Unsri Students Develop Drought Prediction Technology

This innovation uses the Long Short-Term Memory (LSTM) method, a deep learning algorithm known for its reliability in analyzing time series data. LSTM is then combined with the Standardized Precipitation Index (SPI), an index commonly used to measure drought levels. The combination of these two methods is expected to produce more accurate predictions than conventional approaches. This research received funding support through the 2025 PKM-RE program from the Ministry of Education, Culture, Research, and Technology of the Republic of Indonesia (Kemendikbudristek RI).

The research team is led by Annisa Reida Raheima (Computer Engineering), with members consisting of Evan Febrian (Computer Engineering), Nabila Kurnia Aprianti (Computer Engineering), and Dio Subayu Jonathan (Soil Science), and is supervised by Ms. Alvi Syahrini Utami, S.Si., M.Kom.

Ogan Ilir Regency was selected as the research location because this region frequently faces drought problems that directly impact people’s lives, particularly farmers. The unpredictability of the rainy and dry seasons in recent years has made it difficult for many farmers to determine their planting schedules. Additionally, droughts often lead to problems with clean water availability.

“Through this research, we hope to make a tangible contribution to the community, especially farmers who are most affected by drought. With this prediction system, they can be better prepared to face the dry season—for example, by adjusting their planting patterns or preparing water reserves,” explained the research team leader.

According to him, using LSTM was the right choice because this method can identify rainfall patterns that fluctuate from year to year. Meanwhile, SPI is used to classify drought levels—whether mild, moderate, or severe. As a result, the research findings are not merely predictive but can also serve as a basis for decision-making by local governments.

The Student Creativity Program they are running also receives support from the university. Sriwijaya University considers this type of research important because it directly addresses real-world problems in the field.

“Unsri students are not only expected to excel academically but also to produce innovations that benefit the broader community. This research on drought prediction is a concrete example of how science can help in everyday life,” said one of the team members.

In addition to providing short-term benefits, this research is also believed to support national programs related to food security and climate change adaptation. More accurate rainfall and drought predictions will help the government formulate disaster mitigation strategies, especially in drought-prone areas.

This innovation by Unsri students also serves as proof that the younger generation is capable of combining advanced technology with the needs of the community. The hope is that this research can continue to be developed to produce an application that can be directly accessed by farmers and relevant agencies. That way, communities in Ogan Ilir and other regions can experience the tangible benefits of research conducted by the nation’s youth.