Exploring the promise of data-driven animal feeding

Learn how new technologies will provide feed management insights, allowing farmers to take immediate action to improve feed intake, profitability.

0911 F Maidanconnolly1 Headshot
Creating Software And Website For Different Digital Platform Des
Andrey Suslov | Bigstock.com

New technologies offer feed management insights, allowing farmers to take immediate action to improve feed intake, profitability

It is humorously said that there are three feeds; the one the nutritionist formulates, the one that is made at the feed mill and the one the animal eats. As with all jokes, there is more truth in the idea than most are willing to admit.  The challenge of controlling all three points in the chain is even greater in an era when farms and feed mills are in the middle of pandemic.

COVID-19 has raised some ugly questions. What happens if nutritionists, veterinarians and extension agents are banned from visiting farms? Or what happens if visits are just substantially curtailed? Can new feed additives, technologies or innovations be introduced to a business where the need for face to face visits have been so much a part of our business culture?

I believe the feed industry needs to focus on five deliverables:

  • Safety
  • Transparency
  • Sustainability
  • Government regulations
  • Prosumers

Safety has taken on a new meaning with the advent of COVID-19, emphasizing the need for visibility and real-time information, which can be enhanced through sensors, internet-connected devices and artificial intelligence (AI). Clearly the adoption of robots in feed factories and on farms will be driven by recent events, since removing human labor can reduce the effects of staff shortages, or passage of infections, during a lockdown.

Transparency is becoming a consequence of increasing consumer demands and government regulation. The desire to have ingredient origins be traceable is to support faster implementation of immutable record-keeping such as through blockchain.

Sustainability in feed production and feeding animals on farm is best modeled through the analysis of big data, and predictive models, and large food companies are demanding this from all stakeholders in the chain. Government regulations are becoming more onerous, even in the United States and, in my opinion, this is a consequence of ever greater analytical capacities to identify problems.

Prosumers proactively are driving the agenda when it comes to the food chain, demanding a food system that fits with their expectations and values, and social media gives them a unique platform, but equally social media has allowed greater interaction between farmers and consumers, with the opportunity to set the record straight.

The emergence of Agriculture 4.0 can be seen in the growth of investment in artificial intelligence, especially in food tech and crop tech — but now increasingly in livestock tech.

Ag tech in feed production

The feed industry faces challenges addressing the “three feeds” and using technology to respond.

The feed we formulate:

  • Can data allow us to feed for maximum performance?
  • What impact does nutrition have at gene level, e.g. Alltech’s Nutrigenomics?
  • How do we feed the microbes in the rumen, in the gut, and not focus on just the animal’s needs, e.g. the work done at the University of Illinois?
  • What is the impact of nutrients in combination rather than rationalizing formulations in terms of ingredients individually?

The feed we make:

  • Near-infrared spectroscopy (NIR) has made real-time information on nutritional content of feed at the mill simpler and is becoming increasingly routine.
  • The use of hand-held NIR in the field allows the analysis of feeds, forages, e.g. NutriOpt by Nutreco.
  • Big data and connectivity to the cloud connects into real-time formulation adjustments, e.g. the Max System and Dynamic Nutrition from Cargill.

The feed they ate:

  • Camera/machine learning allows us to digitize the images of consumption and show what was actually consume, e.g. Cainthus’ Alus Nutrition on large U.S. dairy farms, with sensors being used to evaluate feed intake in pigs in real time.
  • In-vitro analysis for monogastric feeds rely upon big data to predict the impact of feed additives such as enzymes, e.g. those being developed by AB Vista.
  • Fermentrics is a Canadian startup already feeding 700,000 cows where the digestibility of the feed is evaluated as a whole, modeled using AI, and not based on single ingredient digestibility

All of the systems above are underpinned by the use of AI.

AI comes in many forms and even primitive systems such as the first calculators could be said to have been based on artificial intelligence, but the key here are systems that are going beyond simply replicating human cognitive function to bring insights the human mind cannot comprehend let alone calculate, generating information on the complex interactions nutrition can on the animal, its genome and its microbiota.

The development of AI-based solutions differs from traditional nutritional innovations in that it is increasingly not being tested and developed in universities or research institutions but being developed directly on farm. AI products by their nature require real world data, and self-learning and self-correcting occurs when more farms, and more real-life experiences (data) are required to generate robust solutions.

The future

In a world where growing restrictions will make it difficult for people to visit farms, what does this mean for feed production and feed mills? Real time 24/7 insights into what exactly livestock consume through AI may allow us to change the nutrient profile, the ingredients we use, when we feed, how we feed.  Such insights generate remotely can go beyond traditional nutrition and result in improved performance, or at a lower cost, while responding better to consumer demands.

Page 1 of 8
Next Page