REPROsheep: A model that integrates individual variability to optimise hormone-free reproduction management strategies for a dairy sheep flock

• Published in: Computers and electronics in agriculture 2021-10, Vol.189, p.106412, Article 106412
• Main Authors: Laclef, E., Debus, N., Taillandier, P., González-García, E., Lurette, A.
• Format: Article
• Language: English
• Subjects: Age; Agent-based model; Agricultural sciences; Animal production studies; Animal reproduction; Animal welfare; Context; Dairy sheep; Flock performances; Hormone-free reproduction; Hormones; Individual variability; Life Sciences; Management strategies; Milk; Optimization; Public health; Sensitivity analysis; Sheep; Synchronism
• ISSN: 0168-1699; 1872-7107
• DOI: 10.1016/j.compag.2021.106412

•A model was built simulating hormone-free dairy sheep flock performance over time.•An agent-based modelling method was used to integrate individual oestrus variability.•Ewe body condition score, age and milk-yield seem to impact flock performance. Managing reproduction of a dairy sheep flock without using hormones for oestrus induction and synchronisation is an alternative method that complies with agroecological principles, animal welfare and public health concerns. However, shifting from conventional reproductive management to a hormone-free philosophy means leaving more room for individual intraflock variability in oestrus occurrence. The success of establishing alternative hormone-free reproduction systems relies on managing the variability of ewes’ individual responses, which are themselves influenced by individual characteristics. Therefore, the so-called REPROsheep model was built to study the effects of individual ewes’ characteristics, such as age, body condition score (BCS) or milk yield, on individual responses, as well as the overall productive and reproductive performances of the flock. This dynamic agent-based model (ABM) represents individual reproduction and lactation processes in a hormone-free management context. Outputs at the flock level were simulated using this model. Two dynamic outputs (weekly number of ewes lambing and weekly milk yield) and two punctual outputs (total number of ewes lambing and total milk yield) were simulated. Sensitivity analysis showed that the studied outputs were affected by ewe age, pre-mating BCS and milk-yield distributions in the flock. Increasing average age or pre-mating BCS by 0.5 from a starting point of 2.5 and average milk-yield by 50 L from a starting point of 195 L appeared to increase, for one breeding season, the number of ewes lambing, the concentration of those lambing events over one period, as well as total milk yield, whereas decreasing the average value of these factors by the same amount appeared to produce the opposite effect. Management strategies favouring an optimum age and pre-mating BCS, as well as good milk yield for the individuals of the flock, may thus be a lever for optimising performance, especially lambing distribution over time. Therefore, the REPROsheep model seems relevant to determine how a management strategy (affecting flock structure in terms of age, body condition or production potential) can impact the performance of a dairy sheep flock in the context of hormone-fre