Improved model based decision support using advanced operations research methods – PigIT, University of Copenhagen

Forward this page to a friend Resize Print Bookmark and Share

PigIT > About PigIT > Methodology and results > Improved decision support

Improved model based decision support using advanced operations research methods

A special challenge in relation to decision support models is how to integrate decisions at animal, pen, section and herd level. The pig production systems can be addressed as a general animal replacement problem [15], however, decisions at one level (e.g. animal or pen) must be linked to the capacity constraints at a higher level (e.g. section or herd). In addition, the decisions concerning intervention to avoid welfare and productivity problems will often face different constraints and time horizons, than those aiming at optimizing an ongoing production. Thus, the integration of welfare and productivity related decision making is potentially too complex to produce the optimal decision strategy at all levels simultaneously. Thus, there is a need for approximate methods, which e.g. relax the requirement of an optimal solution and search for a better than current solution [9] and [14]. But also methods which provide better on-line integration with data and address how much the information structure or state-space for the decision complex can be decomposed across and between levels with different time horizons [16, 22, 26] need further development.