Advancing zoo animal welfare through data science: scaling up continuous improvement efforts
Publication Type: |
Journal Article |
Year of Publication: |
2024 |
Authors: |
Liptovszky,Matyas |
Publication/Journal: |
Frontiers in Veterinary Science |
Keywords: |
animal welfare, artificial intelligence, continuous improvement, data science, evidence-based decision making, machine learning, zoo |
ISBN: |
2297-1769 |
Abstract:
Advancing animal welfare in zoos is a multifaceted endeavor that lies at the core of their conservation and educational missions. While zoos have made significant strides in improving the wellbeing of their animal residents, challenges persist in ensuring a comprehensive, evidence-based approach to continuous improvement. Some of these challenges stem from the exponential growth of animal welfare in the last few decades, making it difficult to keep up with ever evolving theories and practical assessment methodologies. Others are related to the inherent resource limitations of any organization, including staff time and financial resources dedicated to animal welfare. A significant limitation with the latter is staff time allocated to animal observation and other data collection.
Untapped potential currently exists for data science as a transformative tool to overcome some of these challenges and further advance animal welfare practices in zoos. Data science is an interdisciplinary field that involves the extraction of valuable insights and knowledge from large and complex datasets through a combination of statistical analysis, machine learning, and domain expertise. It encompasses various processes, including data collection, cleaning, and transformation, as well as the application of algorithms and machine learning models to uncover patterns, trends, and correlations within the data. Data science leverages programming languages and tools to process and visualize data, enabling data scientists to interpret and communicate findings effectively. It empowers organizations to make informed decisions, predict future trends, and solve complex problems across diverse domains, from business and healthcare to environmental sciences and beyond. As data continues to proliferate in the digital age, data science plays a crucial role in extracting meaningful knowledge and unlocking the potential for innovation and advancements in numerous fields.
The data science lifecycle starts with problem identification and business understanding: the need to deeply understand the operational environment where the data collection and analysis (and any subsequent predictions) will be carried out, and the reasons why we would do that. This provides opportunities for zoos to consider various approaches to how data science can be implemented within existing systems. This can be simple, by analyzing single data sources, or complex, aiming at comprehensive welfare assessments based on all available evidence. A key to the success of this step is domain expertise, in our case expertise from animal care, veterinary and animal behavior and welfare staff. Emphasizing this key first step is also likely an important solution to reduce resistance toward novel technologies and methodologies, as it ensures that the inherent knowledge and expertise of the zoo team are captured.
By adopting a data-driven approach, zoos can capitalize on the wealth of information already available from diverse sources, including readily available daily keeper reports, animal health records, behavioral observations, enclosure data, as well as through emerging technologies such as CCTV footage, environmental sensors, and acoustic recordings. Harnessing the power of data science could allow for efficient and in-depth assessments of animal welfare, enabling zoos to extend their focus to a larger number of animals and make evidence-based decisions for their care. This shift toward data-driven decision-making not only optimizes resource allocation but also leads to a more substantial and positive impact on animal welfare.
Furthermore, integrating data science into daily operations and research can foster collaboration between experts from various fields, including data scientists, academics, and zoo professionals. By encouraging the exchange of knowledge and expertise, this collaborative effort will help zoos address common challenges. Embracing technological change and fostering collaboration will lead to an evidence-based approach for the continuous improvement of animal welfare in zoos. Ultimately, this approach not only benefits the animals under our care but also empowers staff working toward increasingly better animal welfare outcomes and provides vital proof of these efforts for visitors and the broader society.