With modern technology playing an increasing role in the day to day management of farms, what knowledge can be extracted from the huge volumes of data we generate as an industry? ‘Big Data’ is the buzzword of the IT and corporate world at the moment, but how can poultry breeders take advantage?
“You can’t manage what you don’t measure.” This quote by two famous consultants, W. Edwards Deming and Peter Drucker, shows the importance of measuring and analysing information. The information being currently collected in the poultry industry allows managers to measure their farm production trends, which ultimately leads to improved decision making and bird performance.
The use of Big Data has the potential to transform poultry business. It may even offer a stronger competitive advantage as the concept is slowly adopted by others. As we’ll discuss in more detail, this Big Data revolution is more powerful than the analytics used in the past. We can measure, predict, and find cause and effect relationships more precisely than ever before, allowing us to make smarter decisions. We can target more effective customer support using accurate data analytics instead of “gut” intuition.
As tools to manage large amounts of data improves, they will challenge conventional management practices. Smart leaders will be using Big Data as it was intended: a management revolution. But the challenges of becoming a Big Data enabled organisation can be enormous and require hands on leadership. So what do we mean by Big Data? Really it is about handling large volumes of data quickly from a range of different sources and then analysing the data to make management decisions.
As technology enables us to transfer information from the farms, processing plants and hatcheries into servers in corporate offices quicker and easier than ever before, the amount of information collected has increased exponentially. This has provided companies with an opportunity to work with a huge volume of data for a better understanding of their own operation(s) by understanding response to changes made in production over time and seeking to repeat historical success without complete reliance on research conducted in environments which often don’t reflect current reality.
For many applications, the speed of data collection is even more important than the volume. The speed at which data can be gathered and analysed allows a company to respond to issues before they become a bigger and costly problem. However, this is still a far reality for some in the poultry industry as farms are sometimes in remote areas lacking network connections to servers.
Big Data is a conglomerate of information from various sources such as weather stations near production farms, environmental control systems at the farm, and machines at a hatchery. As costs of all the elements of computing (memory, processing, and bandwidth) continue to decline, previously costly data-intensive approaches are quickly becoming more affordable. Although the data available are often unstructured (i.e. not organised in a database), there’s a huge amount of potential information waiting to be extracted from the noise.
Being a primary breeder, Aviagen has been using data analytics to make breeder decisions for many years. However, over the past few years, we have decided to extend these data analysis services to our customers as well. This decision resulted in the production of a set of statistical software designed to analyse the “ocean of data” collected by our customers.
We invested in developing tools to:
While data allow us to identify problem areas that impact production, it does not replace the need for skilled managers and consultants, as they continue to deeply understand the biological changes of the modern bird. Poultry companies will continue to rely on their expertise to determine where the biggest opportunities and challenges lie, and their experience will be invaluable in guiding the company towards their next plan of action. However, Big Data can greatly enhance the decision making process. As the Big Data movement advances, the role of consultants will shift. They’ll continue to be valued because they know what questions are the right questions to ask during analysis.
As data become cheaper, the specialists who work with data become more valuable. A good example are data scientists with a poultry background and other professionals skilled at working with large quantities of information. Even though statistical knowledge is important, many of the key techniques for using Big Data are rarely taught in traditional statistics courses. The skills for cleaning and organising large data sets are extremely important as data rarely come in structured formats. Along with data scientists, new generations of database experts are bringing techniques for working with very large data sets. Expertise in the design of trials can help close the gap between cause and effect. The best data scientists are also comfortable speaking the language of the poultry business and translating the results in a way the audience can easily understand.
The tools available to handle all aspects of Big Data have improved greatly in recent years, and the price of these technologies are less expensive than ever. However, the skills required to integrate all the relevant internal and external source of data into one location to be available for analysis are new to most IT departments.
Effective organisations place information and decision makers in the same location. The best leader will create an organization flexible enough to minimise the “not invented here” syndrome and maximise cross functional co-operation. It is essential to bring together the right data with the problem solvers and those who understand the biological impact.
The first question a data driven organisation asks itself is not “What do we think?”, but “What do we know?” This requires a departure from acting solely on hunches and instinct. It also requires companies to realise they often pretend to be more data-driven than they actually are. Too often we see managers who spice up their reports with lots of “selective data” that support decisions they have already made using the intuition approach. There is no question that difficulties to succeed remain. There are too few data scientists with poultry background to go around. Some groundbreaking technologies to analyse poultry data continue to arise. The cultural challenges are vast, and, of course, privacy concerns are only going to become more important. But the underlying trends, both in the technology and in the return on investment, are unmistakable. The evidence is clear: Data-driven decisions offer many advantages over the intuition approach. Companies that are able to successfully combine domain expertise with data science will outperform their competitors.