Efficient management of the perishable food supply chain, particularly in the dairy industry, plays a key role in reducing costs, increasing delivery speed, and controlling emissions. In this study, an integrated optimization model based on the Internet of Things (IoT) and dynamic routing is proposed for the distribution network of dairy products in smart cities. This model leverages real-time data from IoT sensors to continuously update vehicle routes in response to changing traffic and demand conditions. To evaluate the performance of the proposed model, nine practical scenarios were simulated, including baseline conditions, heavy traffic, peak demand, sensitive products, and time window constraints. The results indicated that the time window scenario achieved the best performance, exhibiting the lowest cost, minimal emissions, and the smallest objective function value, whereas the peak demand scenario resulted in the highest cost and emissions. Subsequently, a sensitivity analysis was conducted on four key parameters: product perishability rate, freshness degradation penalty, transportation cost, and emission coefficient. The proposed model can serve as a foundation for developing smarter systems using real-world data and machine learning techniques. Furthermore, integrating such an IoT-based dynamic routing approach with predictive analytics can significantly enhance supply chain resilience against unforeseen disruptions such as sudden traffic congestion or equipment failure.