ABSTRACT
Penaeus vannamei is one of the most cultured species. The global production of Penaeus (Litopenaeus) vannamei reached 5.8 million tonnes in 2020, contributing to 51.7% of total shrimp production. However, despite its high production, there are still many issues in this industry. One of those is the disease. The disease brings many threats to shrimp farming, such as slowing shrimp growth rate and even mortality. Previous research estimated that global production losses due to disease over the preceding 15 years amounted to approximately US$15 billion. To help the farmers in mitigating the impact of disease we tried to develop a predictive model that is able to give early warning of disease occurrence. We focused on predicting acute hepatopancreatic necrosis disease (AHPND), infectious myonecrosis virus (IMNV), and white spot disease (WS). The research used the Conditional Tabular Generative Adversarial Model (CTGAN) to synthesize the data to improve the data quality and address class imbalance issues. The synthetic data is then used as input for model development. The model algorithm consists of several engineering processes and classifications. We used Random Forest Classifier (RF) as the classifier. Applying the algorithm to 1839 cultivation data that came from 389 farms we managed to achieve F1 scores higher than 0.85 for the three diseases. However, there is a performance issue in IMNV prediction where we only get a 0.78 recall score which indicates a high false negative prediction. But despite the issue, in this research, we get a hint that the disease occurrence can be predicted based on water quality conditions.