OPTIMIZATION OF OYSTER MUSHROOM CULTIVATION MONITORING SYSTEM THROUGH THE IMPLEMENTATION OF DIGITAL TWIN CONCEPT

• Main Author: Nur Lathifah, Syfa
• Format: Theses
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/76658
• Institution: Institut Teknologi Bandung

The agricultural industry has excellent potential to grow because of the vast opportunities and demands. However, despite the significant potential in agriculture, especially mushroom cultivation, there are still some problems, such as a gap between yields and market demand, because mushroom cultivators cannot produce optimal mushroom production in quality and quantity. Proper and effective cultivation processes must be carried out to produce quality oyster mushroom products in large quantities. Implementing the digital twin concept by utilizing Internet of Things and Machine Learning technology will have many good impacts to make it easier for mushroom cultivators to monitor oyster mushroom cultivation. The digital twin concept will focus on characteristics so it can represent physical objects as closely as possible. IoT will play a massive role in providing data in real-time through sensors so that things that cannot be seen physically can be known based on their data and machine learning will support the virtual modeling process so that monitoring data for mushroom growth will be available in real time according to their respective characteristics. Implementing the digital twin concept in the mushroom cultivation monitoring system requires five stages: creation, communication, aggregation, analysis, and insight. Based on the research conducted, the digital twin concept can optimize the monitoring system of oyster mushroom cultivation using hydrometer sensors for temperature and humidity data collection, as well as camera sensors for monitoring the physical growth of mushrooms. In the computational process, the ARIMA algorithm is employed to predict temperature and humidity data with a highly accurate level of precision. Additionally, the YOLO algorithm is utilized for object detection to identify the physical conditions of oyster mushrooms at each growth phase. All the analyzed data results are presented in a dashboard format, allowing mushroom cultivators to access real-time environmental and physical mushroom conditions, along with receiving recommendations for care based on various environmental and physical mushroom characteristics.