"Big Data Management Within Multi-Objective Optimization Modeling for S" by Breanna Paige Modica

Date of Conferral

1-17-2025

Degree

Ph.D.

School

Information Systems and Technology

Advisor

Paul Frankenhauser

Abstract

Producing enough food to feed a growing human population without further degrading the environment through unsustainable natural resource use is a global issue. The purpose of this quantitative nonexperimental study was to examine the extent to which big data management of environmental impact sources, assessment methods, and allocation methods in multi-objective optimization (MOO) modeling affect the consistency and predictability of sustainable food animal feed management practices at the farm level. Representation theory, the theory of effective use of information systems, the dynamic capabilities view, and the natural resource-based view were used to describe the necessity of accurately representing and managing the real-world environmental impact of big data within information systems to increase sustainable food animal production. This study’s secondary data sources included the U.S. Department of Agriculture. Nonparametric statistical tests were used to analyze broiler chicken, beef cattle, and swine feed formulations (N = 111) developed via MOO modeling. The results indicated that environmental impact source, assessment method, and allocation method significantly affected the consistency and predictability of sustainable feed formulation at the farm level. Not all predictor variables significantly affected all criterion variables, such as the assessment method: (a) number of ingredients (p = .005), (b) crude protein content (p < .001), (c) LO (p < .001), and (d) TAP (p = .015). The implications for positive social change include the potential for U.S. food animal producers to apply sustainable food animal feed formulations from environmental impact sources, assessment, and allocation method compatibilities to produce enough food for a growing human population.

Download

Share

 
COinS