Identifying, quantifying, and fortifying the risk of the us cull sow marketing channel, through the use of analytics

Blair, Benjamin William

Permalink

https://hdl.handle.net/2142/116216 Copy

Description

Title

Identifying, quantifying, and fortifying the risk of the us cull sow marketing channel, through the use of analytics

Author(s)

Blair, Benjamin William

Issue Date

2022-07-13

Director of Research (if dissertation) or Advisor (if thesis)

Lowe, James F

Doctoral Committee Chair(s)

Lowe, James F

Committee Member(s)

• Nguyen, Thanh H
• Rowland, Raymond
• Fang, Ying

Department of Study

Pathobiology

Discipline

VMS - Pathobiology

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

Ph.D.

Degree Level

Dissertation

Keyword(s)

• Cull
• Sow
• Swine
• Markets
• Analytics
• Disease

Abstract

Plagued by endemic disease and under the constant threat of novel pathogen introduction, potential losses in the profitability of the U.S. swine industry from disease are staggering. As a result, the industry has continued to place a spotlight on identifying, quantifying, and fortifying potential routes of pathogen entry into farms. While the industry has made great strides regarding the threat that supplies, animals, feed, vehicles, and aerosols pose to pathogen spread, research regarding numerous high-risk avenues remain uncultivated. Gaps within the collective knowledge of the industry regarding the threat of animal markets, namely the cull sow marketing channel, still exist. The cull sow marketing channel moves 3.2 million animals annually. This significant segment of the industry poses an extensive threat to farms nationwide. The movement of animals within this market channel allows pathogens to spread within the channel and back to farms through the various indirect connections with transport vechicles and facilities. The complexity of these movements and the extended time animals spend within the channel after leaving the farm and before harvest creates an efficient means for untraced pathogen dissemination throughout the industry. These results suggest further quantification of the transmission potential of the cull sow marketing channel is necessary to prepare the swine industry for a novel pathogen introduction. Regulators have proposed to limit the risk of slaughter market channels during a disease outbreak through the standstill of movements either regionally or nationally. While intuitively reasonable, the impact of a standstill on trade patterns and pathogen dissemination potential is unquantified. Implementing an augmented gravity model facilitates the quantification of effects that a standstill may have on trade patterns based on the population size and standstill imposition location. The trade variation induced by the closure of or standstill around individual processing facilities (slaughter plants) increases the potential indirect contact between sows within the market and various farm populations. These models suggest that identification of infected farms prior to animal shipment is essential before the implemention of a regionalized standstill if the spread of a pathogen within the U.S. is to be limited. The tools for syndromic surveillance within sow farms are limited. Available mechanisms, derived and adapted from industrial engineering and originally intended to improve physical manufacturing processes, are based on detecting increased variation levels induced by changes in the production process. Due to biological and process implementation variation, the inherent variability of swine production severely limits their effectiveness. The analytical robustness of machine learning can potentially combat the limitations imposed by a highly variable system. The development of a machine learning tool to monitor unexpected reproductive failure within a single farm identifies the introduction of a novel Porcine Reproductive and Respiratory Syndrome virus (PRRSv) approximately two weeks before diagnostic confirmation, which was initiated by human observation of clinical signs. This tool identifies production changes within a farm 2.5 weeks before the previously described EWMA method. While further validation is required, this machine learning tool may serve as an efficient means to quickly and accurately detect production disruptions, commonly disease. This work is the first to identify, quantify, and fortify the risk posed by the cull sow marketing channel. While much information is still unknown, these studies increase the base knowledge of the industry regarding this significant sector. These results will allow regulators and producers alike to make better decisions in ensuring a secure pork supply as continual evaluation of such a dynamic system is needed to ensure a consensus about the risk and potential avenues of mitigation for this marketing network.

Graduation Semester

2022-08

Type of Resource

Thesis

Copyright and License Information

Copyright 2022 Benjamin Blair

Owning Collections