In an era of rapid technological advancement, the manufacturing sector is undergoing a transformative shift toward Industry 4.0, embracing cutting-edge technologies to boost productivity and operational efficiency. However, with progress, new challenges comes, the most important is ensuring the safety, security, and air quality in manufacturing environments. This thesis aims to address these issues by designing novel Artificial Intelligence (AI) solutions that use Long Short-Term Memory (LSTM) networks to detect suspicious behaviors based on skeletal data.
The primary goal of this study is to design AI algorithms that improve safety and security by detecting suspicious behaviors based on skeletal data, with LSTM serving as a powerful tool for sequential data analysis. This algorithm is intended to detect behaviors such as excessive sneezing, frequent body scratching, or fainting, all of which may serve as early warning signs of health issues exacerbated by poor air quality.
Concurrently, the thesis delves into air quality measurement, employing a mathematical model based on short path distance analysis. This model, in addition to the behavioral detection aspect, provides a quantifiable framework for monitoring and assessing environmental conditions within manufacturing facilities. The combination of skeletal data analysis and mathematical air quality modeling represents a game-changing step toward increased vigilance, worker well-being, and efficient manufacturing operations.
Finally, the research aims to improve safety, security, and air quality monitoring, fostering a safer and healthier working environment for manufacturing personnel and optimizing manufacturing processes. This endeavor represents a synthesis of AI, quality measurement, and human activities in manufacturing, with the goal of ushering in a new era in which technology and well-being merge for the benefit of the manufacturing sector.