In recent years, humanitarian logistics have received much attention from practitioners and researchers due to the significant damage from natural disasters on a global scale. This case study investigated the potential of leveraging social media data to enhance the effectiveness of humanitarian logistics in Vietnam after the disaster caused by Typhoon Yagi. The research examined public sentiment about the disaster response efforts, pinpointed the needs of critical relief, and assessed the performance of various machine learning models in classifying disaster-related content on social media. Data was sourced from multiple platforms, preprocessed and then categorized according to the damage types, required relief supplies, and sentiment labels. After that, different machine learning models were utilized to analyze the negative impact of the disaster. The analysis revealed that housing and transportation were the primary sources of negative public sentiment, indicating significant unmet needs in these areas. In contrast, generally more positive responses were received in relation to cash assistance, food, and medical support. A comparative evaluation of 12 machine learning models suggested that conventional algorithms, such as Random Forest, Support Vector Machine, and Logistic Regression, outperformed deep learning models in sentiment classification tasks. These findings shed light on the value of social media as a real-time indicator of public perception and logistical effectiveness. Therefore, incorporating sentiment analysis into the planning of disaster response can support more adaptive, timely, and community-informed decision-making for governments and humanitarian organizations.

Industrial symbiosis (IS) represents a strategic framework for collaboration among companies through innovative partnerships, which aimed at optimizing resource utilization, reducing environmental impact, and promoting sustainable development in line with the principles of circular economy. This study conducted a systematic literature review (SLR) and a quantitative analysis of the effectiveness of IS tools in resource management. Publications from January 2020 to December 2024 were retrieved from the established databases such as SpringerLink, ScienceDirect, EBSCO, and DOAJ, with a focus on industrial engineering, environmental management, circular economy, sustainable development, resource conservation, and recycling. Advanced methodologies including the Fuzzy Analytic Hierarchy Process (FAHP) and the Decision-Making Trial and Evaluation Laboratory (DEMATEL) were applied to evaluate four key dimensions, i.e., Decision-Making (DMD), Geographical Location (GLD), Strategic Planning (SD), and Lean Manufacturing (LMD), along with 21 subcriteria. The results indicated that DMD and GLD functioned as causal dimensions influencing SD and LMD, while alternatives such as Intelligent Waste Recycling Systems (IWRS) and Life Cycle Assessment (LCA) were considered to be highly efficient in resource utilization. The identification of dominant relationships via the threshold value of α = 0.58 highlighted strategic leverage points for implementing sustainable manufacturing practices. These findings emphasize that effective DMD, combined with strategic planning based on geographical considerations and application of technological tools, is critical for optimizing resources, enhancing environmental protection, and fostering economic and social development, thus providing clear guidance for the implementation of IS strategies in industrial settings.