The UN Food and Agriculture Organization predicts that we need to boost worldwide food production by 70 percent over the next several decades in order to feed the anticipated population of 2050. On top of that crop wastage is a significant problem for most developing countries which contributes towards majority of cash crops in the world. A study once represented that a quantity of wheat equivalent to the entire production of Australia goes to waste each year in India. INR 44,000 crore is the value of food grains wasted in India every year. Under such circumstances the need for an automated agribusiness model is the need of the hour to optimize farm outputs and ensure their availability in the market
What it does
The Solution involves automating agro business to ensure an end to end digitization of the monitoring, procurement and distribution of agro products. It comprises of multiple IOT devices collecting and feeding data on several climate, soil and crop conditions to enable the tracking and monitoring of the farm products by a cultivation manager in charge of the application. Depending on the condition of the products, the application then executes smart routing to multiple expert teams who provide specialized care and handling of the crops.
How I built it
We collaborated with the IoT team to integrate Appian with the sensor devices placed across the cultivation areas. The sensors collect real-time data on climate conditions and soil moisture levels to monitor crop conditions. We Leveraged Appian's Goggle Cloud Vision connected system template to perform image analysis for pest detection in the crops. We used Goggle Maps component plug-in & Connected System template to search and show locations for crop batch creation. We worked towards figuring scenarios where Appian dynamic case management, decisions and rules, AI could be used. Appian's analytics framework has been used to build reports to provide actionable insights for efficient tracking of crop cultivation stages.
Challenges I ran into
The biggest challenge was to collect real-time sensor device data and crop images from across multiple real fields/location to test and realize the true functioning of the app. We faced challenges in preparing the required amount of training datasets for training ML model. Building the auto-refresh functionality of Appian screens to reflect real-time updates of Sensor data was also challenging.
Accomplishments that I'm proud of
Integrating with the IoT sensor devices to capture real-time climate conditions to determine Crop conditions, making use of Google Maps component-plug-in and connected system plug-in to bring power of Google place search and map view into Appian to enable convenient location look up, integration with cloud vision to perform image analysis, creating intuitive and beautiful user interfaces using Appian SAIL.
What I learned
If every farm in the country becomes a smart farm with automation capability, reaching a 70 percent increase in food production and minimizing the manual dependency is a certainty.
What's next for E-Agro Digital Solution
Building a more end to end monitoring component by including drone surveillance, Precision seeding equipments to planting seeds at the correct depth, and spacing plants at the appropriate distance apart to allow for optimal growth, Installation of weather stations for preemptive forecasting of climatic conditions and response, Integration with the Track and Trace solution for complete end to end supply chain visibility and Machine learning/AI capabilities to analyze past actions by the specialized teams and take automotive actions in cases of crop wastage.