Recipient Organization
NORTH CAROLINA A&T STATE UNIV
1601 EAST MARKET STREET
GREENSBORO,NC 27411
Performing Department
AGRIBUSINESS, APPLIED ECONOMICS, AND AGRISCIENCE EDUCATIO
Non Technical Summary
Small-scale farms are a major component of U.S. agriculture. According to the 2017 Census of Agriculture, family or individual farms represent approximately 98 percent of all farms, small farms make up approximately 90 percent. Due to the diversity among such farms, it is important that they are viable and sustainable. Several of the NC A&T SU stakeholders fall under the category of farming households with low sales (65 percent of farming households with income less than $63,179). To address these needs, research and extension will be conducted to identify and promote the sustainability of small-scale farms by using Big Data (BD) applications for decision-making.NC A&T SU and the College of Agriculture and Environmental Sciences (CAES) have been committed to exploring multidisciplinary approaches to data science/analytics and to offering faculty positions to cover a cluster of disciplines. In particular, the Department of Agribusiness, Applied Economics and Agriscience Education seeks to expand its research portfolio to explore opportunities to use BD applications in North Carolina small-scale agriculture. Therefore, this project seeks to gain more knowledge of how BD applications can be used in the 2020 Critical Issues and Science Emphasis Areas - a) Improving Plant and Animal Agricultural Systems and/or b) Protecting Environmental and Natural Resources. This project connects with the CAES initiative for the Small-Scale Farming Resource and Innovation Center (SFRIC), which is to strengthen the small-scale operator's ability to compete in the global marketplace. With increasing interest in BD applications in agriculture, the producers' adoption of BD-informed agricultural practices may strengthen the small-scale farmers' ability to commercialize not only locally, but also nationally and globally.We seek to gain a better understanding of BD applications at the small-scale farming level and to deliver science-based knowledge to policymakers, industry, non-profit organizations, and small-scale farming operations. The objectives of the study are twofold: (1) to conduct an exploratory study on understanding appropriate farm management Big Data (BD) applications in North Carolina small-scale agriculture and (2) to provide recommendations to stakeholders for suitable farm management Big Data (BD) applications in North Carolina small-scale agriculture. In fulfilling the first objective, we plan to first (1) request the suitable datasets from the USDA Economic Research Service, (2) attend the appropriate data training workshops required of recipients of ARMS datasets, (3) review and study the dataset in determining the appropriate classifications and methods that should be used (essentially, the data will need to be evaluated based on scale), and (4) analyze proper prediction models based on different learning algorithms. In fulfilling the second objective, we expect to gain more insight into the predictors of profitability and potential financial and economic decision tools for individual small-scale farmers in North Carolina. We also plan to educate stakeholders of the potential economic and market benefits of specific farm management decisions. This will be conducted by (1) present findings at respective professional meetings in the area of Agricultural and Applied Economics, (2) publish findings in suitable journals in Agricultural and Applied Economics, (3) enhance quantitative methods courses for undergraduate and graduate curriculum enhancement, and (4) conduct workshops in collaboration with the Center for Environmental Farming Systems to stakeholders interested in using BD applications to enhance plant and livestock production and/or better conserve natural resources. It is our overall expectation to show the potential economic impacts of these BD applications on small-scale farming operations and to North Carolina agriculture.
Animal Health Component
100%
Research Effort Categories
Basic
(N/A)
Applied
100%
Developmental
(N/A)
Goals / Objectives
Our primary purpose in conducting this research is to understand how BD applications can be best used to support small-scale farming operations in the state of North Carolina. Small-scale farmers have the propensity to produce a more diverse volume of higher quality agricultural commodities than their large-scale counterparts. This provides an opportunity to meet the growing demands of a more global society, which allows for creating and sustaining competitive advantages. The project is to evaluate how BD applications can be used as decision-marking tools to enhance plant and livestock production and/or better conserve natural resources. The project is also to explore the potential economic impacts of these BD applications to decisions on small-scale farming operations and to North Carolina agriculture. Findings will be disseminated using workshops, social media, electronic mailings, professional presentations, and peer-reviewed journal articles. In addition, findings will be used to enhance the undergraduate and graduate curricula for the BS and MS Agricultural and Environmental Systems (Agribusiness and Food Industry Management) programs to better instruct North Carolina citizens for further enhancement of rural and urban community development. Courses such as ABM 406 Quantitative Analysis, ABM 641 Special Problems in Agribusiness Management, ABM 705 Statistical Methods for Agribusiness, and ABM 708 Econometrics in Agribusiness, will be enhanced from the findings on this search. Students showing interest will be invited to participate in quantitative analysis training and mentorship. In addition, graduate and undergraduate students will be mentored throughout their assistance with the project ultimately contributing to North Carolina small-scale agriculture.We are seeking funding to develop a research program in understanding BD applications at the small-scale farming level and to deliver science-based knowledge to policymakers, industry, non-profit organizations, and small-scale farming operations. The objectives ofthe study are twofold:Objective One: To conduct an exploratory study on understanding appropriate farm management Big Data (BD) applications in North Carolina small-scale agriculture.Objective Two: To provide recommendations to stakeholders for suitable farm management Big Data (BD) applications in North Carolina small-scale agriculture.
Project Methods
In this study, we propose to take a closer look at potential BD applications that enhance small-scale producer decision-making capabilities. Therefore, the overall purpose of the study is to understand how BD applications can be best used to support small-scale farming operations in the state of North Carolina.Objective One: To conduct an exploratory study on understanding appropriate farm management Big Data (BD) applications in North Carolina small-scale agriculture.Procedure One - We propose to evaluate how BD applications can best be used as farm management decision-making tools to enhance plant and livestock production and/or better conserve natural resources. As noted in Section 3, BD is widely privatized due technological innovation property rights of the various capturing devices such as land-based sensors and aerial drones. However, we plan to access public data gathered from the U.S. Department of Agriculture's Agricultural Resource Management Survey (ARMS) (U.S. Department of Agriculture, 2020). The data captured through the ARMS survey reflects information on the production practices, resource use, and economic well-being of America's farms and ranches. It is the only source of information available for objective evaluation of many critical issues related to agriculture and the rural economy.We plan to utilize ARMS data and to explore the use of Machine Learning methods to successfully predict factors that lead to small-scale farms' profitability. Machine Learning focuses on the development of computer programs that can access data and use it to learn. The process begins by learning from the input data and then interpreting and analyzing the input and output data to create machine algorithms. The algorithms then construct a system model, which is used to predict future values. Supervised Learning is widely used with relatively large dataset and then inputs are fed into the neural network with a desired known value. Artificial Neural Networks (ANN), a type of Supervised Learning method, are widely used in agriculture has been be used to address Agricultural Economics issues including the prediction of farmland values, weather forecasting, crop yield prediction and crop selection, irrigation systems, crop disease prediction, agricultural policy and trade, production management and profitability (Coble et. al, 2018). The attractiveness of ANN begins with the fact that it has an advantage over multiple regression. It can select an independent variable in the data, learn complex relationships, and does not place strict requirements a priori on a functional form.In this project, we plan to use ANN primarily to address decisions influencing profitability. ANN has been used to predict credit demand (Ifft et. al, 2017), to identify factors affecting profitability of poultry egg farm production (Johnson, 2020), to estimate the profitability of dairy farms (Yli-Heikkila & Tauriainen, 2014), and to evaluate the adoption and profitability of using precision farming (Griffin et. al, 2004).In fact, Griffen et. al. (2004) and Ifft et. al. (2017) utilize ARMS in their studies. We plan to also utilize the data to learn more about small-scale producer decision-making processes. During this exploratory portion of the study, we plan to first (1) request the suitable datasets from the USDA Economic Research Service, (2) attend the appropriate data training workshops required of recipients of ARMS datasets, (3) review and study the dataset in determining the appropriate classifications and methods that should be used (essentially, the data will need to be evaluated on the basis of scale), and (4) analyze proper prediction models based on different learning algorithms.Objective Two: To provide recommendations to stakeholders for suitable farm management Big Data (BD) applications in North Carolina small-scale agriculture.Procedure Two - We expect to gain more insight into the predictors of profitability and potential financial and economic decision tools for individual small-scale farmers in North Carolina. We also plan to educate stakeholders of the potential economic and market benefits of specific farm management decisions. This will be conducted by (1) present findings at respective professional meetings in the area of Agricultural and Applied Economics, (2) publish findings in suitable journals in the area of Agricultural and Applied Economics, and (3) conduct workshops in collaboration with the Center for Environmental Farming Systems to stakeholders interested in using BD applications to enhance plant and livestock production and/or better conserve natural resources. It is our overall expectation to show the potential economic impacts of these BD applications on scall-scale farming operations and to North Carolina agriculture.