Page 37 - AI for Good-Innovate for Impact

AI for Good-Innovate for Impact

For this challenge, this project applies an IoT-based peatland forest management and
monitoring system to use the products of private companies as social practice and proposes
a POC to implement the research results into society. The candidate sites are Kuala Selangor
and Sarawak in Malaysia. Our IoT system collects sensor data in real time and stores it in an 4 - UPM
integrated cloud server where all members can access and analyze the data.
In this POC, we aim to introduce machine learning technology for data evaluation and evaluate
effective data utilization such as CO2 Emission evaluation and ASEAN FDRS.

The novelty of this project is aiming for a platform that accelerates the possibility of horizontal
collaboration centered on data, and the possibility of not only warning systems for local
communities, but also climate change initiatives such as CO2 Emission. This POC will be
evaluated by participating stakeholders such as Selangor Forestry Department (JPNS),
Meteorological Department (METMalaysia) and Global Environment Centre (GEC).

4�3� Use case requirements

ITU-T Supplement Y.71 ITU-T Y.3000 series – Use Cases for Peatland Management Systems

• REQ-01: It is critical that the system must integrate Ground Water Level (GWL) with the
Drought Code (DC) to calculate the Fire Weather Index (FWI), using GWL instead of
traditional temperature and rainfall metrics.
• REQ-02: It is critical that the LoRa-based IoT system must be deployed for continuous
monitoring and management of peatland conditions at Raja Musa Forest Reserve (RMFR)
in Kuala Selangor, Malaysia.
• REQ-03: It is critical that the system must ensure robust data collection capabilities and
the feasibility of the collected data must be verified by comparing it with data from the
Malaysian Meteorological Department (METMalaysia) to establish high correlation and
accuracy.
• REQ-04: It is critical that the machine learning model using neural networks must be
implemented to predict GWL from the IoT system data and use these predictions to
compute Fire Danger Rating System (FDRS) indices such as DC, DMC, and FWI.
• REQ-05: It is critical that the IoT system should be enhanced by incorporating CO2
sensors to measure carbon emissions from peatland soil and from human activities, which
aids in environmental monitoring and management.
• REQ-06: It is critical that the enhanced FWI system must be applied to effectively
predict and reduce the fire risk in tropical peatlands, thereby aiding in the prevention of
transboundary haze in the region.

21