A sow herd decision support system based on an embedded Markov model

A sow herd decision support system (SH-DSS) was developed, based on an embedded Markov process, representing sow herd dynamics and was implemented using object-oriented programming. It is primarily intended for pig managers, advisers and extension specialists. The SH-DSS is aimed to analyse production factors associated with reproductive and replacement management strategies of sow farms. Decision support is based on a comparative stationary analysis performance from model outputs. The user can explore and evaluate different reproductive and replacement management strategies and therefore can anticipate problems and make the most appropriate decisions. A graphic user interface provides the interaction with the system. The SH-DSS provides several tools to test the model's suitability in relation to a user's historical farm data. Its inputs include technical and economic parameters that characterise specific farm production situations. Technical parameters are those defining sow production (abortion rates, sow mortality and fertility rates, prolificity) and management characteristics of the farm (lactation length, maximum number of matings allowed, maximum number of farrowings). Economic parameters are related to the farm economic environment (feed prices, value of weaned piglets, value of purchased gilts, value of slaughtered sows). The outputs include technical and economic indices, herd structure and the sensitivity analysis of strategic parameters. The model assumes that the herd population is in steady-state.

[1]  Lluis M. Pla,et al.  A Markov decision sow model representing the productive lifespan of herd sows , 2003 .

[2]  Jan Tind Sørensen.  Validation of livestock herd simulation models: a review , 1990 .

[3]  Ruud B.M. Huirne,et al.  An application of stochastic dynamic programming to support sow replacement decisions. , 1993 .

[4]  J.A.L.M Kamp.  Knowledge based systems: from research to practical application: pitfalls and critical success factors , 1999 .

[5]  Martin L. Puterman,et al.  Markov Decision Processes: Discrete Stochastic Dynamic Programming , 1994 .

[6]  Jean Naveau.  Optimierung der zuchtung innerhalb einer geschlossenen zuchtlinie mit schwacher inzuchtrate , 1976 .

[7]  M. A. Allen,et al.  A simulation model for a swine breeding unit producing feeder pigs , 1983 .

[8]  R.B.M. Huirne,et al.  Computerized management support for swine breeding farms. , 1990 .

[9]  Efraim Turban,et al.  Decision support and expert systems , 1993 .

[10]  Aalt A. Dijkhuizen,et al.  An economic comparison of management strategies on reproduction and replacement in sow herds using a dynamic probabilistic model. , 1992 .

[11]  Aalt A. Dijkhuizen,et al.  Dynamic probabilistic modelling of reproduction and replacement management in sow herds. General aspects and model description , 1992 .

[12]  A. W. Jalvingh.  The possible role of existing models in on-farm decision support in dairy cattle and swine production , 1992 .

[13]  J. Pomar,et al.  Sow Model for Decision Aid at Farm Level , 1998 .

[14]  Simon French,et al.  Applied Decision Analysis. , 1985 .