Due to insufficient integration of spatial constraints, existing methods often fail to achieve global optimisation in balancing multiple objectives such as ecological benefits, landscape accessibility, and construction costs, thereby hindering the scientific nature of design and its implementation efficiency. This paper aims to address the challenge of difficult collaborative optimisation of spatial patterns and component parameters due to the separate application of BIM and GIS in landscape infrastructure design, proposing a BIM-GIS deep coupling optimisation method based on multi-level spatial constraints. This method constructs a unified coding system through spatial semantic matching and 3D geometric alignment, and uses a joint similarity function to achieve high-precision mapping between BIM components and GIS elements. The experiment shows that the average geometric alignment error of this method is 0.08 m, the average convergence iteration is 42 times, and the maximum performance degradation rate under disturbance robustness is only 8%.
The lack of integration between the tourism industry and smart cities affects the tourist experience and the efficiency of urban management. This paper collects data on tourist behaviour and urban operations through IoT sensors and mobile internet data. It uses support vector machines and random forest algorithms to perform demand forecasting and behaviour analysis. Then, convolutional neural networks and long short-term memory networks are used to optimise personalised service recommendations for tourists and to schedule urban resources. The experimental results show that average tourist satisfaction has increased by 8 points, and the average urban management response time has been reduced by 102 seconds. Through this intelligent, collaborative approach, this paper has achieved deep integration between tourism and smart cities, providing a feasible solution for future smart city development.
Although industrial factories can rapidly promote economic development, they have caused significant damage to rural ecosystems. Based on the principle of ecological sustainable development, this paper analyses the characteristics of rural resources and combines IoT sensor technology to develop green industries such as tourism and e-commerce, in order to solve the problems of rural ecological balance and industrial development. This paper compares the situation before and after the implementation of rural green industry planning in Wuyuan County and Haifeng County, respectively. The experimental results show that in Wuyuan County, the average economic growth rate of traditional rural enterprises is 34.08%, while after the implementation of rural green industry planning, this growth rate has increased to 72.04%. In Haifeng County, the average economic growth rate of traditional rural enterprises is 42.04%, but after implementing the rural green industry plan, the growth rate increases to 64.64%.
This paper addresses the critical challenge of integrating artificial intelligence (AI) into green ecological construction for promoting environmental sustainability. It proposes an innovative AI-enhanced slack-based measure data envelopment analysis (SBM-DEA) model to quantitatively evaluate regional ecological efficiency. Focusing on Shenqiu County, a representative agricultural region in central China, the study combines multi-source data - including satellite-derived NDVI, IoT-based PM2.5 monitoring, and socioeconomic inputs - to assess ten townships with diverse economic functions. Results reveal significant efficiency disparities, with only the ecological conservation township achieving full efficiency. Key inefficiency drivers include excessive energy consumption and elevated PM2.5 pollution levels, particularly in industrial port townships. The slack analysis quantifies that these inefficient townships require targeted reductions in PM2.5 of up to 18.5% and in energy use of up to 32.0% to achieve optimal performance. The paper concludes by proposing tailored policy pathways and an AI-driven dynamic governance framework to bridge technical potential with local implementation, offering a scalable model for sustainable ecological management.
Digital transformation (DT), as a key driving force for enterprise innovation and development, has been widely applied in various fields such as agriculture, industry, and services. Although DT has achieved significant results in improving operational efficiency, optimising resource allocation, and reducing costs, its potential negative impact on the ecological environment is gradually becoming apparent, especially in terms of increased energy consumption, intensified resource waste, and rising pollutant emissions. This study investigates the economic and ecological impacts of DT under a sustainable framework, based on surveys and performance data from agricultural, industrial, and service enterprises. Results show improved revenue and reduced costs across sectors, with agricultural firms reporting a 13.87% revenue increase and 5.69% cost reduction in the third quarter, and industrial firms showing an 8.73% revenue increase and 5.89% cost reduction in the fourth quarter. Service firms demonstrated slower yet positive changes. However, 96.13% of employees perceived negative environmental consequences, particularly in air pollution, water shortages, and reduced forest coverage. Findings indicate that while DT enhances economic performance, it imposes environmental pressures. Aligning digital strategies with ecological goals is essential for achieving sustainable enterprise development.
This paper presents a lightweight dynamic feature pyramid network (DFPN) for intelligent landscape feature recognition, tackling computational redundancy and boundary ambiguity in complex scenes. DFPN employs an adaptive scale-aware module to dynamically weight and fuse ResNet-18's multilevel features (C1-C4), while a squeeze-and-excitation (SE) channel attention mechanism amplifies critical feature responses. Deformable convolutions enable adaptive receptive fields, enhancing multi-scale landscape perception and computational efficiency. For sharper boundaries, DFPN integrates atrous spatial pyramid pooling (ASPP) with edge priors and adopts a boundary-aware dice loss to mitigate edge ambiguity. On urban, complex, and rural landscape datasets, DFPN attains 93.82% pixel accuracy and 79.1% IoU at 42.3 FPS. In cross-dataset evaluation, its mIoU drops only 12.3% on the City dataset - substantially better than U-Net's 18.7% - confirming superior generalisation. Through co-designed architecture and loss optimisation, DFPN strikes an effective balance between accuracy and efficiency, proving robust in diverse, complex landscapes.
Addressing reinforcement learning's failure to model high-dimensional dynamic interactions in seedling cultivation, this paper proposes an intelligent regulation strategy integrating species embedding and proximal policy optimisation (PPO) for precise, dynamic environmental management of diverse seedlings. The study constructs a 12-dimensional state space encompassing initial biological traits, real-time environmental parameters, and species encoding, designs a 243-dimensional discrete action space encompassing temperature, humidity, light, water, and CO2, and introduces a multi-objective reward function to co-optimise growth rate, survival rate, resource efficiency, and stress avoidance. The strategy improves the average daily growth rate, with a final survival rate of 95.2%, an average reduction in electricity consumption per unit biomass of approximately 0.4 kWh/g, and an average reduction in stress events of 19.86. Furthermore, a transfer learning mechanism enables fine-tuning for new tree species in just seven days. This study provides a generalisable, efficient, and personalised regulation paradigm for intelligent seedling cultivation.
To address the challenges associated with a low per capita GDP, employment rates, environmental quality indicators, and sustainable development metrics, alternative approaches are required; a new research method for high-quality sustainable development measures for social economy and environment has been proposed. Using principal component analysis to evaluate the level of social-economic and environmental development, and determine the coupling coordination between social-economic and environmental factors. Targeted measures for high-quality sustainable development aspects have been proposed based on the degree of coupling and coordination, including the formulation of scientific and reasonable plans, strengthening technological innovation, improving policy and regulatory systems, and enhancing international cooperation and exchanges. Case analysis results show that the per capita GDP in the study area varies between 108,000 yuan and 117,000 yuan, the maximum employment rate is 84%, the maximum environmental quality index is 0.83, and the sustainable development index varies between 0.68 and 0.91.
Hydroponics, also termed aquaponics, nutri culture, or soilless culture has formed a revolution in the agricultural industry. Aggregating AI with hydroponics aids in tackling global food security. In line with this vision, a hydroponic system was constructed utilising the deep water culture method for the green chilli plant. Hydrobuddy, an open-source application, identified the proper nutrient to prepare the solution. IoT devices were used to collect and monitor the data: EC level, pH level, humidity, and air temperature at regular intervals. To augment our efforts, a machine-learning model to predict the dry weight of the tomatoes with the open-source datasets (OpenAg) was formulated to study the growth of tomato plants. Through this initial study, the optimal EC and pH values for the development of two plants (green chilli and tomato) belonging to the Solanaceae family were similar. Considerable results with better analysis of the fundamental traits that are necessary for the optimal growth of the chilli plant are formalised. By amalgamating hydroponics, AI, sensor technology, and machine learning, the model is poised to revolutionise the field of agriculture and effectively address the global challenges of food security.
