sow replacement: a comparison of farmers' actual decisions and model recommendations

Livestock Production Science, 23 (1989) 207-218 207 Elsevier Science Publishers B.V., Amsterdam - - Printed in The Netherlands

Sow Replacement: A Comparison of Farmers' Actual Decis ions and Model Recommendat ions

A.A. DIJKHUIZEN, R.M.M. KRABBENBORG and R.B.M. HUIRNE

Department o/Farm Management, Wagenlngen Agricultural University, Hollandseweg 1, 6706 KN Wageningen (The Netherlands)

(Accepted for publication 9 March 1989)

ABSTRACT

Dijkhuizen, A.A., Krabbenborg, R.M.M. and Huirne, R.B.M., 1989. Sow replacement: a comparison of farmers' actual decisions and model recommendations. Livest. Prod. Sci., 23: 207-218.

Sow culling data were analyzed from 1617 culled sows on 12 Dutch farms, using the VAMPP herd health and management information system. The annual culling rate was found to be 50%, more than half of which was due to reproductive problems and low productivity. The risk of removal increased considerably with age, except for second parity sows.

The data were economically evaluated, using the PorkCHOP replacement model. The calculated losses caused by premature disposal turned out to average Dfl. 124 per culled sow, or Dfl. 9300 per year in a 150-sow operation, which equals about 16% of pig farm income. The highest economic losses were found in cases of"lameness/leg weakness", averaging Dfl. 183 per culled sow.

In comparing the model recommendations and the actual decisions to cull sows because of low productivity there was a striking resemblance. Culling the elder, poorly producing sows directly after weaning was the right decision economically. With respect to reproductive problems, however, sows showing no heat were culled too early after weaning and sows that failed to conceive too late. Moreover, the time of culling hardly differed with the sow's age and productive performances. Further research is desired to determine whether computerized decision support can help the farmers to reduce the income losses when sows fail to conceive.

INTRODUCTION

In The Netherlands about 50% of the sows are replaced annually (Te Brake, 1986). The average productive lifespan of the sows, therefore, is about two years or four litters only.

From a farm management point of view, the total annual culling rate can be considered as the combined result of unplanned removals on the one hand (e.g. in the case of health problems and lameness), and more economically based decisions on the other. The latter applies especially to sows with insufficient productive and reproductive performances, and accounts for more than half of

0301-6226/89/$03.50 © 1989 Elsevier Science Publishers B.V.

2 0 8 A.A. DIJKHUIZEN ET AL.

the annual cullings (Dagorn and Aumaitre, 1978; Kroes and van Male, 1979; van der Steen, 1984; Singh, 1986).

In previous research (Dijkhuizen et al., 1986) an economic replacement model named PorkCHOP, was designed to optimize the replacement decision for sows with poor productive and/or reproductive performances. The model has been constructed as a template for an electronic spreadsheet program, which can be run on a microcomputer. It runs with default values, but allows the user to enter data for all the variables considered, and thus can easily be adjusted to individual farm conditions world-wide.

In the literature no attention has yet been paid to whether there is any resemblance between farmers' actual sow culling decisions and economically op- t imum decisions provided by model calculations, as was done before in dairy cattle (Dijkhuizen, 1983). Such an analysis is especially of interest to gain more insight into the underlying - economic - framework of the farmers' decision-making process. Tha t may help to get a better idea of the potential ben- efits of computerized support, and can also contribute to a further improve- ment of the theoretical concept in modelling the economics of sow replacement decisions.

In this paper, results of such a combined analysis of field data and economic modelling are presented and discussed. Therewith, special attention will also be paid to the - parity-specific - extent and reasons of sow removal.

MATERIAL AND METHODS

Description of the farms involved

Standard Dutch record-keeping systems for swine breeding herds do not provide individual sow performances and removal data, as are required for the analysis in this paper. Access was available, however, to the database of the State Veterinary Faculty, which includes farms using the VAMPP herd health and management information system (Buurman et al., 1987). In total 12 farms had accurate and complete data from 1617 sows, culled during the years 1985 and 1986. To gain more insight into the level of management on these farms, some main herd characteristics are presented in Table 1 and compared with typical commercial swine breeding herds in The Netherlands.

As shown in Table 1, the 12 farms involved in the analysis are bigger, and raise more pigs per sow per year as a result of more litters per sow per year and a lower pig mortality. The annual sow replacement rate, however, is somewhat less favourable. Altogether, these farms are to be characterized as performing "slightly above average".

sow REPLACEMENT: ACTUAL AND MODEL RECOMMENDATIONS 209

TABLE 1

Herd characteristics of the 12 VAMPP farms involved in the analysis, compared to commercial Dutch farms using standard record-keeping systems

12 VAMPP farms Typical farms 1 1985-1986 1985-1986

Number of sows f a rm- 1 145 Litters sow- 1 year - 1 2.22 Pigs born alive l i t ter-1 10.3 Percentage pig mortality 12.0 Pigs weaned lit ter-1 9.1 Pigs weaned sow- ~ year - 1 20.2 Annual sow replacement rate 50

126 2.13

10.4 14.5 8.9

18.9 47

1Almost 2000 commercial swine breeding herds (Arkes et al., 1987).

The logic of the PorkCHOP model

The underlying economic criterion for culling decisions does not differ from that described before in dairy cattle (Zeddies, 1973; Dijkhuizen et al., 1985; van Arendonk, 1985; Kristensen, 1987): a sow of a particular age should be kept in the herd as long as her expected profit for the next parity (marginal profit) is higher than the per parity lifetime average return from a replacement gilt (average profit). This income potential of the gilt cannot be realized as long as the sow is retained, and so can be interpreted as the opportunity cost of postponed replacement.

In determining the optimum lifespan for individual sows, the model starts by calculating this opportunity cost. This calculation is based on the average performance in the various parities of sows present in the herd, assuming this to be the best estimate for the expected future performance of gilts. Future revenues and costs are discounted to make profits occurring at different points of time completely comparable. In averaging these future profits, each is weighed by a probability of sow survival, calculated from the current parity-specific risk of removal in the herd.

Subsequently, the model predicts the expected marginal profits in future parities of the sows currently present in the herd, taking into account variation in expected litter size due to differences in the number of pigs born alive. The repeatability of litter size in individual sows is commonly accepted to be low, but significantly above zero (van der Steen, 1984). In predicting the percentage deviation of the typical parity-specific litter size, information is used as follows from up to three previous parities of the sow:

dpj =0.20Xdpj_ 1 +0.15 ×dpj_2 +O.05 x dpj_3

210 A.A. DIJKHUIZEN ET AL.

where dp = percentage deviation of the typical parity-specific number of pigs born alive (dpj_i=O if ( j - i ) < 0 ) , j = p a r i t y number (2 to 10).

The economic opt imum lifespan then - excluding the risk of removal - is the last parity with a positive difference between expected marginal profit of the present sow and average profit of the replacement gilt.

After a sow's opt imum lifespan has been determined in this way, the model calculates the total extra profit to be expected from trying to keep her until that opt imum and not replacing her immediately, taking into account the risk of premature removal of retained sows. This total extra profit - called retention pay-off (RPO) - is an economic index, which makes it possible to rank sows within a herd on future profitability. The higher the RPO, the more valuable the sow. A value below zero means that replacement is the most profitable choice. It also represents the economic loss when a sow has to be removed involuntarily at normal slaughter value.

The RPO index is calculated in the model at the average time after weaning when sows conceive in the herd. An extra delay in conception, therefore, will decrease the RPO, since the sow concerned will have more open - unproductive - days. By relating the initial value of the RPO index and the calculated financial loss of a delay in conception the model gives an indication of the maximum allowable number of breedings in case of failure to conceive.

A more detailed and mathematical description of the model has been described elsewhere (Dijkhuizen et al., 1986).

R E S U L T S

Farmers' actual culling decisions

Reasons/or culling As mentioned before, a total of 1617 sows were culled on the 12 farms in-

volved in 1985 and 1986, representing an average annual replacement rate of about 50% (Table 1 ). Some average data per reason are presented in Table 2.

As shown in Table 2, reproductive problems are the major reason of removal, with failure to conceive as its major sub-reason. Low productivity is the second most important reason, representing 17% of the total. Together, insufficient productive and reproductive performances account for more than half of the annual culling rate, which closely agrees with Kroes and van Male (1979) and van der Steen (1984).

The average parity at culling for all sows turned out to be 4.3, as shown in Table 2. Taking into account the average number of 2.22 litters per sow per year, this (also) leads to a productive lifespan of about 4.3/2.22= 2 years, or an annual removal rate of about 100/2 = 50%. Sows culled for low productivity (i.e. small litters and old age) realize the highest productive lifespan, whereas culling for "no oestrus" and "lameness/leg weakness" occurs especially in the

SOW REPLACEMENT: ACTUAL AND MODEL RECOMMENDATIONS 211

TABLE 2

Culling data of all 1617 culled sows

Reasons of culling % of total Average Moment of Relative culled parity culling ~ production sows at culling level (%)2

Reproductive problems 34.2 3.4 72 100 Failure to conceive 25.2 3.7 85 99 No oestrus 9.0 2.5 35 101

Low productivity 17.2 7.3 10 86 Old age 11.0 8.5 10 94 Small litters 6.2 5.1 11 71

Sickness/accidents 16.0 3.8 44 96 Mother characteristics 13.9 4.3 13 88 Lameness/leg weakness 10.5 2.9 21 98 Miscellaneous 8.2 4.0 29 94

Average of total 100.0 4.3 40 94

1Days after weaning. 2Based on pigs born alive from up to three previous litters, and weighed according to the formula presented in the section "The logic of the PorkCHOP model".

younger sows, resulting in an average productive lifespan of 2.5 and 2.9 parities, respectively.

The time of culling after weaning varies according to the reason: only 10 days for "low productivity" and 13 days for "mother characteristics" to 85 days for "failure to conceive". The overall average turned out to be 40 days.

The relative production level of the culled sows turned out to be below average, at a level of 94%. The lowest level (71%) was found for sows culled owing to "small litters", as could be expected. None of the reasons especially affected the high-producing sows, as shown in Table 2.

Parity-specific risk of removal The parity-specific (or marginal) risk of removal can be defined as the per-

centage chance for a particular sow of being culled from a certain age category. This chance is to be determined in different ways, as shown in Table 3. Method I relates the number of culled sows and the number of farrowed sows in each of the parities considered. Method II derives the chances from the age structure of the sows currently present in the herd, while Method III is based on the age structure of the culled sows. In a stable situation concerning herd size and culling pat tern all three methods provide the same results.

As shown in Table 3, the outcome of the methods considered does not differ widely, indicating a rather stable situation for the data set as a whole. The calculated risk of removal increases considerably with age, as could be expected. It is interesting to note the exception in the case of second-parity sows.

212

TABLE 3

Parity-specific risk of removal

A.A. D I J K H U I Z E N E T AL.

Parity Number Number Risk of removal ~ n of sows of sows

farrowed removed I II III 1 2 3 4 5 6

1 1698 353 0.21 0.18 0.22 2 1385 218 0.16 0.11 0.17 3 1235 220 0.18 0.23 0.21 4 948 179 0.19 0.24 0.22 5 724 138 0.19 0.18 0.21 6 591 123 0.21 0.31 0.24 7 407 111 0.27 0.31 0.29 8 282 96 0.34 0.37 0.35 9 179 67 0.37 - 0.37

10+ 253 112 0.44 - 1.00

Total 7702 1617

10

1col. 4n=col . 3. /col . 2.; col. 5~=(col . 2~-co l . 2 .+1)/col . 2n; col. 6n=(col . 3 J ~ col. 3n) / n = l

n-- 1 10

( 1 - • col. 3 . / 2 col. 3.) . n ~ O n = l

The relative risk of removal across parities varies per reason, as shown in Fig. 1. In the lower parities reproductive problems account for almost half of the culling rate, while culling for low productivity hardly occurs. The relative importance of culling for lameness/leg weakness decreases with age, as is the case for reproductive problems. "Sickness/accidents" and "miscellaneous" have the most stable relative importance across parities, as should be expected.

Model recommendations

Simulation of the individual farm situation From an economic point of view, culling decisions can only be affected by

those revenues and costs that differ between the various parities of sows. The PorkCHOP model, therefore, takes into account parity-specific values for sow removal probabilities, farrowing efficiency, piglet production, feed consumption and slaughter value of the sows. Moreover, the cost of replacement gilts and a discount rate are included in the calculations (Dijkhuizen et al., 1986).

Most of these factors are directly available in the VAMPP record-keeping system of the farms involved. As shown in Table 4, the current set of input values makes it possible to closely imitate the actual farm performances. Prices, as indicated, had to be gathered from common statistics and entered separately

SOW R E P L A C E M E N T : ACTUAL AND MODEL R E C O M M E N D A T I O N S

100

80

213

F - ' - ] Hi see I I aneous

Lameness/leg weskn.

Hother character ist.

Sickness/ acc i dent s

Low product ivi ty

~'~-'~ Reprod. problees

6O

Q-

tO

0 1 2 3 4<5 6+7

Parity Fig. 1. Relative importance of culling reasons across parities.

into the model. Feed consumption and liveweight of culled sows were set to default values, representing typical Dutch conditions (Dijkhuizen, 1989).

RPO index for the individual culled sows To evaluate the farmers' actual decisions, the PorkCHOP replacement model

was used to calculate the RPO index for all the culled sows, as mentioned before. Results are presented in Table 5.

As shown in Table 5, there is a striking resemblance between actual and recommended decisions for sows culled because of low productivity. The average RPO index was Dfl. 21 only, while 70% Of these sows had a value below zero, indicating replacement to be the most profitable choice. For all other reasons the calculated average RPO index was much higher, as could be expected. The highest economic values, and thus the highest economic losses in case of removal, occur in cases of " lameness/ leg weakness", averaging Dfl. 183 per culled sow. The overall losses were found to be Dfl. 124 per culled sow.

214

TABLE 4

Simulated and actual values of some major herd characteristics

A.A. DIJKHUIZEN ET AL.

Simulated Actual farm values values

Pigs born alive litter-1 10.3 10.3 Percentage pig mortality 12.4 12.0 Pigs weaned lit ter- 1 9.1 9.1 Litters sow- 1 year- 1 2.22 2.22 Pigs weaned sow- 1 year- 1 20.1 20.2 Annual sow replacement rate 52 50 Selling price pigs at 23 kg (Dfl) 113 1 Cost of replacement gilts at age 250 days (Dfl) 520 - Average slaughter value culled sows (Dfl) 479 - Annual feed consumption of sows (kg) 1114 -

1Not available in the VAMPP recording system.

TABLE 5

The RPO index for all the culled sows, calculated at the time of weaning

Reasons of culling RPO index (Dfl)

Average % > 0

Reproductive problems 161 88 Failure to conceive 153 87 No oestrus 185 92

Low productivity 21 29 Old age 14 24 Small litter 32 37

Sickness/accidents 147 87 Mother characteristics 92 69 Lameness/leg weakness 183 90 Miscellaneous 117 79

Average of total 124 74

Most sows were removed fairly shortly after weaning, except those culled because of reproductive problems (Table 2). In cases of "failure to conceive" the interval between weaning and culling was almost three months. The longer it takes to get a sow pregnant, the more urgent becomes the question: is the loss less when the farrowing interval is longer or when the sow is culled? By relating the initial value of the RPO index at the time of weaning and the calculated financial loss of a delay in conception the PorkCHOP model gives

SOW REPLACEMENT: ACTUAL AND MODEL RECOMMENDATIONS

TABLE 6

Actual and allowable number of days between first and last breeding

215

Failure to conceive No oestrus 1

% of Actual Allowable % of Actual Allowable total days days total days days

Par i ty 1 25 64 55 45 24 54 2-3 30 73 51 34 20 47 > 4 45 74 18 21 14 20

Total 100 71 37 100 21 45 Rel. prod.level (%)

<90 31 65 17 29 18 18 90-110 38 75 36 37 21 39 > 110 31 74 60 34 23 74

Total 100 71 37 100 21 45

INo breeding was carried out. The - theoretical - interval between first and last breeding, therefore, was derived from the interval normal first breeding (12 days after weaning) till removal.

an indication of the maximum allowable number of days for (re-)breeding in case of failure to conceive. Results are presented in Table 6.

As shown in Table 6, the actual interval between first and last breeding for sows culled due to failure to conceive averages 71 days (or slightly more than three re-breedings), and hardly depends on parity number and productive quality of the sows concerned. The calculated allowable interval, however, averages 37 days only (or less than two re-breedings), and differs considerably across parity number and productive quality of the sows, as could be expected. Apparently, farmers pay too little attention to age and productivity when de- ciding whether to re-breed or to replace these sows. The same applies to sows which show no heat. The calculated allowable interval exceeds the actual interval, however, especially for the younger and high-producing sows.

DISCUSSION AND CONCLUSIONS

Economic modelling of health and fertility problems in pigs has received much less attention than in dairy cattle, although the potential economic impact would justify the opposite situation (Dijkhuizen, 1989). A major reason for this inconsistency may be the serious lack of accurate epidemiological data of the various health and fertility problems in pigs, including their impact on productive performances. During the last few years much effort has been put into designing and implementing integrated veterinary, zootechnical and economic record-keeping systems (Stein et al., 1983; Buurman et al., 1987). In


the future, systematic epidemiological and economic analyses of these data bases should be given high priority. Knowledge obtained from this kind of research is necessary in order to increase the realism of economic modelling, while model calculations can quantify the significance of remaining gaps in veterinary and zootechnical knowledge.

In this paper a first a t tempt at such an analysis was made, focused on sow removal. A detailed - parity-specific - insight into the extent and reasons of removal could be provided. Entering these data into the PorkCHOP replacement model, developed in previous research (Dijkhuizen et al., 1986), makes it possible to quantify the economic losses of premature disposal. The highest economic losses were found in cases of "lameness/leg weakness", averaging Dfl. 183 per culled sow. The overall losses turned out to be Dfl. 124 per culled sow. Taking into account an annual replacement rate of 50%, this implies an average loss of Dfl. 62 per sow present in the herd, or Dfl. 9300 in a 150-sow operation. This is about 3% of the gross production value or 16% of income for a typical farmer (Dijkhuizen, 1989). Improving animal health and fertility, therefore, can play a major economic role in this context.

In comparing the model recommendations and the actual decisions to cull sows because of low productivity (i.e. small litters and old age) there was a striking resemblance, as indicated by the average economic value (RPO index ) of only Dfl. 21 per culled sow. Culling the elder, poorly producing sows directly after weaning was economically the right decision. Further research is of interest to investigate whether replacement of retained sows would also have been the best decision. Additional calculations showed that, from an economic point of view, sows culled because of old age could equally well have been replaced one parity earlier, this did not apply to sows culled because of small litters.

The most important reasons for culling sows were reproductive problems: no heat and failure to conceive. In general, it turned out that sows showing no heat were culled too early, compared with model recommendations. In the model calculations, no repeatability of these problems across parities was taken into account. When the problems have to be expected in future parities as well, the allowable number of rebreedings is decreased considerably (Dijkhuizen, 1989). Sows that failed to conceive, however, were culled too late, even when no problems in future parities of the sows were assumed (repeatability = 0). Moreover, the interval between first and last breeding scarcely differs with age and productive performance of these sows. It should be considered whether computerized decision support can help the farmers to reduce the income losses when sows fail to conceive. Further research is underway to widen the (theoretical) base for such a support (Huirne et al., 1988).

ACKNOWLEDGEMENT

The authors are indebted to J.C.M. Vernooy (Department of Preventive Veterinary Medicine and Reproduction from the State Veterinary Faculty in

SOW REPLACEMENT: ACTUAL AND MODEL RECOMMENDATIONS 217

U t r e c h t ) fo r p r o v i d i n g t h e n e c e s s a r y d a t a o u t o f t h e V A M P P h e r d h e a l t h a n d m a n a g e m e n t i n f o r m a t i o n sy s t e m .

REFERENCES

Arkes, J.G., Baltussen, W.H.M., Ogink, G.J.A. and Schneider, J., 1987. Results on Dutch swine breeding herds. Reports, Swine Extension Service and the Institute of Agric. Economics (in Dutch).

Buurman, J., Van Leengoed, L.A.M.G., Wierda, A., Vernooy, J.C.M. and Van der Valk, P.C., 1987. VAMPP: a Veterinary Automated Management and Production Control Program for swine breeding herds. I. Main features. Vet. Q., 9: 15-23.

Dagorn, J. and Aumaitre, A., 1978. Sow culling: reasons for and effect on productivity. Livest. Prod. Sci., 6: 167-177.

Dijkhuizen, A.A., 1983. Economic aspects of diseases and disease control in dairy cattle. Ph.D. Thesis University of Utrecht, 193 pp (in Dutch, English summary).

Dijkhuizen, A.A., 1989. Economic aspects of common health and fertility problems for the individual pig producer: an overview. Vet. Q., 11: 116-124.

Dijkhuizen, A.A., Renkema, J.A. and Stelwagen, J., 1985. Economic aspects of reproductive failure in dairy cattle. II. The decision to replace animals. Prev. Vet. Med., 3: 265-276.

Dijkhuizen, A.A., Morris, R.S. and Morrow, M., 1986. Economic optimization of culling strategies in swine breeding herds, using the "PorkCHOP computer program". Prev. Vet. Med., 4: 341- 353.

Huirne, R.B.M., Hendriks, Th.H.B., Dijkhuizen, A.A. and Giesen, G.W.J., 1988. The economic optimization of sow replacement decisions by stochastic dynamic programming. J. Agric. Econ., 39: 426-438.

Kristensen, A.R., 1987. Optimal replacement and ranking of dairy cows determined by a hierarchic Markov process. Livest. Prod. Sci., 16: 131-144.

Kroes, Y. and van Male, J.P., 1979. Reproductive lifetime of sows in relation to economy of production. Livest. Prod. Sci., 6: 179-183.

Singh, D., 1986. Simulation of swine herd population dynamics. Agric. Syst., 22: 157-183. Stein, T.E., Morris, R.S. and Williamson, N.B., 1983. The computer as a core of a planned ap-

proach to health care for food animals. Proc. Symp. Comp. Appl. Vet. Med., 207-222. Te Brake, J.H.A., 1986. Culling of sows and the profitability of piglet production. Neth. J. Agric.

Sci., 34: 427--435. Van Arendonk, J.A.M., 1985. Studies on the replacement policies in dairy cattle. Ph.D. Thesis,

Agricultural University, Wageningen, 96 pp. Van der Steen, H.A.M., 1984. Prediction of future value of sow productivity. 35th Annual Meeting

of the European Association of Animal Production, The Hague, The Netherlands. Zeddies, J., 1973. Zur Bestimmung der optimal wirtschaftlichen Nutzungsdauer yon landwirt-

schaftlichen Nutztieren. Agrarwirtschaft, 22: 7-16.

RESUME

Dijkhuizen, A.A., Krabbenborg, R.M.M. et Huirne, R.B.M., 1989. Remplacement des truies: une comparaison des ddcisions effectives de l'dleveur avec les recommandations du module. Livest. Prod. Sci., 23:207-218 (en anglais).

Des donndes provenant de 1617 truies rdformdes dans 12 dlevages hollandais qui utilisaient le


syst~me d'information VAMPP sur l'dtat sanitaire et la conduite ont dtA analysdes. Le taux de rdforme annuel ~tait de 50%, plus de la moiti~ des animaux dtant dliminds pour des probl~mes de reproduction et une productivit~ faible. Le risque de rdforme augmentait consid6rablement avec l'hge, sauf dans le cas des truies qui se trouvaient en deuxi~me portde.

L'~valuation 4conomique des donn~es a dtd rdalis~e grace au module de remplacement Pork- CHOP. Les pertes calculdes dues h une dlimination pr~maturde s'dlevaient en moyenne h 124 Dfl. par truie rdformde, ou 9300 Dfl. par an pour une opdration portant sur 150 truies, ce qui dquivaut

environ 16% du revenu de l'~levage porcin. Les pertes dconomiques les plus importantes inter- venaient dans le cas de "boiterie/faiblesse des pattes" oh elles s'dlevaient en moyenne h 183 Dfl. par truie rdformde.

La comparaison des recommandations du module et des d~cisions effectives de rdforme des truies pour faible productivitd conduit h des ressemblances tr~s nettes. L'~limination aussitSt apr~s le sevrage des truies les plus figdes qui ~taient faibles productrices ~tait la bonne solution au plan dconomique. Cependant, en ce qui concerne les probl~mes de reproduction, les truies ne prd- sentant pas d'oestrus dtaient rdform~es trop t6t et celles qui n'dtaient pas fdconddes ~taient ~li- min6es trop tard. De plus, le moment de la rdforme ne diffdrait gu~re suivant l'fige et les performances de prodt~ction de la truie. D'autres recherches sont ndcessaires pour ddterminer si une d~cision bas~e sur le cdlcul peut aider les dleveurs h r~duire les pertes de revenu lors.que les truies ne sont pas f~conddes.

KURZFASSUNG

Dijkhuizen, A.A., Krabbenborg, R.M.M. and Huirne, R.B.M., 1989. Sauenremontierung: Ein Ver- gleich aktueller Entscheidungen der Bauern mit Modellempfehlungen. Livest. Prod. Sci., 23: 207-218 (auf englisch).

Mit Hilfe des VAMPP-Gesundheits- und Herdenmanagement-Informationssystems wurden 1617 abgegangene Sauen aus 12 niederl~indischen Betrieben untersucht. Die jiihrliche Merzungs- rate betrug 50%, davon fiber die H~ilfte wegen Reproduktionsproblemen und niedriger Leistung. Das Ausfallrisiko erh~hte sich stark mit zunehmendem Alter, auger fiir Sauen zum zweiten Wurf.

Die ~konomische Bewertung der Daten erfolgte mit dem PorkCHOP-Remontierungspro- gramm. Die fiir vorzeitige Merzung berechneten Verluste betrugen Dfl. 124 je gemerzte Sau oder Dfl. 9300 pro Jahr ftir einen 150-Sauenbetrieb, was etwa 16% des Betriebseinkommens aus Schweinen ausmacht. Die gr6f~ten wirtschaftlichen Verluste von Dfl. 183 pro gemerzte Sau wurden f'~r die Ursache "Lahmheit/Beinschwiiche" gefunden.

Oberraschende ~hnlichkeit ergab sich zwischen den Modell-Empfehlungen und den ta t~ch- lichen Merzungsentscheidungen wegen geringer Produktivith't. Die Merzung der Sauen direkt nach dem Absetzen war fiir iiltere Sauen mit niedriger Leistung die 5konomisch riehtige Entsch- eidung. Hinsichtlich der Reproduktionsprobleme jedoch, wurden nicht rauschende Sauen zu frfih nach dem Absetzen gemerzt und umrauschende zu sp~it. Darfiber hinaus schwankte die Merzung- szeit wenig mit dem Alter und der Produktivit~t der Sauen. Weitere Forschung ist n~tig, um zu bestimmen, ob Entscheidungen aus Computerprogrammen den Landwirten helfen k6nnen, die Einkommensverluste durch umrauschende Sauen zu reduzieren.

sow replacement: a comparison of farmers' actual decisions and model recommendations

Documents