Intelligent Fisheries Information Network - iFIN

Data-driven decision-making IoT app for efficient and sustainable fishing

It is estimated that over 600 million people are supported by the fishing industry, an industry that is directly impacted by climate change and changing environments. For example, fish movement patterns, rising sea levels, temperature and salinity conditions may all impact the ability of fisherman to efficiently fish. Typically, fishermen incur around 40% fuel and effort loss while searching for fish due to a lack of guidance at sea.

Our team seeks to increase the viability and scale of sustainable economic activity from oceans, and enable coastal communities, governments, and corporations to use data to understand and make complex decisions around sustainable and resilient development. We propose a data-driven artificial intelligence (AI) based platform including a mobile application that will inform fishermen (and other stakeholders) about efficient fishing grounds and routing. The platform will reduce both search time and costs endured by fisherman. Additionally, the platform may provide government stakeholders with information about compliance with quota restrictions and provide alerts of any impending changes observed in the natural system.

The platform may be deployed across a wide variety of coastal areas. For example, the platform may be applied where a quota-based management system is in place (e.g., northeast US) and where no rigorous management practices are in place (e.g., coastal regions of India). In the former, we provide the information of where to fish directly to fishermen through an app as a paid subscription. Where a quota is not generally enforced as in the Indian context, we provide the information to local co-managers (e.g., NGOs) through the app, who in turn provide the fishermen with information as a reward for sustainable fishing behavior. The funding for this application is planned through private-public partnerships involving philanthropies, mobile service providers and governments.

On the technical side, our solution utilizes a combination of ocean biogeochemical-physical equations, climatological oceanographic data and historical fish catch information, and incorportates data-driven fish models to make 4-D (3-D in space and 1-D in time) a-priori fish probability maps. Then, real-time data acquired from satellites, ground based radars, nearby metocean platforms (NOAA/Indian buoys, moorings, Argo floats, shipboard, etc.), IoT devices, and live fisher feedback about catch is utilized in an optimal Bayesian approach. The approach will enable the app to obtain updated a-posteriori fish probability maps. From these maps, efficient target fishing grounds are identified and passed on to customers. As populations migrate toward the coastline in some regions and away from the coastline in others, the prosperity of nations, states, and communities will increasingly depend on our collective ability to balance economic imperatives and environmental constraints. To co-exist with nature and for a sustainable tomorrow, it is important to improve the present efficiency of resource management and utilization with the help of advanced technological solutions such as IoT and information gathering, analyzing and interpreting in real-time. Through this solution, we anticipate reduction in green house emissions, and improved viability of fishing as a livelihood under constraints imposed to manage diminishing fisheries resources. 

Traditionally, fishermen have relied on their experience (skills, passion and technique often passed down through generations) to search for fish at sea. This solution is an innovative application of such knowledge in fisheries, recent advances in ocean and data sciences, AI and IoT paradigms to predict fishing grounds and migration patterns. Our predictive models are built on climatological, remote sensing and in-situ data along with fundamental governing equations of ocean physical and biogeochemical processes that affect fish population and track ecosystem changes. Real-time IoT data and feedback from fishermen will allow the novel AI system to adaptively update predictions.

From gathering data to running the model for predictions to relaying actionable information back to the fishermen, technology is the hands and brain of our solution. The core back-end is an AI computational engine which runs a data-driven primitive equation dynamical ocean model, data-driven fish models, and other machine learning algorithms for predicting probable fishing areas. The user facing side of our solution will be web and mobile apps. This front-end will also interface with oceanographic instruments onboard the fishing vessel to send real-time data to our backend computational engine. Such a technology is not available at present.

Over the next 12 months, we intend to pilot the solution both in the US (Massachusetts) and in India (Lakshadweep). In the US, we will partner with Cogitur Inc. to discover early adopters and customers from our beachhead market of Georges Bank fishermen. In India, we will partner with Dakshin (an NGO working with fishing communities), Airtel (a mobile service provider), Tata Trusts (a philanthropic agency) and INCOIS (a governmental agency) to pilot a co-management regime with efficient fishing to increase profits of marginal fishermen. We will also refine the model and app and identify the paying customer in India.

Our vision is to reach fishermen in all the major commercial fishing centers in the eastern US and in India. In the US, we plan to grow as a technology service provider with $60MM in annual revenues. In India, we plan to grow as a social service provider with a vision to improve the quality of life of the 14 million fishing families in the nearly 4000 fishing villages in coastal India. We wish to develop a viable financial model where mobile service providers, NGOs and government together provide a win-win-win sustainable and efficient fisheries management. 

Our solution will be deployed as web and mobile applications to be accessed through handheld devices. In the US, fishermen will have direct access to the fishing ground predictions on the app. We will reach the fishing communities by working directly with fishery associations at one of the largest commercial fishing ports in the USA (New Bedford, MA). In India, fishermen will have access to the predictions through a fisheries manager. We will reach the fishing communities through Dakshin, our NGO partner. Our plan is to educate the users of the financial and sustainability of adopting this technology.

Currently, our solution is in the stage of developing the prototype. 

When deployed in the US, the key sector of the population that will benefit from our solution is the commercial fishermen. In India, our solution will benefit marginal and underserved fishing communities who will now have access to modern technology and fisheries management paradigms that secures their future with more accurate fishing information in real-time.

In the coming months, we plan to serve commercial fishermen in Massachusetts, and underserved fishing communities in India. Within the year, we plan to reach 800 commercial fishing boats in MA. Thereafter, we will expand operations to other states in the next three to five years. In India, our target is marginal and underserved fishing communities who are yet to reap the benefit of advances in information technology such as the IoT app developed here. Initially, we will focus on showing the proof of concept for a small community and then scale to all coastal areas within 3 years.

Our team includes senior science members, Avijit, Pierre, and Pat, who are experts in synoptic ocean prediction with over thirty years of experience and strategic collaborations with Harvard, MIT, and the UMass. Our technology team includes Deepak, a research scientist, and Chinmay and Abhinav, PhD candidates, at MIT. Our business and strategy team members include Mike, an entrepreneur focused on sustainability, and Nakul, who is involved with Fish Forever, McKinsey, and Tata Trusts. Our consulting team includes Amritaa, an MIT and Penn alum with business and intellectual property focus, and Onkita, a Cornell alum with data analytics experience.

We plan to pursue two different revenue models in the US and in India. In the US, we plan to operate for-profit and charge $3000 per fishing season per boat leading to a revenue of $2.4MM per fishing season from 800 fishing boats in Massachusetts alone. In due course, we plan to capture a market with $60MM in annual revenue from all major commercial fishing centers. In India, we plan to operate on a not-for-profit model where we would collect enough revenue to cover operational costs. We envision a private-public-people partnership, where the paying customer is different from the end user. For example, in one model, the mobile service providers would pay us to issue the fisheries advisories in exchange for increased customer base among the coastal populations. Additional funding can come from fisheries and ocean agencies of the Government of India, state fisheries departments and sustainable livelihoods programs of philanthropic organizations. 

We are participating in Solve to gain valuable experience in taking our solution from the lab to the market. Solve can help by promoting our team and solution through media exposure and business connections. 

One key barrier for our solution to succeed is the "trust issue". As the key aspect of our solution is the science based data driven predictive model for the ocean and fisheries, the success of our solution is directly linked to the goodness of our forecast. Also, the competitive nature of fisheries and the problem of unsustainable fishing practices can be barriers for the utility and efficacy of our solution.