General introduction to Simherd III Advanced Herd Management A dynamic, mechanistic and stochastic Monte Carlo model prediction the production and states of the herd time General introduction to Method Static Dynamic Determ Stoch Stoch Optim. Simul. Prob. Rand. Simherd III Gross margin anal. X X X Partial budgeting X X X Cost-Benefit anal. X X Decision analyses X X X Linear Programming X X X Dynamic Program. X X X Østergaard et al. 2003 Markov Chain sim. X X X Monte Carlo sim. X X X General introduction to General introduction to Simherd III •What does the model do: A dynamic, mechanistic and stochastic Monte Carlo model prediction the production and states of the herd time � Simulation of technical and economical consequences of production strategies in dairy herds ’89-’91: Research tool: Jan Tind Sørensen, DIAS •Production strategie: postpone insemination start ’91-’92: Further developed as advisory tool •Technical: disease prevalence, replacement rate ’98: Simherd II, mimic four production diseases: Søren Østergaard •Economical: Margin per cow-year, income from milk sale ’98-’03: Simherd III: 8 diseases, modification feed intake, efficiency and body condition score. Østergaard, Sørensen and Hans Houe Outline of SimHerd III More projects... Many different projects and publications Herd data • Meaning of disease leves (1988) • Mortality and disease prevalence (1991) Statistical analysis • SimHerd I + feedings, reproduction and culling strategy (1992) • Increase dry period (1993) • Replacement and reproductive strategies (1995) SimHerd Simulation and • BVD (1995) SimCow HerdView presentation of annual Description of Description of • Use of Total Mixed Ration-1 (1996) results. feeding and event rates for • Grazing intensity (1997) other production the production Simulation of samples. strategy. strategy. • Organic vs. Conventional production (1998) Description and • Time of first insemination (1998) Presentation of Presentation of editing of initial herd. effects at cow short term • Preventive strategies against stafylokokmastitis (1999) level. simulations. Short term simulation • SimHerd II + Interaction between feeding, health and production (2000) of cow events. • Use of natural bull (2002) • AMS and replacement (2002) • Prolonged lactation (2003) SimTest Technical-economic comparison of production strategies • SimHerd III + control strategies against Milk Fever (2004) based on samples simulated by SimHerd. • control strategies against ParaTB (2004) • Value of progesteron measurements (2004) Overview of the Simherd-programmes and their interrelations • Early treatment of mastitis and ketose (2004) 1
Input and output Input and output Input • State of nature: • What does the model do: – Complete set of input parameters – Simherd: List of the cows’ and heifers’ ”starting values” on • Simulation of technical and economical the starting date of simulation consequences of production strategies in dairy herds • Parameter values for relations in the model – Biological parameters •Production strategie: postpone insemination start – Parameters describing production system (capacity of the •Technical: disease prevalence, replacement rate stable) – Parameters describing production strategy (feeding plan, •Economical: Margin per cow-year, income from milk sale culling decisions) Output: • Technical annual results – 10 years, 500-1000 replications Overview of steps in the simulation Overview of steps in the simulation • The herd at each week-step • Each cow at each week-step (dynamic) – Replacement (Max cow number, Culled cows, Available down calving heifers, Strategy of buying and selling – Lactation stage heifers) – Reproduction status (Heat; Pregnancy; Abortion; Calving) – Diseases, death, culling for replacement and involuntary replacement • State is updated and production and – Net energy intake = Feed intake capacity + Feed consumption are calculated available – Milk yield and weight gain = Utilized net energy - • The herd after each year of simulation Maintenance – Fetus – File annual results • Each heifer at each week-step • Ten years of simulation – Age, Reproduction, Replacement, Feeding – Analyze averages of the last 5 years Outline of SimHerd III Outline of SimHerd III Events: stochastic Mechanistic – Probability of events happening is calculated with a logistic regression model – Events on cow level determine ’behaviour’ on herd level • Conception • Culling – Replacement rate of the herd, determined by: • Disease • Culling strategy of the farmer (min. milk yield level, max. days open) • ... • (Re)production of the individual cow – Drawing samples from relevant probability distributions Functioning on cow level x number of year cows (årskøer) – Number generation by computer: X diseases does not occur ≠ Functioning on herd level 2
Outline of SimHerd III Outline of Simcow: initial cow Herd data Statistical analysis SimCow: simulates the Sim Herd production of an individual cow Sim Herd Cow View given a production strategy SimTest -initial cow -weekly results -lactation statistics -milk production Initial settings standard cow in Simcow Outline of Simcow: weekly results Outline of Simcow: lactation statistics Max. yield possible and yield currently realised Outline of Simcow: milk production Outline of SimHerd III Herd data Statistical analysis Simherd: defining initial herd Herd Sim Sim Herd and strategy Cow View SimTest -biological variables -feeding and drying-off strategy -replacement strategy -milk production Actual daily yield as function of weeks after calving 3
Simherd: defining initial herd and strategy Difference with SimFlock? Biological Variables 291 variables! Abortion (pct) 25 Time for 50% abortion 1 : 17 SimHerd Stillbirths 1. lactation (pct) 10,9 Mature weight 630 kg Stillbirth older cows (pct) 6 Involuntary culling (pct per cow-year) 18 SimFlock Mortality in cows (pct per cow-year) 2 Age at first heat (days) 280 Marture weight mean 1637 gram St.dev. first heat (1,2 or 3) 2 standard deviation 143 gram Change of pregnancy (pct) 50 Gestation length (days) 280 Prop. Heifer calves (pct) 50 Yield level 34,0 (kg ECM 2 1-24 wpp, 3 rd lact) St. dev. Individual yield level 2 St. dev. Individual yield level 2 Mature weight (kg) 630 Simherd: defining initial herd and strategy Simherd: defining initial herd and strategy Feeding and drying of strategy Reproduction strategy yield at 1st step., 1st lact. (kg/day)) 23,0 291 variables! Start breed of heifers (days) yield at 1st step., older cows (kg/day) 28,0 450 Start breed of 1 st lactation (days after calving) yield at 2nd step., 1st lact. (kg/day) 18,0 40 yield at 2nd step., older cows (kg/day) 22,0 of which Start breed of other cows (days after calving) 40 yield at drying off 1st lact. (kg/day) 5,0 Heat detection eff. in winter, heifers (pct) 50 yield at drying off older cows (kg/day) 5,0 Heat detection eff. in summer, heifers (pct) 40 70 decision Heat detection eff. in winter, cows (pct) 50 variables! Heat detection eff. in summer, cows (pct) 50 Pregnancy detection cows (days after insemination) 42 Pregnancy detection cows (days after insemination) 42 First calendar day in summer season, cows 119 Last calendar day in summer season, cows 238 First calendar day in summer season, heifers 119 Last calendar day in summer season, heifers 238 Decision variables concerning health Simherd: defining initial herd and strategy Modeling risk of disease at the cow level Replacement strategy 291 variables! •Base risk (probability for parity 3, average producing, no previous cases) Max number of cows 120 Min number of cows 100 •Parity: 1, 2, 3, >3 Max days open before culling 1 st lact 427 of which •Milk yield capacity Max days open before culling older 427 Max age of open heifers before culling 805 •Lactational recurrence of the disease Max days open bef. cull. decision 1 st lact 245 70 decision • Other diseases Max days open bef. cull. decision older 245 variables! •Body condition score Max days open bef. cull. (low yield) 1 st 203 Max days open bef. cull. (low yield) old 203 •Season Threshold low yield 1 st lactation (kg 24) 25 Threshold low yield 2 nd lactation (kg 24) 33 Threshold low yield older cows (kg 24) 34 Logistic regression model and random numbers Milk quota: max tons per year 0 Min tons per year 0 4
Recommend
More recommend
