Revolutionizing Vault Security with Smart IoT, Weight Analytics and Machine Learning
The persistent use of physical money, despite the rise of digital payment methods, poses security challenges for vaults storing banknotes and coins. Traditional vault security measures, including physical barriers, time locks, dual control systems, and surveillance, are susceptible to sophisticated attacks and insider threats. This paper introduces a novel approach to enhance vault security by incorporating smart Internet of Things (IoT) devices and machine learning algorithms to monitor the weight of banknotes on vault shelves. By tracking and analysing weight variations, this system aims to detect discrepancies and potential theft. The system employs various machine learning models, including Linear Regression, Lasso Regression, K-Nearest Neighbors (KNN), Support Vector Machines (SVM), and Random Forest, to predict the number of banknotes based on weight and denomination. The evaluation demonstrates that Linear Regression and Lasso Regression achieve the highest accuracy, making them the most effective models for this application. Challenges such as limited data, computational resource constraints, and the need for more refined features are discussed, alongside potential improvements like data augmentation and enhanced interpretability. This approach offers a significant advancement in vault security by integrating modern technology to safeguard physical money against theft and unauthorized access.