Multi-level hierarchic Markov processes as a framework for herd managementsupport

A general problem in relation to application of Markov decision processes t o real world problems is the curse of dimensionality, since the size of the state space grows to prohibitive levels when information on all relevant t raits of the system being modeled are included. In herd management, we face a hierarchy of decisions made at different levels with different time hori zons, and the decisions made at different levels are mutually dependent. Fu rthermore, decisions have to be made without certainty about the future sta te of the system. These aspects contribute even further to the dimensionali ty problem. A new notion of a multi-level hierarchic Markov process special ly designed to solve dynamic decision problems involving decisions with var ying time horizon has been presented. The method contributes significantly to circumvent the curse of dimensionality, and it provides a framework for general herd management support instead of very specialized models only con cerned with a single decision as, for instance, replacement. The applicatio nal perspectives of the technique are illustrated by potential examples rel ating to the management of a sow herd and a dairy herd.