Mechanistic epidemiological modelling is often used in the detection and prevention of livestock diseases, accounting for realistic farming practices. It helps understand the spread of pathogens and compare intervention scenarios. However, it is difficult for decision makers to manipulate mechanistic models and interpret outputs by themselves. To overcome this obstacle, artificial intelligence and software engineering techniques can make mechanistic epidemiological models more accessible to farmers and veterinarians, by transforming such models into user-friendly decision-support tools (DST). To facilitate the co-construction of DSTs, we have established a domain-specific language to specify model components as well as tool features through a textual formalization and automatic code generation. This helps nonmodeller stakeholders to read, assess, and revise the model assumptions or scenarios and the tool structure at any moment, making the whole process more collaborative and accessible. We have illustrated this approach using an epidemiological model of Bovine Respiratory Disease (BRD) in fattening farms. This disease affects young bulls shortly after being allocated into pens. BRD is often difficult to anticipate and control because it is a multi-factorial and multi-pathogen disease, which leads to massive use of antimicrobials. The decision-support tool developed from an existing mechanistic BRD model allows users (farmers, veterinarians...) to describe their farm conditions, possible pathogens, and configure different intervention scenarios. Based on that, they can evaluate the epidemiological and economics outcomes associated with different farming practices, and finally decide how to balance the reduction of disease impact and the reduction of antimicrobial usage. As these methods are generic, they can apply to various infectious diseases and farming practices. Additionally, they require little coding (usually none at all), which is likely to foster a broader use of mechanistic models in veterinary epidemiology in general, and more specifically their further use to support decision-making in practical situations by farmers, veterinarians, or other stakeholders