Background: According to the 2023 World Economic Forum report, the impact of Artificial Intelligence (AI) and automation on the job market was more significant than originally projected. Although 2018 research forecasted significant job losses balanced by job creation, current data indicates otherwise. Between 2023 and 2027, it is anticipated that 69 million new jobs will be created due to advancements in AI, however, this will be offset by the loss of 83 million jobs, leading to a net decrease of 14 million jobs worldwide. Roles related to AI, digitalization, and sustainability, such as AI specialists and renewable energy engineers are expected to grow, while those in clerical and administrative sectors are most at risk of decline. This shift underscores the need for reskilling and adapting to evolving fields, as nearly 44% of workers skills will face disruption by 2027. The demand for analytical thinking, technological literacy, and adaptability will grow as companies increasingly adopt frontier technologies. Objectives: (1) identify key variables influencing adaptability of college graduates in Indonesia, (2) quantify the strength of relationships between these variables to understand the combined effect on graduate adaptability. The research also aims to (3) develop theoretical and practical recommendations to strengthen ICIL policy and equip students with the relevant skills needed to thrive in an ever-changing job market. Methodology: The research focuses on predicting future employment trends, adaptability, and learning agility (LA), along with the implications for improving the Independent Campus Independent Learning (ICIL) policy. It focused on the significant unemployment rate among college graduates, along with the lack of research on the relationship between job change predictions, graduates’ adaptability, and the impact on graduates’ general well-being. The mixed-method strategy with quantitative analysis was used to conduct this research with data collected from 284 ICIL participants through online survey. The gathered data was evaluated using Structural Equation Modeling (SEM) with Lisrel version 10. Results: The result showed that job trend projections significantly influence responsiveness, which demonstrated a robust association between employment trend predictions and LA. Responsiveness significantly influenced learning agility which indicated no significant direct association between job trend projections and graduate adaptability. Conclusion: The research emphasized the need to consider adaptability as a concept with multiple dimensions. It proposed incorporating these factors into strategies for education and human resources development in order to better equip graduates for the demands of a constantly changing work market. Unique contribution: This research focused on adaptability as a multifaceted concept that consist of the ability to forecast job trends, be sensitive, and possess LA. It offered a deeper understanding of the relationships between these variables as discussed in the human resources literature. Technology, corporate culture, and training played a critical role in connecting employment trend prediction with the ability to respond effectively. Key recommendation: Institutions should implement a comprehensive approach to the development of human resources, with emphasis on fostering critical thinking, analytical abilities, and the practical application of information. By employing these tactics, higher education institutions may effectively equip graduates with both academic proficiency and the ability to adapt and thrive in quickly changing organizational environments, leading to the production of robust and versatile workers.

Food safety in supply chains remains a critical concern due to the complexity of global distribution networks. This study develops a conceptual framework to evaluate how food safety risks influence supply chain performance through predictive analytics. The framework identifies and minimizes food safety risks before they cause serious problems. The study examines the impact of food safety practices, supply chain transparency, and technological integration on adopting predictive analytics. To illustrate the complex dynamics of food safety and supply chain performance, the study presents supply chain transparency, technological integration, and food safety practices and procedures as independent variables and predictive analytics as a mediator. The results show that supply chain managers' capacity to anticipate and control risks related to food safety can be improved by predictive analytics, leading to safer food production and distribution methods. The research recommends that businesses create scalable cloud-based predictive model solutions, combine data sources, and employ cutting-edge AI and machine learning tools. Companies should also note that strong, data-driven approaches to food safety require cooperative data sharing, regulatory compliance, training initiatives and ongoing improvement.

Heat removal has become an increasingly crucial issue for microelectronic chips due to increasingly high speed and high performance. One solution is to increase the thermal conductivity of the corresponding dielectrics. However, traditional approach to adding solid heat conductive nanoparticles to polymer dielectrics led to a significant weight increase. Here we propose a dielectric polymer filled with heat conductive hollow nanoparticles to mitigate the weight gain. Our mesoscale simulation of heat conduction through this dielectric polymer composite microstructure using the phase-field spectral iterative perturbation method demonstrates the simultaneous achievement of enhanced effective thermal conductivity and the low density. It is shown that additional heat conductivity enhancement can be achieved by wrapping the hollow nanoparticles with graphene layers. The underlying mesoscale mechanism of such a microstructure design and the quantitative effect of interfacial thermal resistance will be discussed. This work is expected to stimulate future efforts to develop light-weight thermal conductive polymer nanocomposites.

This research examines three data mining approaches employing cost management datasets from 391 Thai contractor companies to investigate the predictive modeling of construction project failure with nine parameters. Artificial neural networks, naive bayes, and decision trees with attribute selection are some of the algorithms that were explored. In comparison to artificial neural network’s (91.33%) and naive bays’ (70.01%) accuracy rates, the decision trees with attribute selection demonstrated greater classification efficiency, registering an accuracy of 98.14%. Finally, the nine parameters include: 1) planning according to the current situation; 2) the company’s cost management strategy; 3) control and coordination from employees at different levels of the organization to survive on the basis of various uncertainties; 4) the importance of labor management factors; 5) the general status of the company, which has a significant effect on the project success; 6) the cost of procurement of the field office location; 7) the operational constraints and long-term safe work procedures; 8) the implementation of the construction system system piece by piece, using prefabricated parts; 9) dealing with the COVID-19 crisis, which is crucial for preventing project failure. The results show how advanced data mining approaches can improve cost estimation and prevent project failure, as well as how computational methods can enhance sustainability in the building industry. Although the results are encouraging, they also highlight issues including data asymmetry and the potential for overfitting in the decision tree model, necessitating careful consideration.

This study conducts a comparative analysis of various machine learning and deep learning models for predicting order quantities in supply chain tiers. The models employed include XGBoost, Random Forest, CNN-BiLSTM, Linear Regression, Support Vector Regression (SVR), K-Nearest Neighbors (KNN), Multi-Layer Perceptron (MLP), Recurrent Neural Network (RNN), Bidirectional LSTM (BiLSTM), Bidirectional GRU (BiGRU), Conv1D-BiLSTM, Attention-LSTM, Transformer, and LSTM-CNN hybrid models. Experimental results show that the XGBoost, Random Forest, CNN-BiLSTM, and MLP models exhibit superior predictive performance. In particular, the XGBoost model demonstrates the best results across all performance metrics, attributed to its effective learning of complex data patterns and variable interactions. Although the KNN model also shows perfect predictions with zero error values, this indicates a need for further review of data processing procedures or model validation methods. Conversely, the BiLSTM, BiGRU, and Transformer models exhibit relatively lower performance. Models with moderate performance include Linear Regression, RNN, Conv1D-BiLSTM, Attention-LSTM, and the LSTM-CNN hybrid model, all displaying relatively higher errors and lower coefficients of determination (R²). As a result, tree-based models (XGBoost, Random Forest) and certain deep learning models like CNN-BiLSTM are found to be effective for predicting order quantities in supply chain tiers. In contrast, RNN-based models (BiLSTM, BiGRU) and the Transformer show relatively lower predictive power. Based on these results, we suggest that tree-based models and CNN-based deep learning models should be prioritized when selecting predictive models in practical applications.