beFarm - The genesis of a story that has yet to be written

Vem Sistemi, the group I work for, is a beautiful reality founded in the 1980s in Forlì, an ICT system integrator that has grown progressively over the years; when I joined in the remote 04/24/1999 it was a small company concentrated in a warehouse in the industrial area of Forlì, today we are around 400 people in 6 offices distributed in central and northern Italy. A healthy, young reality, a partnered customer base, and constant growth over the years, which invests in training and in the territory.

In 2017 we created a work team dedicated to innovation, of which I assumed the operational leadership. In August of the same year we convinced one of our historic clients, one of the main Italian players in the Agrifood sector, to experiment on a project that had as its ingredients some trends of interest to us: IoT, Cloud Computing, Machine Learning and Big Data Analytics.

The client offers us the opportunity to get our hands into broiler farms. We organize workshops, technical meetings in which we begin to frame the context, understand how the production cycles are structured, which are the most relevant "stomach aches" for technical management, the cost items, finally arriving at the analysis of the main productivity indicators, among which the infamous Food Conversion Rate, FCR, stands out, i.e. the quantity of feed converted into animal weight. Since the cost of feed is approximately 70% of the total costs of a breeding cycle, it is clear how important this indicator is.

They provide us with a proprietary breeding farm managed by a passionate and competent breeder, who welcomes us with enthusiasm and, over time, transforms a team of IT engineers and analysts into competent breeding technicians, capable of understanding the dynamics of the context, to identify the main inefficiencies and attempt to remedy them thanks to digitalisation.

At the same time we begin to install field sensors capable of collecting real-time data on impactful topics such as: environmental parameters within farms, consumption, weight gain, air and water quality. We begin to record the production results at the end of the cycle; we also integrate via software with the processing plants, to also record the qualitative results on live animals delivered from the farms, i.e. the presence of plantar crusts on the legs, various lesions, and ammonia residues in the meat, all indicators attributable to potential animal suffering during the breeding cycle.

We collect data for years, production cycles that went well and cycles that went less well.

I read an enlightening university study on the impact of heat stress on broiler performance.

In the study, a sample of 250 broilers underwent three different heat stress tests in the second week of life; the fourth test was performed keeping the animals in comfortable conditions.

At 21 days, performance was evaluated in terms of Food Conversion Ratio, FCR. The study demonstrates that performance is unequivocally compromised in the presence of thermal stress in the first weeks of life, as the animals' thermoregulatory system is not fully developed.

The best performance on the FCR was observed for the temperature of 30 ºC, confirmed as thermal comfort for the second week of life of the chickens.

The worsening of the FCR, in the different stress conditions, from cold, 24 and 27°C, and from heat, 33°C, was 14.8, 8.6 and 10.1% respectively. Essentially, in conditions of cold stress, the food consumed by animals is mainly spent to keep warm, not so much to gain weight.

We are starting to understand that two issues that are hypothetically at odds with each other, productivity and animal welfare, can actually run parallel, if guided by careful management and supported by data and digitalisation.

We work closely with the University of Bologna, departments of Computer Science and Engineering, and Agro-Food Sciences and Technologies, for the creation of statistical models capable of identifying potential anomalous behavior of animals and defining target curves and reference baselines on issues impacting animal welfare, such as environmental parameters, consumption, growth, bedding and lighting management.

By collecting millions of data we are able, for example, to define a reliable formula to estimate the perceived temperature, which also takes into account ventilation, relative humidity, weight and density of the animals; and which often differs quite a bit from that detected using the thermometers typically used on farms.

We apply the models to our pilot farm and, as if by magic, with each production cycle, we obtain optimal performance in terms of FCR and animal welfare, completely eliminating all indicators of suffering recorded prior to the delivery of the animals.

The incorrect management of environmental parameters within the farms leads to stratified, wet and ammonia-saturated bedding, which necessarily must be milled several times during the production cycle (even 3-4 times in a few days); milling is lethal, causing the inevitable release of ammonia and volatile organic compounds. From the data collected, we have demonstrated that milling causes stress for the animals, a systematic increase in mortality, and annoying odor emissions released into the environment by ventilation systems, with consequent damage to neighboring communities.

Otherwise, careful and competent management of the environment allows you to preserve compact and dry bedding for the entire duration of the production cycle, maximizing animal well-being and minimizing odor emissions released into the environment. Well-preserved litter can be sold to biodigesters at the end of the cycle, while wet litter, remuffed and saturated with harmful gases, must go through an expensive drying process and subsequent disposal.

During the lockdown we work hard, keeping our sights on a powerful "lighthouse in the fog of innovation", i.e. a simple formula:

if we are able to identify situations of animal stress and suffering, we offer farmers the possibility of intervening promptly, preserving well-being and maximizing production performance. Beautiful!

We understand that in addition to the temperature perceived by the animals, it is essential to comprehensively monitor the temperature and humidity of the litter inside the farms. We are looking for devices on the market capable of measuring, accurately and reliably over time, the relevant environmental parameters at animal height. We don't find anything, so we decide to invest, we involve three partner companies with whom we create and patent two innovative devices, Artemis and Tiresia, specifically designed to work in a challenging environment and collect the required data 24 hours a day.

At the same time, we are giving more and more shape and value to the beFarm software platform which, starting from the data collected in the field, offers a simple and intuitive interface, capable of guiding farmers in their daily operations, highlighting potentially critical situations, and guaranteeing standardization and efficiency thanks to digitalisation. A powerful Innovability tool, capable of combining innovation and environmental sustainability.

The beFarm project, born as an experiment incubated in the innovation team of an ICT company, has become a market solution. We are excited to share it with the whole world.