Mechanistic models (MMs) have served as causal pathway analysis and ‘decision-support’ tools within animal production systems for decades. Such models quantitatively define how a biological system works based on causal relationships and use that cumulative biological knowledge to generate predictions and recommendations (in practice) and generate/evaluate hypotheses (in research). Their limitations revolve around obtaining sufficiently accurate inputs, user training and accuracy/precision of predictions on-farm. The new wave in digitalization technologies may negate some of these challenges. New data-driven (DD) modelling methods such as machine learning (ML) and deep learning (DL) examine patterns in data to produce accurate predictions (forecasting, classification of animals, etc.). The deluge of sensor data and new self-learning modelling techniques may address some of the limitations of traditional MM approaches – access to input data (e.g. sensors) and on-farm calibration. However, most of these new methods lack transparency in the reasoning behind predictions, in contrast to MM that have historically been used to translate knowledge into wisdom. The objective of this paper is to propose means to hybridize these two seemingly divergent methodologies to advance the models we use in animal production systems and support movement towards truly knowledge-based precision agriculture. In order to identify potential niches for models in animal production of the future, a cross-species (dairy, swine and poultry) examination of the current state of the art in MM and new DD methodologies (ML, DL analytics) is undertaken. We hypothesize that there are several ways via which synergy may be achieved to advance both our predictive capabilities and system understanding, being: (1) building and utilizing data streams (e.g. intake, rumination behaviour, rumen sensors, activity sensors, environmental sensors, cameras and near IR) to apply MM in real-time and/or with new resolution and capabilities; (2) hybridization of MM and DD approaches where, for example, a ML framework is augmented by MM-generated parameters or predicted outcomes and (3) hybridization of the MM and DD approaches, where biological bounds are placed on parameters within a MM framework, and the DD system parameterizes the MM for individual animals, farms or other such clusters of data. As animal systems modellers, we should expand our toolbox to explore new DD approaches and big data to find opportunities to increase understanding of biological systems, find new patterns in data and move the field towards intelligent, knowledge-based precision agriculture systems.

The Kempen system is a dairy feeding system in which diet is provided in the form of a compound feed (CF) and hay offered ad libitum. Ad libitum access to CF and hay allows cows in this system to achieve a high DM intake (DMI). Out of physiological concerns, the voluntary hay intake could be increased and the consumption pattern of CF could be manipulated to maintain proper rumen functioning and health. This study investigated the effects of an artificial hay aroma and CF formulation on feed intake pattern, rumen function and milk production in mid- to late-lactating dairy cows. Twenty Holstein–Friesian cows were assigned to four treatments in a 4 × 4 Latin square design. Diet consisted of CF and grass hay (GH), fed separately, and both offered ad libitum, although CF supply was restricted in maximum meal size and speed of supply by an electronic system. Treatments were the combination of two CF formulations – high in starch (CHS) and fibre (CHF); and two GH – untreated (UGH) and the same hay treated with an artificial aroma (TGH). Meal criteria were determined using three-population Gaussian–Gaussian–Weibull density functions. No GH × CF interaction effects on feed intake pattern characteristics were found. Total DMI and CF intake, but not GH intake, were greater (P < 0.01) in TGH treatment, and feed intake was not affected by type of CF. Total visits to feeders per day, visits to the GH feeder, visits to the CF feeder and CF eating time (all P < 0.01) were significantly greater in cows fed with TGH. Meal frequency, meal size and meal duration were unaffected by treatments. Cows fed CHF had a greater milk fat (P = 0.02), milk urea content (P < 0.01) and a greater milk fat yield (P < 0.01). Cows fed TGH had a greater milk lactose content and lactose yield (P < 0.05), and milk urea content (P < 0.01). Cows fed TGH had smaller molar proportions of acetic acid and greater molar proportions of propionic acid compared with UGH. In conclusion, treatment of GH with an artificial aroma increased CF intake and total DMI, but did not affect hay intake. Additionally, GH treatment increased the frequency of visits to both feeders, and affected rumen volatile fatty acid profile. Type of CF did not affect meal patterns, ruminal pH, nor fermentation profiles.

From 2007 to 2010, the largest reported Q-fever epidemic occurred in the Netherlands with 4026 notified laboratory-confirmed cases. During the course of the epidemic, health-seeking behaviour changed and awareness among health professionals increased. Changes in laboratory workflows were implemented. The aim of this study was to analyse how these changes instigated adjustments of notification criteria and how these adjustments affected the monitoring and interpretation of the epidemic. We used the articles on laboratory procedures related to the epidemic and a description of the changes that were made to the notification criteria. We compared the output of a regional laboratory with notifications to the regional Public Health Service and the national register of infectious diseases. We compared the international notification criteria for acute Q-fever. Screening with ELISA IgM phase II and PCR was added to the diagnostic workflow. In the course of the epidemic, serology often revealed a positive IgG/IgM result although cases were not infected recently. With increasing background seroprevalence, the presence of IgM antibodies can only be suggestive for acute Q-fever and has to be confirmed either by seroconversion of IgG or a positive PCR result. Differences in sero-epidemiology make it unlikely that full harmonisation of notification criteria between countries is feasible.