Applications of Machine Learning and Data-Driven Models to Agriculture Risk Management

Each planting season, farmers are exposed to the uncertainty that their crops will yield poor results, leading to unpredictable revenue. The seasonality of their product lends itself to price volatility as a result of unpredictable factors such as weather, pests, supply and demand. In an industry where small price fluctuations can have a big impact on profitability, risk management is extremely important. As farm input costs continue to increase, farm profitability has not increased correspondingly. The objective of our research is to adapt advanced analytical methods to develop novel algorithms that generate cutting-edge price predictions for agricultural commodities. These price predictions will be used to develop hedging strategies and increase farm risk management practices. With better predictions, cheaper hedge strategies, and more timely data, farmers will have increased financial stability and the opportunity to raise profits in an industry where price fluctuations make already slim profit margins unpredictable.

Faculty Supervisor:

Alex Melnitchouck

Student:

Partner:

Algo-Rythmn

Discipline:

Computer science

Sector:

Professional, scientific and technical services

University:

Olds College

Program: