Big Data and AI Revolution in Precision Agriculture: Survey and Challenges

Foodborne pathogens are a major contributor to foodborne illness worldwide. The adaptation of a more quantitative risk-based approach, with metrics such as Food safety Objectives (FSO) and Performance Objectives (PO) necessitates quantitative inputs from all stages of the food value chain. The potential exists for utilization of big data, generated through digital transformational technologies, as inputs to a dynamic risk management concept for food safety microbiology. The industrial revolution in Internet of Things (IoT) will leverage data inputs from precision agriculture, connected factories/logistics, precision healthcare, and precision food safety, to improve the dynamism of microbial risk management. Furthermore, interconnectivity of public health databases, social media, and e-commerce tools as well as technologies such as blockchain will enhance traceability for retrospective and real-time management of foodborne cases. Despite the enormous potential of data volume and velocity, some challenges remain, including data ownership, interoperability, and accessibility. This paper gives insight to the prospective use of big data for dynamic risk management from a microbiological safety perspective in the context of the International Commission on Microbiological Specifications for Foods (ICMSF) conceptual equation, and describes examples of how a dynamic risk management system (DRMS) could be used in real-time to identify hazards and control Shiga toxin-producing Escherichia coli risks related to leafy greens.

Download Full-text

Applications of Machine Learning For Precision Agriculture and Smart Farming

The International FLAIRS Conference Proceedings ◽

10.32473/flairs.v34i1.128497 ◽

2021 ◽

Vol 34 (1) ◽

Author(s):

Sai Gurrapu ◽

Nazmul Sikder ◽

Pei Wang ◽

Nitish Gorentala ◽

Madison Williams ◽

...

Keyword(s):

Machine Learning ◽

Big Data ◽

Supply Chain ◽

Precision Agriculture ◽

Farm Management ◽

Time Data ◽

Quantitative Metrics ◽

Smart Farming ◽

Applications Of Machine Learning ◽

Financial Indices

Recent deglobalization movements have had a transformativeimpact and an increase in uncertainty on manyindustries. The advent of technology, Big Data, and MachineLearning (ML) further accelerated this disposition.Many quantitative metrics that measure the globaleconomy’s equilibrium have strong and interdependentrelationships with the agricultural supply chain and internationaltrade flows. Our research employs econometricsusing ML techniques to determine relationshipsbetween commonplace financial indices (such asthe DowJones), and the production, consumption, andpricing of global agricultural commodities. Producersand farmers can use this data to make their productionmore effective while precisely following global demand.In order to make production more efficient, producerscan implement smart farming and precision agriculturemethods using the processes proposed. It enablesthem to have a farm management system that providesreal-time data to observe, measure, and respondto variability in crops. Drones and robots can be usedfor precise crop maintenance that optimize yield returnswhile minimizing resource expenditure. We developML models which can be used in combinationwith the smart farm data to accurately predict the economicvariables relevant to the farm. To ensure the accuracyof the insights generated by the models, ML assuranceis deployed to evaluate algorithmic trust.

Download Full-text