TE-FOOD's Livestock Management and Epidemic Control solution enables government authorities and farmers to recognize, contain, and eliminate epidemic animal diseases, by using the latest technologies like blockchain, artificial intelligence, and mobile apps.
The third pillar enables government bodies and academia participants to estimate the economic consequences of the epidemic, and the potential counter actions. The concept was to give an outlook about how the pandemic affects food supply, as well as how the potential interventions affect participants of the food industry on a financial level.
This concept is technically based on the second pillar's ability to analyze the spread of an animal disease, simulate forecasted scenarios, and rate them by using AI driven probability calculations.
The analysis and simulation is based on historical outbreak data, as well as collected information and general parameters, like
- biosecurity rating of the farm
- transmission rate of the disease
- transmission methods of the disease
- historical transmission rate amongst farms and logisitcs providers
- rate of infections on neighbouring farms
- transmission range
- latency period of the disease
- disease recovery rate
- disease death rate
- rate of losing immunity
- length of being infectious
- efficiency ratio of a vaccine (if exists)
- time and ratio of related vaccinations on the farm
- inbound and outbound transport rate of the farm
- common inbound and outbound transport locations of the farm
- rate of infection on the farm
- number of days since the farm is under quarantine
- number of days since the last illness has been reported from the farm
- actual regional quarantine level
- previous cullings on the farm in different time intervals
- ratio of culling efficiency on the local infection rate
- disinfection levels of logistics providers
The third module basically an extension of the analysis module, which adds an economic layer to the calculated data. Various calculated or simulated data can be generated for any geolocation.
Basic calculations like value of livestock lost, and loss of revenue help government officials to see the pandemics' impact on the food supply, the food industry, and enable them to prepare compensation packages.
Simulated scenarios enhanced with the financial layer can primarily forecast potential food supply issues, and enable them to compare the total cost of lost revenues and the potential costs of higher level interventions (e.g. preliminary vaccination, compartmenting, safety culling).
The neural network algorithms calculate the potential loss of livestock on farms in case of intervention or no-intervention.
Logically, as the simulations extrapolate in multiple steps, the certainty of the simulation decreases, but still provides trends to the researchers to estimate the consequences of potential actions. This means that estimating and scoring the medium term consequences of an outbreak not only requires huge technical resources, but generates a growing number of possibilities, thus, the certainty of the estimations decrease. This is why currently, the simulations are limited to provide as much certainty as possible, while providing a potential future outlook.
In practice, the users can set up a potential scenario, where they order safety culling on farms, where the infection rate grows, or has the potential to grow to a specific level, and - using AI based algorithms - estimate the financial consequences of the safety cullings and infections. In another scenario, they can replace the safety culling action with a farm lockdown action, and the system estimates the financial losses in that scenario. Comparing scenarios with different government actions the authorities can be prepared how the pandemic might spread, and what economic consequences they can accept in case they initiate different actions.