Publication

Abstract : The rapid expansion of online meal delivery services has posed considerable issues in preserving the freshness, quality, and safety of perishable food items throughout transportation. Conventional logistics models are deficient in real-time monitoring and predictive functionalities, frequently leading to food waste, quality deterioration, and consumer discontent. This study presents an IoT-driven intelligent delivery service model that incorporates sensor nodes, edge processing, cloud analytics, and mobile applications to ensure comprehensive visibility of food conditions. To anticipate the danger of spoilage and make delivery operations better, real-time environmental and logistical data such temperature, humidity, gas concentration, and GPS location were analyzed using machine learning and hybrid ensemble methods.   The testing showed that the hybrid ensemble model was 98.2% accurate. This was better than the SVM (97.5%) and KNN (88%) classifiers on their own.   The comparison analysis demonstrated that the proposed method diminished spoiling by 15–20%, enhanced freshness retention by 20–25%, and facilitated adherence to food safety regulations more effectively than prior methods.   The results suggest that IoT-enabled predictive frameworks can revolutionize how perishable food is distributed by making it more open, cutting down on waste, and helping people trust the system.