While the country has witnessed tremendous growth in digital infrastructure to support farmers, the innovations have primarily focused on pre-harvest and harvesting phases (advisory on sowing and crop cultivation; knowledge on fertilisers, seeds, and machinery; weather forecasts), or tailored to larger farmers thereby leaving out marginalised farmers and traders. Meanwhile, postharvest losses and commercialisation of crops – a significant step in ensuring farmers’ income is commensurate with the increase in their farm output – have received only limited attention under the app-driven agricultural revolution.

The Basel Agency for Sustainable Energy (BASE) and The Swiss Federal Laboratories for Materials Science and Technology (Empa) began working towards addressing this gap in January 2021 under the Your Virtual Cold Chain Assistant (Your VCCA) initiative as a part of the data.org global innovation challenge. After a year and a half of interdisciplinary teamwork including strong local partnership development, data collection, user interface design, the development of machine learning and physics-based models as well as of the software back-end and front-end, and on-ground user testing, the team is excited to release the first version of its open-access, data-science-based mobile application called Coldtivate. This version of the app is being tested by our local cooling companies and operators. Updated versions of Coldtivate will be released in late 2022, enabling farmers to access the existing and additional features on their smartphones.The BASE and Empa teams are grateful for the excellent collaboration with Relief Applications for the software development. 

“We have developed an open-source app that enables cold room operators to manage inventories and monitor occupancy digitally, and cooling beneficiaries to track the quality of their crops in storage in real-time. By combining this with the upcoming feature of staying updated on market prices, this information empowers people to make informed decisions on when and where to sell their crops, strengthens their bargaining position to negotiate for better prices, and prevents distress selling.”

Roberta Evangelista, Technical Lead

The app facilitates the implementation of the Cooling as a Service (CaaS) business model, wherein local cold room providers retain ownership and maintenance responsibilities of the infrastructure while farmers use the facilities on a pay-per-crate basis for a nominal fee. The app is a plug-and-play tool for local cold room providers to offer the CaaS model quickly and efficiently. To make the process of tracking the crates entering and leaving the room less cumbersome, Coldtivate is a first-of-its-kind app to digitalise inventory management in the cold room, which is presently carried out manually and is prone to error. Using this feature, cold room operators can check-in and check out the cooling users’ (marginal farmers and small-time traders) produce. Digitalising the inventory management system allows operators to monitor the current and predicted occupancy of the room, notify users of their pickup days and keep track of the prices received by the farmers for their produce stored in the cold rooms in a hassle-free manner. Cold room providers who own and operate a fleet of cold rooms can assign operators to specific cold rooms and keep an overview of their operations.

Even as cold storage capacity is progressively increasing in the country, backed by policy interventions, inadequate trust in the potential of cooling to augment income has hindered farmers’ uptake of these solutions. Coldtivate attempts to bridge this trust deficit with another of its key features, enabling farmers to track the shelf-life of their stored produce in real-time. For each checked-in crate, the app computes the ideal time for farmers to pick up their crops, i.e., it predicts the number of days that the fruit can last under the current storage conditions with a buffer time for bringing the crops to the market at ambient temperature. To make these calculations, the app integrates a crop-specific physics-based model known as ‘digital twins’ developed by Empa, which simulates how crops in a particular crate age inside the cold room in real-time. These calculations depend on factors such as the type of commodity, room temperature and humidity, and the initial quality of the produce. The relevant data is gathered through different means. For instance, the temperature and humidity are recorded by the sensors installed in the room. The initial quality of the produce is determined during the check-in of the produce by noting the harvesting dates.

Since the app’s current version caters to cold room operators, they can monitor the shelf-life of each farmer’s produce and inform them of the ideal check out date as per the app’s predictions via SMS or in-person. In another four months, a new version will be released allowing farmers to see this information for themselves and decide when to check out their crates.

The journey of improving farmers’ welfare through postharvest management goes beyond merely curbing quantitative (decrease in weight and volume) and qualitative (cosmetic imperfections and fall in nutritional value) losses in crops and increasing their shelf lives. Surrounded by uncertainty, farmers make sales decisions in a dynamic environment against many competing short-term tactical and long-term strategic decisions. Although vast amounts of data are available to optimise agriculture sales decisions, the information is typically raw and scattered, offering few useful insights. The next steps for Coldtivate involve empowering farmers with market price forecasting.

The upcoming versions of the app will also feature daily market price forecasts for different commodities using machine learning and open-source data from markets near the cold rooms across India. This information, combined with updates on the remaining shelf-life of the crops, help farmers with better price discovery by choosing the ideal time and location of sales. Processed data represented in actionable formats fixes information asymmetry in the market, mitigating exposure to manipulation by intermediaries and ensuring that the farmers secure a relatively larger portion of the profits.

“It has been a long and exciting journey: from software design, development and testing, to data acquisition, model development and their integration into the app. We’re thrilled to finally make this a reality and bring the power of data science to farmers in an open-source solution. We leverage sensor’s temperature data with our physics-based digital twin models to predict the remaining shelf life of crops in cold storage, and we’re also in the final stages of the development of a machine learning model that leverages public market data to make 2 weeks daily market price forecasts at the market, district and state levels. We believe these tools combined in Coldtivate create a unique added value and have the potential to disrupt the agri-business sector.”

Joaquin Gajardo, Technical Co-Lead

In addition to using data-science to assist farmers in navigating an increasingly complex value chain, Coldtivate includes a Knowledge Hub that contains information about the optimal storage temperature and average storage time for each commodity and will be regularly expanded and enriched to reflect the best practices in postharvest management as per the latest research. To universalise access to this crucial information and create engaging learning environments, the knowledge hub will also be available offline in interesting formats such as comic books and manuals in the coming months. After the addition of further functionalities, Coldtivate will also be tested with cold room operators in Nigeria by September 2022, where the Your VCCA project is deployed in partnership with Coldhubs. 

Coldtivate helps cold room operators and providers to offer sustainable cooling technology under a servitisation business model, and helps farmers to better leverage the benefits of cooling, reducing food loss and securing higher prices.

Coldtivate is now available for download for free on Android and iOS.

1. Li D., Gajardo J., Volpi M., Defraeye T. (2022), Using machine learning to generate an open-access cropland map from satellite images time series in the Indian Himalayan Region. engrXiv. preprint DOI.

As a part of the open-access map of India on Google Earth Engine, a new layer was developed and documented that identifies cropland from satellite images in India at a 10-meter resolution using machine learning. 

2. Defraeye T., Shoji K., Schudel S., Onwude D., Shrivastava C. (2022), Evaporative coolers for postharvest storage: where to best use them and how well do they work? engrXiv. preprint DOI.

3. Defraeye T., Schudel S., Shrivastava C., Motmans T., Umani K., Crenna E., Shoji K., Onwude D. (2022), The charcoal cooling blanket: A scalable, simple, self-supporting evaporative cooling device for preserving fresh foods. engrXiv. preprint DOI.

4. Evaporative Cooling Potential Map: https://empasimbiosys.github.io/evapo_cooling_map/

A new type of passive evaporative cooler was developed for smallholder farmers that helps them store their fresh produce after harvest. In warm areas, often, a lot of quality is lost immediately after harvest. However, in remote areas, energy to cool produce is not always available. This cooler only needs water to work and is easy to produce using just textile and charcoal. DIY instructions with a video are online now. In addition, the team developed and documented open-access maps to identify how good evaporative coolers work in the world. That way, these evaporative coolers are only used where and when it makes sense to use them.

5. You L., Schudel S., Defraeye T. (2022), Developing of biophysical food for monitoring postharvest supply chains for avocado and potato and deploying of biophysical apple. engrXiv. preprint DOI.

Finally, the team developed a newer version of their artificial food sensor device that helps cooling service providers to monitor better the cooling process in their micro-scale solar-powered cooling facilities. The artificial fruits and vegetables can be packed deep inside the cargo and help detect problematic locations that cool slower than others. Next to apple, banana, mango, and citrus, now also artificial avocado and potato sensing devices were developed.

This gap arises because different datasets needed to translate numbers into action often fail to interact with one another. Furthermore, the data can be challenging to visualise and understand. Fresh food supply chains are complex systems comprising diverse individuals, organisations, resources, activities and technologies, and the previous problems arise within a variety of actors looking to leverage data-driven decision making. Even when focusing on a smaller section of the supply chain, i.e., cold storage, a range of factors such as access to roads and the electrical grid, proximity to mandis (big markets), number of households engaged in farming have to be factored for successful decision-making by supply chain stakeholders (farmers, cold storage providers, retailers, and policymakers). Using widely available open-source agricultural data from different domains, the BASE and Empa teams have developed a multi-layer, interactive map of India (view here) to help players along the fresh food supply chain with decision-making.

“We have created an accessible tool to dynamically compare multiple locations across India. Our goal is to empower cold chain providers to make the best decisions for their business and the farmers they are serving.”

Roberta Evangelista, Sustainability Data Science and Digitalisation Specialist at BASE

India’s hinterlands are home to nearly two-thirds of the population, of which 70 percent rely on agriculture for their livelihoods. A large majority of these farmers operate on less than two hectares of land, making it crucial for them to maximise the sale of their produce for sustenance. However, the reality is far from the ideal – nationally, farmers lose out on revenue from the post-harvest spoilage of about 20-30 percent of their produce. The Your Virtual Cold Chain Assistant (Your VCCA) project being implemented by BASE and Empa aims to reduce food loss by expanding access to pay-per-use cooling services as well as by developing an app that digitises inventory management and provides pre- and post-harvest market intelligence for proper handling, storage, and protection of the crops. 

ORGANISING OPEN-SOURCE DATA IN A SINGLE MAP

There have been several prior initiatives in India to combat the lack of data in rural areas in India. However, many available data are not easy to access or visualise, and may require a certain degree of processing before they can be turned into interpretable information for decision-making. One such notable example is Mission Antyodaya, launched in 2017 by the Indian Government to collect data at the village level on, inter alia, agriculture, land improvement, infrastructure development, rural electrification, and growth in non-conventional energy sources. Although digitally recorded, this open-source data is scattered and needs to be processed into geospatial format and analysed to maximise its utility. The interactive web map developed by BASE and Empa as a Google Earth Engine application offers one such synthesis by collating and visualising different raster and vector data layers related to India’s fresh produce supply chain. Relevant spatially distributed data on cropland mask, elevation, predicted electricity network lines, crop production of fruits and vegetables, temperature, solar radiance, roads, availability of markets, households engaged in farming, distribution of farmer producers’ organisations, and mobile band coverage have been visualised at a district level (or block-level, for some states) through this one-of-a-kind map.

WHAT PURPOSE DOES THE MAP SERVE?

This visual representation of the selected data layers serves as a decision-making tool for stakeholders form the public and private sector in the fresh produce supply chain, including social enterprises like farmer producer organisations and policymakers, as well as financial institutions, cooling providers, NGOs, and government bodies.

“Our mission is aligned with the targets of the SDG 2, and is to improve the income of small-scale farmers around the world with the benefits and opportunities that cooling brings. We believe that the open-source tool we have developed is a key step in that direction, allowing cooling service providers to unlock geospatial insights when looking to scale their cooling rooms offering.”

Joaquin Gajardo, Data Scientist and Software Developer at Empa’s SimBioSys group.

A practical example of this is identifying where cooling could provide the highest shelf-life gains and where to site future cold storage facilities. Cold storage rooms can add value to farming products, thereby witnessing an increasing demand nationwide. However, these cold rooms can adequately benefit farmers only when they are conveniently placed and meet all criteria to protect the crops. The deliberations of where to place a cold room depend on several factors, which may vary by state. Identifying promising cooling unit locations demands expert guidance and technical know-how, which due to the infancy and nicheness of the cooling industry in India, may be difficult to find. The map simplifies this process for cold room operators by computing ideal locations within a 20 km market radius, at most 500 m away from the road, with stable network coverage, and other state-wise considerations. This is the first step towards an application where users are able to compute promising cold room locations by dynamically deciding which layers to include and their thresholds.

Furthermore, cold storage can increase the number of days farmers take to secure profitable sales for their produce instead of liquidating their crops at low rates before it spoils. The map represents the shelf-life gain for selected crops (potatoes, apples, and bananas) across India. These calculations offer farmers and cold storage operators critical information on the added benefit of using (respectively, offering) cold storage with an actionable metric: the number of shelf-life days gained by storing fruits and vegetables in a cold room vs at ambient temperature.

“Simulating how fresh produce loses its quality is our expertise. We are very excited that this knowledge in this web map is bridging the gap between science and practice.”

Kanaha Shoji, Environment Scientist at Empa’s SimBioSys group

The learnings from this map will be complemented by an easy-to-use Your VCCA app to help cold room operators digitally check-in and check-out crates brought in by the farmers, indicate the number of remaining shelf-life days, and provide market intelligence. Stay tuned for the release of the app’s first version in April 2022.

The project is funded by the data.org Inclusive Growth and Recovery Challenge, which was launched in partnership with the Rockefeller Foundation and the Mastercard Centre for Inclusive Growth.

Read the full paper ‘Increasing accessibility and usability of open-source data through a web map for better decision-making in India’s cold chain of fresh produce’ here.

FACTSHEET LAUNCH

Over the past six months, Your VCCA has seen continued and significant progress on multiple fronts. Since its inception, the project team has grown by six members, forged partnerships with various cooling solution providers and other local organisations and collected open-source data for India. Using the data, the team has developed digital food twins for several fruits to map their ageing patterns in food supply chains. Furthermore, a preliminary user interface for the mobile application, making it intuitive and easy to use, has also been designed. 

Today, the BASE and Empa teams are excited to introduce Your VCCA’s visual identity, striking a balance between the two worlds of agriculture and digitisation that intersect within the app. As a part of the launch, the team is excited to release a factsheet that summarises the context, objective, and methodology of the project.  

PROJECT INSIGHTS

In terms of pilot projects and implementing partners, BASE and Empa have started formal partnerships with two local implementing partners, Coolcrop and Oorja. 

The project team is working closely with Coolcrop towards starting a pilot in Himachal Pradesh’s booming apple industry. The partnership seeks to adapt the post-harvest storage and handling, one of the most power-hungry parts of the agricultural value chain, to energy efficiency. Targeting the Farmer Producing Organisations (FPO) as the beneficiary, Coolcrop will own, operate, and maintain the cold room, charging the users a small fee per box of apples stores. It will simultaneously test the effectiveness of some key components of the Your VCCA app, including an image-based machine learning model to grade fruits after harvest, and gauge their receptivity among farmers. Current solutions that grade fruit based on smartphone images are less tailored to the needs of smallholder farmers. To translate the data into actionable metrics that allow farmers to identify the best time to sell their produce, the team is developing visual, physics-based food twins that predict the expected shelf-life of the stored fruit and age along with it in real-time. Coolcrop and the Your VCCA team are further collaborating to gather and label images of stored commodities to better train and improve the accuracy of the machine learning model. Alongside these developments, an in-person survey is being conducted together in the apple and potato supply chains in Himachal Pradesh, Maharashtra, and Karnataka, respectively.

Oorja is using the Cooling as a Service (CaaS) business model in Muzaffarpur, Bihar, where they own, operate, and maintain a cold room. This cold room was supplied by Ecozen and provides cooling for horticulture crops in a market visited by approximately 500 smallholder farmers. This project serves as a pilot on which Your VCCA project team is currently collecting hygrothermal and usage data, intending to upcycle this data and provide insights on optimal storage conditions and expected shelf-life of crops. Together with Oorja, the project team will test an interactive map that will facilitate the identification of promising cold room storage locations throughout India.

TAPPING INTO THE POWER OF DATA

The team has collected planned datasets on weather, agricultural food production and harvesting, market locations and pricing, farmer census data, and satellite images via open-source repositories. These multi-disciplinary datasets have been depicted in a multi-layer map of India, assisting cooling service providers with better scoping sites to install cooling facilities. policymakers gain insights on different aspects related to food cold storage. This tool can aid decision-making in finding sites with the most considerable potential to save food by cooling, thereby increasing the smallholder farmer’s revenue. 

The team has already built digital twins of different fruits, with which the remaining shelf-life of these foods can be predicted based on the sensor data measured in the cold store or throughout the complete supply chain. This is essential for farmers to recognise the ideal time window to sell their products before they spoil. These food twins will be made operational via a web-based platform to farmers and cooling service providers worldwide or via an executable version that can run on mobile devices. In Your VCCA, these food twins and their shelf-life predictions will be combined with market price forecasts to advise farmers on the best time and market to sell.       

A preliminary architecture of the mobile application user interface, from both the farmer and the cooling service provider’s perspectives, has been designed, and its user-friendliness will be tested for improvement through surveys during the coming months.

Your VCCA has received early support from a variety of organisations. Earlier this year, the Global Innovation Lab for Climate Finance has selected Your VCCA with CaaS as the financial instrument they will help replicate in India. They are supporting the development of the pricing model and facilitating interviews with local experts and leaders. In addition, Columbia University has recently published a working paper, “ICT in Agriculture Value Chain, especially during post-harvest operations in India”, by Abhishek Beriya, featuring Your VCAA.

This promising progress in India in only six months has motivated BASE and Empa to start planning the project’s expansion to other countries, with the team already exploring prospects in Nigeria. It is also planning to set up an Incubator, which would enable the project team to train more companies to use the Assistant and implement CaaS.

ADDITIONAL RESOURCES

For more information on Your VCCA – feel free to download the Factsheet of the project. You can reach us at info@yourvcca.org