Résumé
Models and scientific software are currently hardly shared among and (re-)used by stakeholders in the fooddomain, even though they are research products that can support the transfer of knowledge from academia toindustry or to students. Scientific software generally incorporates models that capture fundamental domain in anexplicit format, i.e. math equations, this makes them relevant knowledge transfer medium, under the conditionthat sufficient contextual information is provided along the software to favour the effective diffusion re-use byusers.Hence, what academia should do to support a better re-used of the scientific models ?To provide an answer to this question, we reviewed selected approaches, best practices, hurdles and limitationsregarding knowledge transfer via software and the mathematical models embedded in it, to provide points ofreference for the food community. In this purpose, we focused on three aspects. First, the publication of digitalobjects on the web, which offers valorisation software as a scientific asset. Second, building transferrablesoftware as a way to share knowledge through collaboration with experts and stakeholders in the domain of foodscience and technology. Third, developing food engineers’ modelling skills through the use of food models andsoftware in education and training.By highlighting miscellaneous approaches regarding scientific software, this communication aims at promotingknowledge transfer between and within academia, industry and other stakeholders, and at opening prospects forsynergistic efforts that will allow the food community to face the oncoming challenges