Affected by advances in artificial intelligence and awareness of green environmentalism, consumers’ purchasing behavior of fresh agricultural products has been transformed. New agricultural value chains are being created, such as the new retail model. This study used principal component analysis (PCA) to select essential factors of online consumer behavior and adopted multilevel structural equation modeling (MSEM) to analyze the consumer behavior of Fresh Hema, a new retail format completely reconstructed by Alibaba for offline supermarkets, under different levels of the new retail model. The study adopted an online questionnaire of the consumer behavior of Fresh Hema in 2022. The results show that playfulness and epidemic prevention positively impacted customer satisfaction; convenience benefits and green logistics had the most significant positive impact on customer loyalty.

The aim of this study was mainly to analyze the operating efficiency of Green Hotels during the epidemic, and to explore whether hotels that pay more attention to green energy and saving energy would be more economically efficient. The object of this study is the top 20 Green Hotels voted on the European Union Internet Forum from 2018 to 2022. According to the empirical results, all Green Hotels were affected by COVID-19, and their operating efficiency became worse. This study further used Tobit regression to find out the factors that affect the operating efficiency, and performs regression analysis on the efficiency values of the green energy environment variables. The findings showed that the green energy equipment of EU-certified Green Hotels had a negative effect on the operating efficiency, with a coefficient value of −0.362, indicating that the investment in green energy equipment cannot be achieved in a short period of time. In addition, the education level of managers and hotel green energy licenses have positive effects on the operating efficiency of Green Hotels. In particular, when the Green Hotel licenses are the most important, they have the highest coefficient value of the efficiency of Green Hotels at 0.484. This means that the higher the education level of Green Hotel managers and the higher the level of the Green Hotel license, the higher the operating efficiency. These findings suggest that Green Hotel management should pay attention to its green certificate level and the education knowledge of managers, so as to achieve sustainable energy saving and to improve operating efficiency.

The rise of vertical farms offers new opportunities for climate change and agricultural development. Vertical farms use greenhouses and AI technology applications to overcome the agricultural problems of climatic factors and labour shortages. It also leads to the problem of high electricity bills and other operating costs, which affects the operating efficiency. This study aims to analyse the operational efficiency and input-output resource allocation of vertical farms. This study uses a dynamic network DEA model to analyse the first-stage energy efficiency stage, second-stage operating efficiency and overall efficiency of the world’s top 8 vertical farms from 2018 to 2022. The research results are as follows: 1. The first-stage input (solar power generation and water-saving equipment) is sufficient, and the intermediate output of the vertical farm (the higher the power generation, the greater the water saving) is prone to economies of scale, and the impact on the vertical farm in the first stage will be Brings positive operational efficiencies to the department. 2. The COVID-19 epidemic has had a significant positive impact on the power generation and energy saving efficiency of the first phase of vertical farms, which means that the energy input and output of vertical farms have been better distributed during the epidemic, and indirectly affected the second phase. stage operational efficiency as well as the overall operational efficiency of the vertical farm. 3. According to the slack variable analysis, in order for vertical farms to achieve the best overall operating efficiency, both input and output factors need to be reallocated and adjusted. Among them, operating costs (power costs) need to be reduced by 38.68%, which is the largest adjustment among all input factor resources. Vertical farms use modern, high-tech artificial intelligence to stabilise crop value supply chains and provide many other functional benefits. However, their operational efficiency is severely hampered by high electricity costs. Therefore, effectively reducing the electricity costs of vertical farm operations will be the key to their success or failure.

The application and development of AI intelligent technology is the key to solving the current shortage of manpower in online malls and industries, and also improving the overall efficiency of online shopping Platforms operations. Modern people?s living and consumption habits have changed, and online shopping focuses on delivery speed, accuracy and delivery quality. Therefore, Intelligent warehousing management plays an important role in the operation of online shopping Platforms. This study mainly uses the network DEA model to analyze the operating efficiency of online shopping platforms Integrated with AI-Powered Intelligent Warehouses. The online shopping platforms operation and management is divided into the AI intelligent warehousing stage and the sales department stage, and based on the operating efficiency weights of the two stages, the total efficiency of the online shopping platforms is measured by a weighted average. According to the empirical analysis results, the AI-Powered intelligent warehousing efficiency weights of all online shopping platforms are higher than the sales efficiency weights, indicating that the key to the overall efficiency of online shopping platforms is the AI-Powered intelligent warehousing operation and management efficiency. Although D1 (Shopee Taiwan) has the highest sales efficiency value and annual transaction volume, its efficiency value in the AI-Powered intelligent ware-housing stage is lower than D3, resulting in the total operating efficiency being lower than D3. Therefore, it can be seen that the operating efficiency of the AI-Powered intelligent warehousing department is currently The key to the success or failure of online shopping platforms operations.

Since the traditional Structural Equation Modeling is unable to correctly grasp the preferences and behaviors of consumers at multiple levels, this study combined the virtual reality experience model to analyze the motivations and cognitive behaviors of the public towards sports and fitness. The factor analysis was adopted to measure the latent variables of the virtual reality sports experience function, and then Multilevel Structural Equation Modeling was conducted based on the latent variables to find out the optimal model combination and the best path relationship between people?s experience situations and latent variables, to truly grasp the motivations and goals of people?s sports experience. The empirical results showed that the main factors of the latent variables of people?s satisfaction and loyalty of the group level were different from those of the individual level, which indicated that there was a difference between the group and individual sports preferences in virtual reality sports experience, and it indirectly affected the consumption intention of the group and the individual for virtual reality sports experience.