A considerable portion of the world frequently experiences flooding during the monsoon season. As a result of this catastrophic event, hundreds of individuals have become homeless. In addition, rescuers are not usually effective enough to rescue the majority of victims. This is due to inadequate rescue operations infrastructure, a severe flaw in today's technologically advanced society. This manuscript proposes a microservice-dependent secure rescue framework that uses geographic information system mapping with a K-Means clustering algorithm to identify flood-prone regions. Numerous microservices, such as fleet management, cloud computing, and data security, integrate and execute the framework in preand post-flood situations. Labeling data from the proposed framework generates a support vector machinebased classifier for predicting flood risk. Furthermore, a hybrid A* algorithm is developed to find an optimal route for the rescue operation. Based on the K-means clustering results, which reduced the variance by 89.2 percent overall, dividing the data into six clusters was the best option for this study. The smoothness of the suggested hybrid algorithm is also used to verify its superiority.

The unwanted vehicular engine irregularities diminish vehicular competence, hinder productivity, waste time, and sluggish personal/national economic growth. Transportation sectors are essential infrastructures that require practical vulnerability assessment to avoid unexpected consequences through risk severity assessment. Artificial intelligence would be vital in the Industry 4.0 era to eliminate these issues for seamless activity and ultimate productivity. This article presents a risk management framework that includes an efficient decision model for monitoring and diagnosing vehicular engine health and condition in real-time using vulnerable components information and advanced techniques. To do this, we used the vulnerability identification frame to identify the vulnerable objects. We created a decision model that used an infrastructure vulnerability assessment model and sensor-actuator data to diagnose and categorise engine conditions as good, minor, moderate, or critical. We used machine learning and deep learning algorithms to assess the effectiveness of the risk management system's decision model. The stacked ensemble of the deep learning algorithm outperformed other standard machine learning and deep learning algorithms in providing 80.3% decision accuracy for the 80% training data and efficiently managing large amounts of data. Anticipating the proposed framework might assist the automotive sector in advancing with cutting-edge facilities that are up to date.