Controlling the spread of Mycobacterium avium subsp. paratuberculosis (Map) within and betweenherds still represents a major economic challenge. However, the succession of infection stages and the individual shedding pattern are not well known for infected animals, whereas this precise knowledge is required to identify effective control measures in order to limit new infections.Combining field observations and epidemiological modelling appears to be a relevant approach toproduce such knowledge. However, most existing models do not account for individual characteristics of animals (health status, shedding level), or neglect shedding by infected calves and Map survival in the environment, the latter also being a pivotal process involved in Map persistenceat the herd scale. We propose a new individual-based model that accounts for all of the available knowledge on Maptransmission and shedding. By considering precisely individual characteristics of animals and their variation over time, this model can be calibrated on observed longitudinal field data. Our individual-based model is a stochastic model in discrete time, with a time step of one week. It has been programmed in C++ to ensure computing efficiency. It couples population and infection dynamics within a dairy cattle herd, and considers explicitly all of the farming environments and Mapsurvival. It accounts for the five routes of transmission (in utero, contaminated colostrum/milkingestion, and ingestion of contaminated faeces present in the local and in the general environments), as well as for the early shedding of Map by infected calves. Animals are individually characterized, with information about their age group (unweaned calves, weaned calves, youngheifers, heifers, and adult cows), their progression between health statuses (susceptible, resistant, transiently infectious, latently infected, subclinically infectious, and clinically affected), their contact with farming environments, their shedding level, and their detection test history. With our individual-based model, we aim at assessing biological assumptions, notably concerning the probability of adult infection, shedding dynamics, and the succession between infection statuses