An evaluation of Key Performance Indicators for beef herds

Hewitt, Sarah (2022) An evaluation of Key Performance Indicators for beef herds. PhD thesis, University of Nottingham.

[img]
Preview
PDF (Thesis - as examined) - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
Available under Licence Creative Commons Attribution.
Download (4MB) | Preview

Abstract

Key performance indicators (KPIs) can be used to monitor progress towards predefined targets. They are used widely across many industries, and although their use in the beef sector currently lags that in the dairy, pork and poultry sectors, it is growing with farmer appetite for data driven decision making. There is limited evidence behind many of the commonly suggested metrics however, and although they have typically been developed through evaluation of expert opinion, analysis of the associations between these metrics and overall enterprise success is lacking. There are several reasons for this; data can be more challenging to capture in more extensive systems (typical of beef suckler enterprises especially). Small herd sizes and a long production cycle also limits the quantity of data available, and the diversity of the sector presents challenges around data continuity. Beef enterprises may operate under tighter margins than other livestock enterprise types, so there may also be a financial barrier to data capture and analysis. Farmers are often unsure of how to make the best use of their data, so in addition to improving recording, there is also substantial value that could be added by making the best use of whatever data is available (such as legally required movement data).

This project used a combination of focus group discussion and a questionnaire to evaluate farmer and adviser opinion around performance metrics for beef herds. Six focus group meetings were held over 18 months and 140 responses from UK beef farms, including 107 suckler farms, were collected by questionnaire survey. This led to the development of a KPI ‘toolkit’ with calculation methods and definitions. In order to demonstrate the value of the metrics, regression analysis was carried out using data from a single beef finishing unit in the East Midlands. The dataset contained 16,248 animal records from 2010 to 2016. Predictors of daily liveweight gain (DLWG) and antibiotic treatments were investigated. Predictors of DLWG included purchase price, month of purchase, source of purchase, breed and age of animal, and whether the animals had been given any antibiotic treatments. Predictors of antibiotic treatment included age at purchase and weight for age at purchase.

Linear regression analysis of an AHDB Stocktake dataset containing 56 suckler and 36 grower/finisher farms between 2013 and 2015 was used to evaluate the associations between performance metrics in the KPI toolkit, and overall enterprise success (defined as net margin per cow bred for suckler herds and net margin per head of output for grower or finisher herds). Metrics such as age at first calving, scanning percentage, weaning weight and mortality rate were found to be significantly associated with net margin per cow bred in suckler units. In contrast, only financial metrics, such as feed cost per head, were found to be significantly associated with net margin per head of output in grower or finisher herds.

To further investigate the relationships between metrics and enterprise success, a stochastic simulation model was developed representing a suckler herd. This was used to generate data from 10,000 herds of 200 suckler cows which could then be analysed using multiple regression. The results of this were used to further influence the structure of the KPI toolkit, and to provide example effect sizes for changes in performance. For example, a change in weaning weight per cow bred from the median (227kg) to the upper quartile (246kg) was associated with an increased net margin per cow bred of £19.96. Relationships between performance indicators and enterprise success such as these could be used to further assist beef farmers with data driven decision making.

A mixed methods approach has been used to evaluate KPIs for monitoring beef herd performance. Focus group discussions and surveys have been combined with both real herd data and simulated data, with the aim of evaluating not only what is possible to monitor and record on a regular basis, but also what is practical and useful. The close involvement of stakeholders has helped to ensure that outcomes are relevant to the beef industry and has facilitated knowledge transfer.

Item Type: Thesis (University of Nottingham only) (PhD)
Supervisors: Hudson, Christopher
Green, Martin
Keywords: Data driven decision making; Data capture; Data analysis; Performance metrics; Stochastic simulation model
Subjects: S Agriculture > SF Animal culture
Faculties/Schools: UK Campuses > Faculty of Medicine and Health Sciences > School of Veterinary Medicine and Science
Item ID: 69382
Depositing User: Hewitt, Sarah
Date Deposited: 29 Jul 2022 04:40
Last Modified: 29 Jul 2022 04:40
URI: https://eprints.nottingham.ac.uk/id/eprint/69382

Actions (Archive Staff Only)

Edit View Edit View