SmartHangar

The decision support system (DSS) for preserving aeronautical heritage, called SmartHangar, is engineered using a multiple-input and multiple-output (MIMO) decision tree model, operationalized through the scikit-learn library in Python. The input-output combinations for learning based on decision tree approach were defined by experts or by state-of-the-art knowledge. Data in SmartHangar undergo a rigorous preprocessing stage to ensure consistency and accuracy, applying moving average filters. After that a resampling is employed to standardize the temporal resolution across disparate data sources so that datasets are aligned to a uniform date-time format. The SmartHangar system incorporates a blend of robust and cutting-edge technologies within the scikit-learn library, namely PostgreSQL, PyCaret, and FastAPI. First, PostgreSQL, the database management system, is used for managing the complex and relational data structures that the DSS requires. PyCaret automates data-driven corrosion model training and evaluation for prediction of corrosion risk. At last, FastAPI is employed for building API to handle asynchronous requests efficiently for real-time data processing and responsiveness. The SmartHangar system gives off outcomes with actions to undertake for preserving the aeronautical heritage. These actions can be both invasive (refurbishment) and non-invasive (preventive). To facilitate the usage of the SmartHangar the Human-Machine Interface (HMI) is developed in the JavaScript framework Vue.js.