Döngüsel tedarik zincirleri: Toplam gıda endüstrisinde bozunmuş ürün tespiti için bir IoT uygulaması

Today, majority of food created is wasted rather than consumed, which has a negative impact on worldwide hunger and the economy. Improvements to aggregate supply chains are at the forefront of the actions needed to meet these problems. One of such improvements noted in this research was aggregate food storage. ESP8266 Microcontroller, along with DHT11 temperature and humidity sensor and MQ3 alcohol sensor, is used to measure the storage conditions of fruit products. Data gathered is sent to AWS-Cloud-Computing-Service via MQTT protocol and is stored in databases using rules defined. As result, the sensor data in the database is examined using AWS IoT Analysis. As temperature rises, crops begin to decompose. Accordingly, a rule in cloud is defined to fire with out-of-range measurements, and AWS Simple Notification Service is triggered to send ambient readings to user via SMS and e-mail. Convolutional Neural Network model was developed to classify fruits based on their variety and freshness using binary and multiclass classification. Models was first taught using images of 1693 fresh apples, 1581 fresh bananas, 1466 fresh oranges, 2342 rotten apples, 2224 rotten bananas, and 1595 rotten oranges and tested with images of 395 fresh apples, 381 fresh bananas, 381 fresh oranges, and 388 rotten apples, 601 rotten bananas, and 530 rotten oranges over 50 epochs. Binary model had 99.15% training and 97.59% testing accuracy whereas multi class model had training accuracy of 99.59% and testing accuracy of 98.99%.