predictions offer a reliable reference for strategic planning in logistics management, enabling companies to optimize automation and innovate supply chain processes to align with evolving trends.

This paper took the China Pavilion of the International Horticultural Exhibition 2019 Beijing China as the subject for study. Based on the influence of cold climate on people’s behaviors and their perception of the environment, the paper proposed designs of green public buildings adapting to the cold climate and analyzed the overall functions of green buildings via tests of temperature, humidity, illumination and total energy consumption to verify the ecological features and energy conservancy of the buildings, forming a closed-loop chain of procedures of the green design.

This project is based on the climate characteristics of Guangxi, combined with regional traditional construction methods and vernacular construction materials, and mainly in the light of the passive green building technology system guided by the principle of less manpower and more natures features, which saves early investment and the cost of operation and maintenance, further verifies the effectiveness of the strategy after completion, providing a reference for the development and application of related passive building technology.