Currently, the diversity of sow herd management strategies has been described but there are no tools that explore how it promotes sow herd performance nor how it or performance are linked to work organization problems. The goal of the current study was to build a herd dynamic, stochastic object oriented model capable of representing the herd dynamics and performance, and to predict the number of events workers will have to deal with. Each sow is individually represented in the model and the model works as a discrete event simulator with a predefined time step of 1 h. At each time step of simulation, the model searches for an event to be processed. An event may imply change of sow physiological state (e.g. oestrus, far rowing and insemination) and/or request an action from a worker (e.g. oestrous detection and far rowing supervision). This action may result in the planning of a new event (e.g. far rowing after mating) and/or modification of sow state (e.g. from oestrus to pregnant).The occurrences of some technical activities such as weaning are defined in time and frequency according to the management strategy of the farmer. The model is stochastic as sow biology is represented by several normal univariate distributions according to parity or by a threshold (fertility,abortion and mortality rates). When sows return into oestrus after mating they can be moved to another batch or culled depending on batch management strategy and culling policy. Outputs of this model focus on productivity of sows and distribution of tasks over the week. Definitions of the duration of simulation and number of replications to obtain the steady state and the variability of results are presented. The model is able to simulate several batch far rowing systems (BFS) and results of 1-, 3- and 4-week BFS are presented. Several simulations with modified management (no oestrous detection during the weekend and change of the weaning day) or with modified sow biology (increased variability of the weaning-to-oestrus interval and lower fertility rate) are performed. Results indicate that these modifications have specific consequences on performance and task distribution according to the BFS. The model provides useful information concerning the effects of herd management strategies on productivity and distribution of events over time and their sensitivity to biological criteria.