If we map the current challenges in agriculture maket, there are many. For example In Thailand, we have Roughly two-fifths land is covered by mountains and hills, the steepness of which generally precludes cultivation. classified roughly 58 percent of mountainous and hilly regions as cultivable of which about 19 to 25% was usable for paddy, 28 percent for upland crops, and 11 percent for both paddy and upland agriculture.
Thailand banks on a 5G future and with Govt and Agriculture organizations , FMCG Organization support, our agricultural system is improving but need attention too. For example, our existing agricultural model lacks knowledge of right species to grow, diseases, epidemiology and commodities as per market needs .
Farm-tech Companies need to deep dive the profile of farmers . Our typical household had about three hectares and staple produce being rubber coconut fruit and rice.
Other point is Soils throughout most of the country are of low fertility, largely as a result of leaching by heavy rainfall and flooding. Thai farmers traditionally still relied on rain and flood water for crops , A smart irrigation and innovative farm irrigation solutions like Water from Air Condensation is a viable solution to explore as we have more then 70% humid climate.
Also Since Thailand isn’t a large country, shifting farmers towards an industrial farming system can prove difficult. Social enterprise is the new model that should be more suitable to Thai farmers. It allows for farmers to become shareholders of a new business method called ‘Smart Farming or Service Farming’. This will enable farmers to operate their own business start-up. We need humble Integrated agri-input management – smart micro-irrigation.
Beside, Farming-as-a-service (FaaS) deliver key parameters for small farmers namely Crop health management: yield forecasting, pest management, smart insurance; Integrated agri-input management: agri-nutrients, irrigation, fintech linkages.
Now Lets have a at a glance view of Smart Farm Infrastructure. Its sure that the Connected infrastructure and IIOT has provided not only a way to better measure and control growth factors, like irrigation and fertilizer, on a farm, it will change how we view agriculture in its entirely. We observe that there are 6-7 key building blocks of smart farm namely 1.IoT & crowd-sourcing Soil moisture and piezometer sensors and mobile app data integrations with ML models Sensors used for soil, water, light, humidity, temperature management. First is Remote sensing , Agritech Software that target specific farm types or Applications agnostic IoT Solutions We need Connectivity like cellular , LoRa etc . We need Location sensing like GPS, Satellite, etc. We need Smart Agriculture Machis./Drone/ Robotics automation through Autonomous tractors, processing facilities, etc. like Drone watershed planning Higher resolution of drone survey digital elevation models will yield accurate results.
Above all being Remote sensing data clubbed with deep tech AI to mentor field like crop sown area and crop water stress. The solution includes Data analytics , AI ML and data pipelines for downstream. AI based weather forecasts Assemble forecast with micro-level grids to ensure data granularity for true-to-ground data. Armed with such tools, farmers are expected to take benefit like monitor field and make strategic decisions for the whole farm or for a single plant/crop cycle. Smart farming follows a cycle like this one:
1. Observation :Sensors record observational data from the crops, livestock, soil, or atmosphere.
2. Diagnostics : The sensor values are fed to a cloud-hosted IoT platform with predefined decision rules and models—also called “business logic”—that ascertain the condition of the examined object and identify any deficiencies or needs.
3. Decisions :After issues are revealed, the user, and/or machine learning-driven components of the IoT platform determine whether location-specific treatment is necessary and if so, which.
4. Action :After end-user evaluation and action, the cycle repeats from the beginning.
Lets talk about Thailand’s agricultural system which is transforming toward the digital world. There are mainly 9 changes in agriculture that are anticipated to arise in the Thai agricultural system:
1. Satellite for Agriculture which can identify minerals within the soil, on land and whether if the area can be utilized. Adopting satellite technology will contribute to better management and cultivation of land.
2. Zoning and Geo Strategy will be another factor to bringing about change in agriculture. Zoning or land management means identifying cultivation areas in different parts of the country; by farming in accordance to market trends. For example, coconuts are a popular product for the Chinese market and will need increased farm land accordingly.
3.Water is the heart of agriculture. lack of water is critical issue. Water management and IoT can help provide solutions to this problem. Smart Irrigation can help calculate the amount of water in the area and help manage water issues.
4. A successful agriculture reqires high quality seeds and soil. Unlike a new ‘Mega Farming’ method that requires mass cultivation areas, There is a strong Role of Data and AI in Agriculture Harvesting Smart Seeds and Machine learning for smarter seed selection
5.traceability of the agriculture produce is a necessity in the 4.0 era. Bringing a viable Blockchain platform will allow for food products to be traced back to where they originated, provide transparency throughout the production process with all its stakeholders, and is in accordance with international standards to ensure safe and high-quality products that is environmentally-friendly and does not violate human rights.
6.We need to test food quality ingredients, with scientific research to uphill specific requirements like build immunity , or antiaging etc.
7.For which Agro and Food r&d is must 8.World economic forum has already anticipated the measures to be taken for effective and sustainable use of land as well as precition indoor urban agriculture formats.
All these touch points we observe as change in food and agriculture, happening due to many reason from thrust of food security vision to nutrition thrust.
First one being Predictive Agriculture: technologies like Camera Technology, Indoor Rails inside green house. Vision and Data Capture, Machine Learning are driving value on software side on vision side for agriculture is increasing acceptable for Boosting productivity and innovation, Managing environmental challenges, 3) Cost savings and business opportunities, 4) Better supply chain management ETC
Second is Research driven next-gen proteins tipped for the mainstream like today’s plant-based proteins driving more then 10.8 billion by 2022, required is next-generation plant-based proteins best-positioned to displace first-generation alternative proteins like pea protien. This selection have to build on a detailed analysis of available technologies using a set of parameters, including functionality, nutrition, scalability, and sustainability
Third is consumer sentiments which is changing a lot , for example Generation Z, the generation after millennials currently in school or university, and their preferences. Today’s consumers are willing to pay a premium for better products and more information about those products. Gen Z-ers are committed to sustainable, organic, and locally sourced foods. In-field sensors, autonomous machines, and management software will help farmers keep track of everything that’s happening on the farm and produce higher yields in the same amount of acreage.
Forth is Sustainable packaging is important because it reduces the ecological footprint
Fifth is process of GMO and genetic inheritance allows farmers and breeders to improve food security by increasing both yields and the nutritive qualities of crop varieties and livestock breeds which need strengthened .
Last but not the list the Novel Farming techniques need a data driven approach A prime example is indoor and vertical farming, which uses LED lighting, sensors, and automation to effectively stack indoor on top of another. We need to see how we make it more sustainable and include maximum crops to indoors.
Use of Robotics in Agriculture
Developing autonomous vehicle technology in other sectors will translate to farm implements. These Driver/Driverless farm equipment will improve to provide more flexibility and efficiency for farmers and save on labor costs. Trucks for crops transportation can reap IoT sensor benefits as well. These sensors can track cargo temperature and send alerts if it becomes too warm or cold (i.e., cold chain). Small mobile sensors such as asset trackers will likely continue to use high-latency technologies like LPWAN. 5G will enable autonomous vehicles with more powerful onboard computers to send and receive larger, ultralow-latency data streams, including video. Beside automate several labour intensive aspects of farming, including weeding, seeding, and harvesting. They help in Agrochemicals use in a targeted manner. The first generation of agricultural robots broadly covers three main tasks: eliminating weeds, monitoring pests and diseases and harvesting specialised crops, e.g., berries or vegetables.
Beside all good points, ground realities, we observe local conditions unpredictable terrain , weather conditions can be harsh, and signal connectivity, High Costs of technology , low reliability, Due to this agriculture has historically been slow to digitise and Compared with other industries, uptake of robotics and other digital technologies has been fairly sluggish in agriculture. Overall, my observation is that the field of agricultural robotics is comparatively immature. However, falling costs of robotics combined with ongoing labour issues are making robotics increasingly attractive to the sector, But The labour issues facing agriculture do not necessarily mean that the transition to robotics and automation will be rapid.
Uncertainty stemming from the COVID-19 pandemic coupled with lower profits may prevent farmers from being able to meet the significant upfront capital costs required to invest in automation and robotics. The success of the agricultural robotics industry will be dependent on more than just technology availability. It will also depend on cost in relation to traditional labour, integration with current farm practices, and business models, e.g., renting and subscription based models vs. farmers owning the robots outright.
One area with Farming Global; Agri Robotic acceptability is high is on Robotics milking which is the most commercially developed region of agricultural robotics around 30% adoption in Europe itself. This is mostly because there are significantly fewer technical challenges to developing a milking robot – milking robots operate in controlled, indoor environments and don’t require autonomous mobility, meaning they require much less sophisticated sensor systems and AI than field robots. before accelerating .
Since Smart agriculture applications are starting to move to the cloud, with the aim of delivering benefits in data access, synchronization, and storage. Applications such as VRT, livestock monitoring, mobile payment, trading and farmers’ helpline provide huge advantages/benefits for telecom operators. We see a huge addressable market size by application as catered by telecom providers where we use, LTE 5F emBB, NB1, NB2, M1, Category 5 G Technologies. Beside for remote areas, Telecom’s Services and Collaboration as LPWA Providers in order to provide low-cost, ubiquitous wide-area connectivity that can cover the remotest places where operators fail to reach.
For agriculture fields, Drones are good part of the solution, Today More thai farmers are using drones to monitor their crops. Drones are less costly than driving tractors and robotics through tarrin fields and provide more targeted information about crop damage and other variables. As a high-bandwidth technology, 5G will enable drones to collect higher-quality video data and convey it faster. This high-speed data transmission capability will enable AI drone technology development and real-time reports. Also it it has largely Eliminated Expensive piloted aircrafts . As well as remote sensing, Traditional scouting requires spending hours and hours of visualizing. Potentials and benefits of UAV Drones for Precision Agriculture, farming, and crop management. UAV Drones provide a lot of live data from a range of sensors (i.e., multispectral, NIR, Lidar, etc) from in-depth analysis of crop health, inventory management, etc. as well as solve time and cost problems with Satellite imaging and piloted aircrafts.
Drone are helping in many aspects like Soil and field analysis and Planting. üDrones can be instrumental at the start of the crop cycle. üThey produce precise 3-D maps for early soil analysis, useful in planning seed planting patterns. üAfter planting, drone-driven soil analysis provides data for irrigation and nitrogen-level management.
Planting: Drone-planting systems that achieve an uptake rate of 75 percent Decrease planting costs by 85 percent. These systems shoot pods with seeds and plant nutrients into the soil, providing the plant all the nutrients necessary to sustain life.
We know that apart from farm dashboarding, weuse AI Ml in many Agriculture function, from Right Seed –> Right Area. Smart Irrigation, Predictive Analysis. … Diagnosing Soil Defects. …Production Forecasting using weather condition: … Weed Detection: … Water Treatment. … Recommender System as well as Analyzing market demand, forecasting prices, and determining the optimal time for sowing and harvesting are key challenges farmers can solve with AI.
Some example I mention here which we realize in Thailand like Pheno-typing for Farm Scouting & Forecasting (Health, Quality & Yield), using satteline imagay. Drone recognition of weeds AND spray selectively in real time . But Farm fata point capture and Prediction is area which is evolving question. Like Could an ML-powered device or service (drone, app) enable much higher yield ?
Analytics example, Disease notice in early phase, Reduce crop failure with early detection of infection, Reduction of water and fertilization, Suggest appropriate amount of water, fertilization for crop cultivation, Optimization of cultivating automation-Automate the facility based on crop sensing analytics.
Lets see the overall framework at a glance .
We need Agriculture Automation , Digital Enablement Tool and Solution Landscape is looking as on today around the farmer requirements in 3 major categories – • 1.Agruculure Digitalization and Production which encompassing much of the IoT, robotics, and automation, and remote sensing activity; 2.Planning and Farm Management: intersecting across digital agronomy, resource management, and business planning and execution; and 3.Market Access and Financing: tools and technologies used by farmers, farm managers, and crop buyers to access markets and financing.
But we face issue of limited data availability and quality are tempering long-awaited mass-market adoption. Though there are few Big incumbent players building data platforms that have the scale to deal with these challenges. We’ve even seen moves from IT leaders like Microsoft with its FarmBeats platform and Google with it’s interest in “computational agriculture.” jumped. But where is the gap ?
We Yet to see business models mature which are still in their infancy, and the appetite is cautious for funding substantial infrastructure investment with an unclear timeline to adoption or profitability.
Second market has yet to commit to common services framework ; this reinforces the industry’s fragmentation and hampers scalability. We have not yet seen a sizable wave. Facing an uncertain future with commodity farmer’s scaling back investment in new tech and capital markets tightening up.
Also we feel that the automation and robotics space is crowded, particularly in high-value (high labor) specialty crops with few clear leaders or exits, beyond some are experimenting with new “Robotics-as-a-Service” models to reduce adoption friction. But the incumbents haven’t overlooked this trend in the traditional Agri value chain as digitally transformed business models enable “XaaS.”
Planning and Farm Management is One of the more challenging dimensions of the Agriculture Digitalization sector has always been the ability to link agronomic activity to farm financials. For a variety of reasons, such as lack of digital standardization, interoperability, data quality, and behavioral friction, understanding the true cost of production and modeling scenarios to improve agronomic decision making has long vexed tech companies and farmers both. Farmer subsector need to integrate of digital agronomy, finance, and some lightweight Farm AI Driven ERP functionality. This space will continue to attract attention from the incumbents and institutional finance with some already notable acquisitions and partnerships
Market Access and Financing is The biggest challenge of unorganized farming sector. Working capital is the lifeblood of farms and the driver of financing too. Still Today farmers, look back to government schemes for their seed to market price parity. In developing countries we see that Most of time, a average farmer does not get adequate price of his yield.
We need availability of data to better understand market insights, risk, and supply and demand as well as utilize smart contracts (blockchain-enabled or otherwise). We see many companies already created a wave focused on helping farmers up their game when it comes to the business of farming. But I feel they need more focus on end to end integration for risk assessment and forecasting system to bring weightage to finance decisions.
Lets also touch livestock automation.
Livestock, Management like shrimp, we need to understand the business, like for feeding, predictability in Fish Appetite Index is important parameter. Another aspect is Yield KPI. For example 10 tons/rai of size < 50 Shrimps. Other KPIs like Higher Survival Rate , Higher Average Daily Growth , Lower Feed Conversion Ratio shall be the parameers of the system. Shrimp gut color changes with major disease they carry which need to be monitorried.
On sensors, We have increased feasibility for 5G low-power and denser sensor networks, animal monitorring sensors will likely stay connected via Wi-Fi, Bluetooth or LTE LPWAN. And in large centralized farms where 5G infrastructure can be built over a small area (e.g., a chicken farm) and track individual animals. Ag-tech developers have created herd management sensors, including smart collars and ear tags, to track an animal’s location and health.
When it comes to Pig / Swine, Disease Symptom Diagnostics and Predictability Smart Sensor Requirements becomes critical like … capture instance if pig eat or not , capture show low rate of interest in feed /water, Skin Proximity Sensor to identify Rapid Breathing or Non Invasive Sensers to identify if indication of a fever, Visual Sensor to identify white skin-colored pigs the skin may become reddish., Droopy ears or ears pointing downwards., Dull Eyes, Dull Skin may be sign of diarrhea which may sometimes be bloody or blood stained. Similarly disease detection and prediction model need to be evolved by leveraging AI and ML for all common diseases like Roundworm, Tapewarm. Mites, Lice, Myiasis,
To realize the promise of data analytics, machine learning and all the sophisticated decision support technologies, Livestock Tech companies must first address the data capture and data cleaning hurdles. The research shows cows can physiologically tell us what and how they are feeling, but we just don’t holistically collect the data today to provide insights to producers to accurately replace a visual inspection. Elements of the “connected cow” are materially further along than other agriculture commodities, and even than most “connected humans”, but there is still much work to do. To make the incremental leaps to more data-driven, digital dairies, some basic building blocks are often still missing for producers:
Financial data is not adequately connected to production activity data. •There is reluctance to give up trusted notebooks or there is inconsistent entering of field or barn-level data into digital devices. You can’t improve what you don’t measure. •Inadequate network connectivity or lack of applications that cache content offline until connectivity is re-established is an issue. •Lack of open data standards compound the difficulty in adopting new technologies.
Lets talk about Precision Agriculture. And see what current and future hold for Smart Farming. We observe that as technology improves, farming is gradually moving away from the in-discriminate constant rate approach that historically dominated agriculture, where the entire field receives the same level of inputs.
Instead, many farms, particularly in the West, use a variable rate technology (VRT) approach, where a field is divided into a series of patches, where each patch is treated differently according to what farm data analytics dictates. Means we can detech plant by plant.
This trend may eventually lead to an ultra-precision approach to agriculture, where each individual plant receives the exact amount of inputs it requires.
We know Smallholder farmers, critical to the global food supply, face growing challenges. Approximately 500 million small and family farms contribute 80% of the global food supply . Despite our global reliance on them, small farmers face many hurdles including increasing climate variability, inability to rely on generational knowledge, and pressures on farmland from urbanization. Furthermore, to feed a population that is expected to exceed nine billion by 2050, food production will need to increase by 60% from business-as-usual growth rates.
Issues – Cost and access driving innovation,£ Ecosystem: Ecosystems of partners , Horizontally-oriented: In the absence of an ecosystem, technology developers tend to vertically orient and take on a variety of roles including device development, farmer recruitment, data collection and analytics, and advice to farmers. , # Sector support and funding: A common challenge is finding committed project partners and funding to support the activities over a duration long enough to see results. We recommend that an IoT for smallholder agriculture forum be established, funders supporting projects for at least five years to build sufficient evidence on the efficacy of the intervention, and hands-on IoT training opportunities be supported to prepare students for careers in this sector.
Challenge: How do farmers’ planting decisions respond to crop prices? How, therefore, do we expect demand-side and policy changes to affect agricultural output and input use?
We need solutions which Apply a system dynamics predictive model to local markets, focusing on the behavioral dynamics of a transition from conventional to organic farming which can help farmers can minimize the losses from this transition.
Challenge: Can we to improve the practice of probiotic design for developing world aquaculture in the context of shrimp farming by taking advantage of modern genomics tools?
We know that Aquaculture is the fastest growing mode of food production in the world, growing at a rate of approximately 10% per year.. There is a groundwork required to develop better probiotics for shrimp farming through the use of genomics and machine learning (ML) techniques. We need AgriTech Provider and their Researchers to collaborate with scientists and shrimp producers to develop improved probiotics that can help shrimp populations more effectively resist pathogens, reducing waste and decreasing the need for the overuse of antibiotics that negatively affect local ecosystems. The team will
develop a culture collection and assess the impact of synthetic microbiomes on farm yield.
Download Slides from here