AI-Driven Meat Food Drying Time Prediction for Resource Optimization and Production Planning in Smart Manufacturing

The meat food manufacturing industries play a crucial role in delivering various meat products to global consumers. However, one of the significant challenges within this industry is optimizing food processing efficiency across various stages, as it directly affects both product quality and production costs. Drying is one of the crucial stages, wherein moisture is extracted from the meat to reach the desired moisture levels. This prevents spoilage and influences product quality, safety, and overall production efficiency. The drying time is variable, contingent on factors such as the type of meat, quantity, environmental factors, and the desired product characteristics. This variability contributes to the complexity and multifaceted nature of the issue. Conventional approaches for estimating drying times often depend on empirical rules or manual observations, which can be time-consuming, subjective, and susceptible to human error. Therefore, implementing an automation solution by developing a predictive model for drying times in meat manufacturing is essential for optimizing the production lifecycle. Recognizing the potential of advanced computational techniques, machine learning algorithms have demonstrated promising results across various predictive tasks in recent years. Building on this, this research paper aims to explore the utilization of machine learning methods in predicting the drying time of meat-based food products incorporating multiple factors including structure and properties of food, environmental factors, food mass, and physical parameters of food containers. Furthermore, the paper explores correlations, performs feature importance analysis, and addresses the challenges and limitations within this context.