AWS Partner Network (APN) Blog

TCS Digital Farming Generates Near Real-Time Insights on Crop Conditions for Actionable Decision Making

By Dr. Suryakant Sawant, Data Scientist – TCS
By Mahesh Sarode, Delivery Manager – TCS
By Rajoy Jose, Sr. Solution Architect – AWS
By Kiran Killedar, Sr. Partner Solution Architect – AWS

Connect with TCS-1

With climate change and population rise, there is increasing pressure on available land resources. Farming communities are constantly exposed to biotic and abiotic crop stresses like pests, disease, drought, natural calamities, and uneven or excess rainfall. This hinders agricultural activities which in turn impacts crop yield and food security.

The Digital Platform for Next Generation Agriculture (DNA) is built by Tata Consultancy Services (TCS) to address challenges faced by farmers and associated enterprises including agri-input companies. TCS is an AWS Premier Tier Services Partner and Managed Cloud Service Provider (MSP).

The DNA platform, powered by Amazon Web Services (AWS), is based on five pillars⁠—social networks, mobility, analytics, cloud, and Internet of Things (IoT)⁠—to create market- and climate-smart entities. It enables the use of predictive analytics for crop acreage, crop yield, crop health, soil moisture status, weather and pest forecasts, and resource quality assessment to help farmers minimize risk.

In this post, we will provide an overview into the Sky-Earth convergence approach developed on AWS. The platform comprises of big-geospatial, IoT, and climate data analysis framework built on AWS to drive critical agri-business decisions.

Solution Overview

The TCS DNA platform enables customers to process data from earth observation satellites to continuously monitor crop health over large areas of agricultural land. This results in reduced risk due to the detection of crop health issues on time, higher productivity of agricultural operations, and better utilization of scarce resources.

The information from various sources (farmers, satellite, weather) is assimilated in real- or near real-time to generate actionable insights within a cropping season.

The figure below gives an overview of the DNA platform, which can transform the existing farming processes, entire farm ecosystems, and ancillary processes. It’s a system which can help monitor the fields and generates recommendations based on real-time verifiable primary data rather than heuristic thumb rule-based data.


Figure 1 – TCS DNA platform capabilities.

Solution Architecture

Figure 2 shows the platform architecture that includes various AWS components connected to:


Figure 2 – TCS DNA platform architecture.

The key services used in the TCS DNA platform are:

  • Registry of Open Data on AWS: This makes it easy to find datasets made publicly available through AWS services. This DNA solution uses the Sentinel-2 dataset which is made sharable via Amazon Simple Storage Service (Amazon S3) buckets. This enables the platform team to spend more time on data analysis rather than data acquisition.
  • Amazon SageMaker: This provides a platform to train, test, and host the proprietary machine learning (ML) model of the TCS DNA platform. SageMaker’s ability to work with third-party models ensures digital farming initiatives (DFIs) have a flexible ML platform to deploy and operate its models efficiently, and at the same time keep its IP protected at all times.
  • AWS Lambda: This is a serverless compute service that lets the application run workflows without having to provision or manage a server. Lambda is used to integrate applications with the ML endpoints hosted in SageMaker, as well as to create the data processing workflow.

Solution Features

  • Efficient data processing: There was no need to download satellite images with all bands, as the Registry of Open Data on AWS has stored spectral bands separately in an Amazon S3 bucket. This helped to get only the required bands (Band 4 and Band 8, and Quality band) and reduced the 40% cost of data transfer and local storage. The total size of one granule (100 sq. km. area) is approximately 800 MB, and with this approach TCS could filter out unwanted data and only ~ 100 MB of data was transferred.
  • Near real-time analytics: As soon as satellite data was a made available, S3 event notifications ensured the data processing could start immediately, thereby reducing the delay between acquisition and pre-processing.
  • Scaled up processing: On-demand provisioning of high-end computing resources helped process large amount of satellite data to get insights over the large areas; for instance, across an entire watershed or country to understand the vegetation health status.
  • Interoperability and standards: The information generated using the platform is interoperable and stored using standards such as Institute of Electrical and Electronics Engineers Standards (IEEE Standards), Open Geospatial Consortium (OGC) Standards, and more. This enables the data and knowledge discovery for other domains such as social science, climate studies, and policy decisions.
  • Monetization potential: AWS opens up avenues for monetizing the proprietary models and algorithms of the platform via AWS Marketplaces for SageMaker models and algorithms and providing data to third parties via AWS Data Exchange.

Solution Coverage

The platform is capable of spatio-temporal scale-up (process satellite data for large region at higher frequency). Each use case has a unique role to gain insight about the local and regional agriculture. The spatial and temporal coverage varies for the business need, such as seed plot monitoring and identify crop coverage.

Following are the spatio-temporal coverage details rolled out to major agri-input supplier:

  • 329 million hectares of land is covered for monitoring various aspects of crop health and crop condition with 10-15 Terabytes of data processed every 7-10 days.
  • Weather forecasts (short, medium, and long window) are provided with thematic representation across country (e.g. India) for better decision making.
  • Seasonal weather anomaly prediction for the next six months is provided to envision future risks and opportunities.
  • Pest and disease forecasts, as well as online dashboards based on machine learning, predict the risk of pests and disease for agri-input planning and managing the regional supply chain of agri-inputs.
  • Country-scale (pan India) offering of use cases like current vegetation health, relative vegetation health/season progress, root zone soil moisture, crop health provides near real-time monitoring of seasonal progress. This is accomplished through remote sensing techniques by processing earth observation satellite images (Sentinel, MODIS, and Landsat) available through Registry of Open Data on AWS.


Considering the dynamic nature of agricultural systems, satellite-based observations help derive spatial and temporal insights at scale, crop health being one of such insights. In this post, we shared insights on TCS’s Digital Platform for Next Generation Agriculture (DNA) that uses the convergence of Sky-Earth technologies to generate the near real-time insights on crop condition for actionable decision making.

The platform helps in planning the stock of agri-inputs, targeted marketing in the regions with high biotic stress, dissemination of information on sustainable crop specific management practices, and production of good quality seeds. The current platform delivers insights for crop monitoring using microservices architecture and is scalable over spatial and temporal resolution.

The AWS Cloud services provide ease in customer delivery and seamless integration of AI/ML algorithms to generate outcome using integration of big-geospatial data, IoT, and weather factors for near real-time agricultural monitoring.


TCS – AWS Partner Spotlight

Tata Consultancy Services (TCS) is an AWS Premier Tier Consulting Partner and MSP. An IT services, consulting, and business solutions organization, TCS has been partnering with many of the world’s largest businesses in their transformation journeys for the last 50 years.

Contact TCS | Partner Overview