Cartography includes two major tasks: map making and map application, which is inextricably linked to artificial intelligence technology. The cartographic expert system experienced the intelligent expression of symbolism. After the spatial optimization decision of behaviorism intelligent expression, cartography faces the combination of deep learning under connectionism to improve the intelligent level of cartography. This paper discusses three problems about the proposition of “deep learning + cartography”. One is the consistency between the deep learning method and the map space problem solving strategy, based on gradient descent, local correlation, feature reduction and non-linear nature that answer the feasibility of the combination of “deep learning + cartography”; the second is to analyze the challenges faced by the combination of cartography from its unique disciplinary characteristics and technical environment, involving the non-standard organization of map data, professional requirements for sample establishment, the integration of geometric and geographical features, as well as the inherent spatial scale of the map; thirdly, the entry points and specific methods for integrating map making and map application into deep learning are discussed respectively.

Micro-mobility has the potential to address first -mile challenges, improving transit accessibility and encouraging public transit usage. However, users’ acceptability of modal integration between various micro-mobility options and public transit remains largely unexplored in the literature. Our study investigates the user behavior for first-mile options, focusing on four alternatives: walking, bicycling, motorcycling, and bus, to access urban mass rapid transit (UMRT) in Hanoi, Vietnam. Based on data collected from 1380 individuals, a Nested Logit Model (NLM) was proposed to analyze the determinants of users’ acceptability under each access mode option as well as evaluate further impacts of shifts in access mode choice on vehicle-kilometer traveled and emissions. The analysis shows that the availability of access modes might increase UMRT use by 47.83%. While this increase further generates additional vehicle-kilometer traveled due to the increase in park-and-ride users, this is offset overall by the large number of motorcycle users shifting to UMRT. Under the most optimistic scenario, modal integration for transit-access trips leads to an average reduction of 17.7% in net vehicle-kilometer traveled or 14.5% in net CO₂ emissions or 10.9% in NO_x from private vehicles. Our findings also imply that the introduction of parking fees for bicycling- or motorcycling-access trips, while impactful, does not significantly change UMRT choice. Therefore, the pricing schemes should be a focus of parking planning surrounding stations. Finally, a number of policy suggestions for parking planning and first-mile vehicles are presented.

The goal of this work was to create and assess machine-learning models for estimating the risk of budget overruns in developed projects. Finding the best model for risk forecasting required evaluating the performance of several models. Using a dataset of 177 projects took into account variables like environmental risks employee skill level safety incidents and project complexity. In our experiments, we analyzed the application of different machine learning models to analyze the risk for the management decision policies of developed organizations. The performance of the chosen model Neural Network (MLP) was improved after applying the tuning process which increased the Test R² from −0.37686 before tuning to 0.195637 after tuning. The Support Vector Machine (SVM), Ridge Regression, Lasso Regression, and Random Forest (Tuned) models did not improve, as seen when Test R² is compared to the experiments. No changes in Test R²’s were observed on GBM and XGBoost, which retained same Test R² across different tuning attempts. Stacking Regressor was used only during the hyperparameter tuning phase and brought a Test R² of 0. 022219.Decision Tree was again the worst model among all throughout the experiments, with no signs of improvement in its Test R²; it was −1.4669 for Decision Tree in all experiments arranged on the basis of Gender. These results indicate that although, models such as the Neural Network (MLP) sees improvements due to hyperparameter tuning, there are minimal improvements for most models. This works does highlight some of the weaknesses in specific types of models, as well as identifies areas where additional work can be expected to deliver incremental benefits to the structured applied process of risk assessment in organizational policies.

Outsourcing logistics operations is a common trend as businesses prioritize core activities. Establishing a sustainable partnership between businesses and logistics service providers requires a systematic approach. This study is needed to develop a more effective and adaptive framework for logistics service provider selection by integrating diverse criteria and decision-making methodologies, ultimately enhancing the precision and sustainability of procurement processes. This study advocate for leveraging industry-based knowledge in procurement, emphasizing the need to define decision-making elements. The research analyzes nearly 300 logistics procurement projects, using a neural network-based methodology to propose a model that aids businesses in identifying optimal criteria for evaluating logistics service providers based on extensive industry knowledge. The goal of this study is to develop and test a practical model that would support businesses in choosing most suitable criteria for selection of logistics service providers based on cumulative market patterns. The results of this study are as follows. It introduces novel elements by gathering and systematizing unique market data using developed data processing methodology. It innovatively classifies decision-making elements, allocating them into distinct groups for use as features in a neural network. The study further contributes by developing and training a predictive model based on a prepared dataset, addressing pre-defined goals, expectations related to green logistics, and specific requirements in the tendering process for selecting logistics service providers. Study is concluded by summarizing suggestions for future research in area of adopting neural networks for selection of logistics service providers.

Technological innovation allows nations to produce sophisticated products more efficiently and at higher quality to increase exports. Countries that aim to produce and export sophisticated products can improve their economic complexity and lead to the country’s economic development. Hence, the study investigates the impact of technological innovation on economic complexity in South Africa. Technological innovation, exports, and manufactured products were used as variables to examine South Africa’s economic complexity index. The study employed the ARDL method to determine the relationship among the variables. The ARDL F-bounds test reflected the long-run cointegration among the selected variables. The study produced long-run positive estimates of technological innovation, exports, and manufactured products on economic complexity, however, manufactured products and exports were insignificant. Granger causality indicated unidirectional causality on economic complexity to manufactured products, exports to technological innovation, and a bi-directional causal effect from exports to economic complexity and technological innovation to economic complexity. The study recommends that South Africa focus on innovation, create more diversified and sophisticated products and processes, and promote more manufacturing firms, particularly Agri-processed products.