26

@ Forest | May-June , 2021

Even though still at its early stage , agri-tech in Malaysia has shown promising results as embodied in the pilot project managed by the local Farmers ’ Organisation / Pertubuhan Peladang Kawasan ( PPK ) – agricultural cooperative reorganised under the Farmers Organisation Act 1 ( 973 ) – in Kuala Langat .

Individual farmers or private limited companies ( i . e ., the smallholders ) come under the care and oversight of the PPK .

In turn , the PPK is under the jurisdiction of the national Farmers ’ Organisation Authority ( Lembaga Pertubuhan Peladang / LPP ), Ministry of Agriculture & Food Industry ( Mafi ).

The pilot project involves collaboration with the Malaysia Digital Economy Corporation ( Mdec ), specifically in its agtech scheme known as e-Ladang . It aims “ to pilot digital technology use cases in transforming traditional farming into a high-income digital economy profession ” and is divided into :

• e-marketplaces ,

• last-mile delivery ,

• smart farming ,

• transporter ,

• warehousing / storage components . For the Kuala Langat PPK pilot project , the focus has been on smart farming . Smart farming , also known as precision farming , involves using the Internet-of-Things ( IoT ) and automation to control and monitor the farming process from fertilisation and irrigation , combined as fertigation to pesticide spraying and harvesting .

Monitoring in real-time

As reported and analysed in the EMIR Research policy article , “ Time to modernise the agricultural sector for food security ” by Ameen Kamal ( Business Today , April 13 , 2021 ), “ smart sensors embedded in each polybag to detect temperature and humidity , providing information on when and how much water is needed . The sensors … can also measure pH of the medium and fertiliser concentration , specific quantities of nutrients are solubilised using pumps in a mixing tank , which are then pumped through small pipes that drip at each polybag ”.

The data and process are not only monitored in real-time but can also be conducted remotely , i . e ., outside the farm itself . Digitalisation saves on costs ( i . e ., in terms of labour and reducing leakages and wastages ) and increases yields and thereby the income of farmers . According to the Chairman of Mdec , Datuk Wira Dr Rais Hussin , profits have increased by 33 per cent , and revenue has gone up by 20 per cent .

Thus , by a single stroke , digitalisation could both address and even “ permanently resolve ” our self-sufficiency issue as well as enabling farmers to enjoy higher income on a sustainable basis .

Also , food safety concerns such as overuse of pesticides can be allayed as the risks are incredibly minimal throughout the sub-process – from entry into the automated sprayers fitted into horizontal rails above the polybags to the

BY JASON LOH actual spraying itself .

In improving and enhancing the IoT system , the use of low-powered widearea / LPWA IoT such as Narrowband IoT / NB-IoT ( on a bandwidth of 180 kHz and latency of fewer than 10 seconds ) should be considered , as proposed by Ameen Kamal .

NB-IoT would promote a more energy-efficient IoT system covering an extended range ( i . e ., long-distance ) for applications on larger farms and plantations . It can also be used for underground coverage when the farming could also be done in tunnels and caverns somewhere in the future .

Or in the immediate context , should outdoor farming become indoor ( greenhouse ), which would minimise pesticide use and afford protection against weather disruptions – where the control panel or dashboard is on another site .

Minimise power replacement

NB-IoT would also enable batteries with a shelf life of up to 10 years , thus helping to save costs and minimise power replacement .

To supplement and complement NB-IoT , especially in minimising latency lag , it could also be modelled along with edge computing , which would bring the IoT system ( currently on 4G LTE networks ) in line with the future 5G technology .

Edge computing brings data processing and analysis closer to the IoT devices rather than being entirely dependent on a local data infrastructure or centralised cloud servers , which would still be needed as backup .

It can serve as a kind of “ secondary ” or a rather parallel sensor that collects and communicates real-time data .

For example , the interaction between soil conditioner types ( in promoting the use of biofertilisers ) and seed varieties to increase yield and control crop characteristics can be done on the spot and then only sent online for further analysis .

Other examples include functioning as an early warning system for meteorological predictions , enabling real-time / on-site detection of crop experimentation progress , etc . ( see “ Edge computing : A tractable model for smart agriculture ?”, Artificial Intelligence in Agriculture by M . J . O ’ Grady , D . Langton , and G . M . P . O ’ Hare ; and “ Overview of edge computing in the agricultural Internet of Things : Key technologies , applications , challenges , IEEE Access by Xihai Zhang , Zhanyuan Cao and Wenbin Dong ).

Such application of edge computing also promotes machine learning and deep learning for artificial intelligence ( AI ) by incorporating accumulated experience alongside tailor-made and context-specific recognition and parsing .

Framework will pave the way

Thus , the IoT devices can perform cloud , server , and data centre functions – a kind of hyper-converged infrastructure ( HCI ) but on a local and customised scale .

Edge computing especially integrates well with hydroponics , aeroponics , and integrated agriculture-aquaculture ( i . e ., the simultaneous breeding of fish and other aquatic organisms such as crustaceans molluscs alongside a certain