A computer program for planning dairy heifer raising

Abstracts

Alimentação e manejo inadequados das novilhas em recria constituem alguns dos entraves ao incremento da produtividade do rebanho leiteiro no Brasil. Nessa fase, o pecuarista deve decidir entre recriar os animais de reposição, adquiri-los ou terceirizar sua criação. Além disso, em se decidindo por recriar suas próprias novilhas, seria importante encontrar a melhor alternativa em termos econômicos. Visando auxiliar o pecuarista a resolver tal questão, apresenta-se um modelo matemático que permite estimar os custos decorrentes da recria de fêmeas de rebanhos leiteiros até atingir o peso recomendado à cobertura. Tal modelo foi inserido em um programa de computador de maneira a permitir a construção de diferentes cenários, considerando-se peso, raça e ganho de peso diário do lote de novilhas. Assim, após contrastar os cenários gerados com as opções de compra das novilhas ou terceirização da recria, o pecuarista aumenta as chances de tomar uma decisão mais lucrativa. Exemplos de utilização do modelo inserido em um programa de computador são apresentados e avaliados.

orçamento; recria de novilhas; programação linear

A computer program for planning dairy heifer raising¹

Moisés de Andrade Resende Filho

CNPq-Brazil scholarship holder, Assistant Professor at FEA/UFJF on leave, Ph.D. student at Department of Applied Economics, University of Minnesota, USA – [e-mail]

ABSTRACT

Inadequate feeding and management when raising heifers are two factors that constrain dairy cattle productivity in Brazil. In order to build their heads, Brazilian dairy farmers must choose between three options: to raise their own heifers to breeding age, to purchase mature heifers from outside sources, or to employ a custom heifer raiser. If the dairy farmer chooses to raise his own heifers, he needs to determine the most economically rational production strategy. To aid the farmer in this determination, a mathematical model was developed to create a budget for raising a group of heifers to their recommended breeding weight. This model was embedded in a computer program making it possible to build different growth scenarios, which can be adjusted by breed, desired daily weight gain, and average weight for a given group of heifers. After contrasting the results of these generated scenarios against the purchase cost of breeding-age heifers or the cost to employ a professional to raise the heifers, it is more likely that the dairy farmer will make the most profitable decision when it comes to increasing his herd’s size. Some examples this computer program’s use are presented and evaluated in this paper.

Key words: Budget, dairy heifer raising, linear programming.

1. Introduction

During the 1990s, Brazil’s dairy agrifood chain went through deep transformations that altered the industry’s consumption, production, industrialization, and commercialization processes. These changes can be attributed the opening of Brazil’s commercial markets, the formation of MERCOSUR (Mercado Común del Sur), and reduced State economic intervention, especially the liberation of dairy product prices at the end of 1991 and the establishment of the Real Plan (a macroeconomic plan for price stabilization) in July 1994 (Gomes, 1996 & 1997).

Gomes found that that throughout this decade of dairy industry transformation, Brazilian milk production constantly expanded. Yamaguchi et al. (1997) point out that milk production has increased because the productivity of dairy cattle has increased and that there exists potential for more productivity gains. Unfortunately, the increase in productivity has been from a very low base. Deresz (1992) asserts that one reason for the relatively low productivity of Brazil’s dairy herd is that the heifers first calve later than they should, a consequence of poor feeding and inadequate management². The author argues that the better the dairy heifers are raised the more profitable the dairy activity will be over the long and short-runs.

Yamaguchi et al. (1997) claim that each dairy farmer must answer a fundamental question: should they raise all their own heifers, purchase them from outside sources, or employ professionals to raise their heifers? To answer this question, the dairy farmer needs to determine how much it would cost to properly raise a group of heifers to breeding age, the cost to purchase breeding-age heifers in the open market, and the cost of contracting a professional to raise the farmer’s own stock. Thus, the farmer would want to examine indices that contrast the market price of heifers, the farmer’s cost to raise his own heifers, and the cost of a professional.

As a way to help farmers generate these indices and determine which method of herd building is best suited to the individual farmer, we have devised a mathematical model that allows its user to simulate a great variety of cattle raising scenarios. These scenarios are constructed to take into account heifer breed, final desired weight, desired daily weight gain, and alternative feedstuffs. [Note: feedstuffs refer to the individual ingredients in each ration.] The model then generates a budget for each scenario. In order to make that model quickly and easily available to potential users, we embedded it in a computer program. In doing so, we are assuming a constant reduction in computer and computer program prices.

The present article is organized in the following manner. Section 2 presents the theoretical and mathematical models used to build the model. The computer program’s structure is presented in Section 3. Lastly, Section 4 presents results from simulations run using the Program, which are then contrasted with technical reports from the fields of animal nutrition and dairy economics to evaluate the model’s accuracy.

2. Methodology

2.1 Theoretical Model

The following assumptions were made regarding market structure:

It is accepted that dairy farmers are price takers in the heifer market, in input markets (feedstuffs, veterinary products, labor etc.), in the market as a whole, and, specifically, in the market for breeding age heifers;

During heifer raising, from weaning to first breeding, there are a number of feedstuff combinations that, after consideration of the animal’s breed and live weight, have the potential to assure a variety of heifer daily weight gains;

The heifer's first breeding is a function of either its size or its live weight (physiologic age), and not a function of its chronological age (Campos e Lizieire, 1995).

The following technical constraints were assumed:

During the pre-puberty phase, the heifer's daily weight gain must not be greater than 900 g (Campos e Lizieire, 1995). To impose this restriction, the computer program does not accept any value inserted as live weight gain greater than 900g during this phase.

After puberty, there is no such a constraint on heifer maximum daily weight gain (Campos e Lizieire, 1995).

Furthermore, we decided to subdivide the heifer's pre-breeding period into three sub-phases. This division was made to accommodate a) the technical constraints’ characteristics and b) the fact that the equations used to predict nutritional requirements and daily ration intake are nonlinear functions of live weight and daily live weight gain for a given heifer pen.

The sub-phases are as follows:

The first sub-phase is delimited by the average initial weight of the heifers (this value should be entered into the Program by the user). The first sub-phase ends at a weight determined by the Program as follows: the heifer’s initial weight is first subtracted from its puberty weight (Table 1), that result is then divided by two, and the result of this division is added to the heifer’s initial weight to give the heifer’s final weight at the end of the first sub-phase. This calculation was implemented to subdivide the pre-puberty sub-phase into two segments, thereby reducing inaccuracies that occur when the heifers’ average weight for the whole pre-puberty period is used as input for U.S. National Research Council’s equations (NRC, 1989).

The second sub-phase is delimited by considering the calculated heifer weight at the end of the first sub-phase as its initial weight in the second sub-phase. The final weight for this sub-phase is defined by the heifer’s technically recommended weight at puberty, as shown in Table 1.

The third sub-phase is the phase from post puberty weight to breeding weight. The heifer’s assumed initial weight in this sub-phase is its technically recommended weight at puberty and the heifer’s final weight in this sub-phase is its recommended weight at first breeding (see Table 1)

2.2. Mathematical Model

In order to build a model that will budget the feed costs to raise a herd of heifers to breeding weight, a minimum cost ration model was devised for each of the previously defined sub-phases using the system of equations proposed by the NRC (1989). These equations are used to calculate a dairy heifer’s minimum nutritional requirements for Total Digestible Nutrients (TDN), Crude Protein (CP), Calcium (Ca), and Phosphorus (P) and thereby establish the lower bounds of a minimum cost ration. These requirements are determined by the heifers' characteristics, which are defined by the program user.

Although it was developed in the United States and is therefore not specifically designed for Brazilian conditions, the NRC (1989) system of equations is often used by Brazilian dairy cattle nutritionists because there is no set of similar equations designed for the Brazilian situation. A general feature of the NRC equations is that they are functions of heifer live weight and daily weight gain adjusted for each breed (see Table 1).

a) Equations to calculate TDN requirement

NEGrgt = (0.035 * LWt^0.75 )* (LWGt / 1000)^1.119 + 1 * (LWGt / 1000) * Coef. NEGrg

NEGrpt = (0.045 * LWt^0.75 )* (LWGt / 1000)^1.119 + 1 * (LWGt / 1000) * Coef. NEGrp

(1)

(2)

Firstly Coef. NEGrg = 0 and Coef. NEGrp = 0 but if small-size female then Coef. NEGrp = 1. else Coef. NEGrg =1;

NEGt = NEGrgt + NEGrpt

ME = FRG * Coef. NEGrg + FRP * Coef. NEGrp

FRG = 800 kg and FRP = 600 kg

NEMt = 0.086 * LWt ^0.75

CNEWGt = NEGt/ (LWGt / 1000)

PRVMt = LWt/ ME

(3)

(4)

(5)

(6)

(7)

If PRVMt < 0.125 then PCMEt = 2.67 else if, PRVMt > 0.125 and PRVMt < 0.75, then, PCMEt = 2.67 – (0.67 * ((PRVMt – 0.125) / (0.75 – 0.125))), else PCMEt = 2.00;

ACEt = PCMEt

CNEMt = ((1.37 * ACEt) – (0.138 * ACEt² )) + ((0.0105 * ACEt³) – 1.12)

CNEGt = ((1.42 * ACEt) – (0.174 * ACEt²)) + ((0.0122 * ACEt³) – 1.65)

DMMt = NEMt / CNEMt

DMGt = NEGt / CNEGt

DMIt = DMMt + DMGt

[ME]t = ACEt * DMIt

CEDt = (ACEt + 0.45) / 1.01

EDt = DMIt * CEDt

TDNRt = EDt / 4.409

TDNt = (TDNRt / DMIt) * 100

(8)

(9)

(10)

(11)

(12)

(13)

(14)

(15)

(16)

(17)

Where:

t defines each sub-phase from 1 to 3;

NEGrgt = required net energy gain for a large-breed female;

NEGrpt = net energy to weight gain for a small-breed female;

LWt = female live weight in kg to sub-phase t. It is calculated as the mean value of initial and final weight for each sub-phase t;

LWGt = daily live weight gain in g;

Coef. NEGrg and Coef. NEGrp, are coefficients to adjust NEG for small-breed and large-breed heifer respectively;

NEGt = required net energy for gain in Mcal;

ME = proportional ME concentration in DM to relative live weight in Mcal/kg;

FRG = maximum live weight for large-breed females in kg;

FRP = maximum live weight for small-breed female as kg;

NEMt = required net energy for maintenance in Mcal;

CNEWGt = NEMt concentration in dietary dry matter in Mcal/kg;

PRVMt = relative weight to FRG or FRP;

PCMEt = proportion of metabolizable energy in the ration’s dry matter with respect to PRVMt ;

ACEt = metabolizable energy adjustment in the ration;

CNEMt = NEM concentration in dietary in Mcal/ kg;

CNEGt = NEGt concentration in dietary in Mcal/ kg;

DMMt = dry matter intake for maintenance in kg/day;

DMGt = dry matter required for gain in kg/day;

DMIt = dry matter total required in kg/day;

[ME]t = metabolizable energy total required in Mcal.;

CEDt = digestible energy concentration of dry matter in Mcal/kg;

EDt = digestible energy total required in Mcal/day;

TDNRt = baseline total digestible nutrients for retention in kg/day;

TDNt = total digestible nutrients as % of dry matter for each sub-phase t.

b) Equations used in the crude protein requirement calculation

PDt = 0.2 * LWt^0.6

PEUt = 2.75 * LWt^0.5

PDAt = PDt / 0.67

PEUAt = PEUt / 0.67

PMFAt = 1000 * [DMIt – TDNRt * (1 – 0.08)] * 0.090

(18)

(19)

(20)

(21)

(22)

{ maintenance}

PMt = PDAt + PROTEUAt

PLRt = [211 – (26.2 * CNEGt)] * LWGt/ 1000

PRPGPt = PLRt / (LWGt / 1000)

PRAt = PLRt / 0.5

PAt = PDAt + PEUAt +PMFAt +PRAt

(23)

(24)

(25)

(26)

(27)

{Determine protein flow}

CPBt = 6.25 * [–31.86 + (26.12 * TDNRt)]

PDRt = CPBt / 0.9

CPVt = CPBt * 0.8

CPDt = CPVt * 0.8

CPIt = CPVt – DBP

PANt = CPBt – CPVt

PANDt = PANt * 1

PANIt = PANt – PANDt

PNDDt = PAt – CPDt

CPNDt = PNDDt/ 0.8

UIPDMt = CPNDt / (1000 / DMIt)

PDIt = UIPDMt – PNDDt

REPt = PDRt * (1 – 0.9)

IPRt = PDRt – PLRt

CPt = (PDRt + UIP) / (1 + 0.15)

RIPt = 0.15 * CPt

CPDt = PDRt – RIPt

DIPDMt = CPDt / (1000 * DMIt)

UIPIPt = CPNDt/ CPt

APIPt = PAt / CPt

CPDMt = CPt / (1000 * DMIt)

CPDMKt = (CPt / (1000 * DMIt)) * 100

PNDNDt = CPNDt / CPt

PDND = CPDt / CPt

(28)

(29)

(30)

(31)

(32)

(33)

(34)

(35)

(36)

(37)

(38)

(39)

(40)

(41)

(42)

(43)

(44)

(45)

(46)

(47)

(48)

(49)

(50)

(51)

{Calculation of crude protein requirements}

IP33DIP = CPNDt / 0.33

IP33UIP = CPDt / (CPDt– 0.33)

CPt = CPt + (0.5 * (IP33DIP – IP))

IPDM = CPt / (1000 * DMIt)

(52)

(53)

(54)

(55)

If IPDM < 0.12, then IPDM = 0.12 else if, IPDM > 0.16, then IPDM = 0.16, else IPDM = IPDM

CP%t = IPDM * 100

(56)

Where:

PDt = scurf protein in g/day;

PEUt = endogenous urinary protein net in g/day;

PDAt = scurf protein absorbed in g/day;

PEUAt = scurf urinary protein absorbed in g/day;

PMFAt = metabolic fecal protein absorbed in g/day;

PMt = maintenance protein absorbed in g/day;

PLRt = retained protein net in g/day;

PRPGPt = retained protein net in live weight gain as a proportion in g/day;

PRAt = retained protein absorbed in g/day;

PAt = absorbed protein total in g/day;

CPBt = bacterial crude protein in g/day;

PDRt = rumen available protein in g/day;

CPVt = bacterial true protein in g/day;

CPDt = digestible bacterial protein in g/day;

CPIt = indigestible bacterial protein in g/day;

PANt = nucleic crude protein in g/day;

PANDt = digestible nucleic protein in g/day;

PANIt = indigestible nucleic protein in g/day;

PNDDt = digestible undegraded protein in g/day;

CPNDt = digestible undegraded protein intake in g;

UIPDMt = digestible undergraded protein/dry matter proportional in g/day;

PDIt = indigestible undegraded protein in g/day;

REPt = rumen efflux protein in g/day;

IPRt = retained protein increment in g/day;

CPt = intake protein in g;

RIPt = rumen influx protein in g/day;

CPDt = degraded protein intake in g/day;

DIPDMt = degradable intake protein per dry matter proportional in g/day;

UIPIPt = undegraded protein/dry matter proportional in g/day;

APIPt = absorbable protein/ intake protein in g/day;

CPDMt = calculate CP concentration in dry matter as proportional in kg;

CPDMKt = calculate CP concentration in dry matter as proportional in kg/kg;

PNDNDt = undegraded protein required in the ration in g;

PDNDt = degraded protein required in the ration in g;

IP33DIP= calculated Intake protein by assuming 0.33 for UIPIP;

IPDM = protein intake/dry matter in kg/kg;

CP%t = crude protein as % of the dry matter to the sub-phase t.

c) Equations and conditions used in the Calcium requirement calculation

If LWt < 250 then Cat = [(8 + 0.0367 * LWt) + 0.00848 * LWGt)] / (DMIt * 10)), else if LWt > 250 and LWt < 400, then Cat = ((13.4 + 0.0184 * LWt) + 0.00361 * LWGt)) / (DMIt* 10)), else Cat = ((25.4 + 0.00092 * LWt) + 0.00361 * LWGt)) / (DMIt* 10))

where: Cat = Calcium concentration as % of dry matter in the ration to the sub-phase t.

d) Equations and conditions used in the Phosphorus requirement calculation

If LWt< 250 then Pt = [(0.884 + 0.05 * LWt) + 0.00486 * LWGt)] / (DMIt * 10)), else if LWt > 250 and LWt< 400 then Pt = [(7.27 + 0.0215 * LWt) + 0.00602 * LWGt)] / (DMIt * 10)), else if Pt = [(13.5 + 0.00207 * LWt) + 0.00829 * LWGt] / (DMIt * 10))

where: Pt = Phosphorus concentration as % of dry matter in the ration to the sub-phase t.

2.2.1. Minimum cost ration formulation model

The linear programming problem (LPP) of ration cost minimization in each sub-phase t is defined as follows:

Where:

t defines each sub-phase from 1 to 3;

CRt = minimum cost of the formulated ration to be provided to the heifers being raised to sub-phase t in a feedlot, in R$/100 kg of dry matter. It is the objective function of the LPP;

Pr_i = market price or production cost for each feedstuff to be used in formulating a minimum cost ration in R$/kg; each model is able to deal with a vector of up to 50 different types of feedstuffs;

X_it = model’s decision variables. Their values are the amount of each feed i in kg of dry matter in the ration during the t sub-phase;

DMi = dry matter for each feedstuff i as a percentage of the feedstuff’s total weight.

The constraints to the LPP in each t sub-phase are presented below:

1) TDN requirement constraint: , where TDN_i = total digestible nutrients in each feedstuff i as % ; 2) Crude protein requirement constraint: , where CPi = crude protein in each feedstuff i as %; 3) Calcium requirement constraint: , where Ca_i = calcium in a dry matter for each feedstuff i as %; 4) Phosphorus requirement constraint: , where P_i = Phosphorus in the dry matter for each feedstuff i as %; 5) ration’s total dry matter constraint . It was imposed in order to insure that the model's decision variables are calculated as percentage of dry matter; 6) The model’s decision variables must be non-negative.

Further constraints imposed to the LPP in accordance wiht Campos et al. (1995):

7) The quantity of poultry litter plus urea in the ration must not exceed 1/3 of the ration’s crude protein; 8) the roughage in the ration must be at least 60% of the ration’s total dry matter; 9) the peanut bran must not exceed 30% of the ration concentrate; 10) the rice bran must not exceed 20% of ration concentrate; 11) non-fatted bran rice must not exceed 20% of the ration concentrate; 12) ground rye must not exceed 40% of the ration concentrate; 13) bran rye must not exceed 40% of the ration concentrate; 14) ground barley must not exceed 40% of the ration concentrate; 15) bran citrus pulp must not exceed 30% of the ration concentrate; 16) poultry litter must not exceed 20% of the ration's dry matter; 17) sodium chloride must be at least 1.5% of the ration's dry matter.

2.2.2 Estimating heifer raising total feeding costs

Once a solution to the ration cost minimization LPP is obtained for each of the three heifer raising sub-phases, the computer program then uses the following equation to calculate the expected total feed cost:

Where:

CTA = total feeding cost estimate in R$;

CA = feeding cost estimate for each sub-phase t;

DMIt = dry mater intake for each sub-phase t according to equation (12);

ND_t = days in each sub-phase t, calculated by dividing the expected live weight gain for the female group in each sub-phase t by the daily weight gain defined by the user;

NA = number of heifers per pen.

3. The computer program

Since it was developed to give support in solving a specific management problem and it is a grouping of computational tools, the developed program can be defined as a Decision Support System (DSS).

In conceptual terms, a DSS is made up of the user, a computer, and the software. The software can be divided into subsystems: the model subsystem, the communication subsystem, and the database subsystem (Turban, 1993 e El-Najdawi e Stylianou, 1993). In the program presented in this article, the model subsystem is composed by the equations and models presented in Section 2.2; the communication subsystem is composed by the input insertion and output presentation screens (see Figures 1 to 4); and the database subsystem is composed by the parameters presented in Table 1 and figures derived from the nutritional composition and costs of 50 types of feedstuffs, which are typically formatted as shown in Table 2. The user may add new feedstuffs and change the previously inserted composition of feedstuffs in the ration. It is important to note that Prices 1, 2, and 3 in Table 2 are the expected costs for heifer raising during sub-phases 1, 2, and 3.

Figure 1
- Click to enlarge

Figure 2
- Click to enlarge

Figure 3
- Click to enlarge

Figure 4
- Click to enlarge

Table 2
- Click to enlarge

The program user must enter the breed and the average initial weight of the heifers' group into the Program using the screen shown in Figure 1. Extrapolating from the animal’s breed and the data presented in Table 1, the Program automatically calculates the animal’s ideal live weight at puberty and its technically recommended weight at first breeding. The user also enters the average daily heifer weigh gain desired in each sub-phase, the start date, and the number of heifers in the group (see: Figure 1). As mentioned in Section 2.1, the Program does not accept weight gain values greater than 900 g/day during the first two sub-phases.

The Program allows the user to choose between using either a single cost for the entire heifer raising period or the cost for each of the three sub-phases. Having chosen a cost period, the user enters the prices of all feedstuffs. Previously entered feedstuff prices can also be altered using this screen.

Given the command, the Program loads the inserted parameters, calculates the total feeding costs for various rations composed of different ratios of dried and natural materials, and solves the minimum cost ration formulation model for each sub-phase or the entire period. If the user wants to estimate the total cost of raising the group of heifers, rather than feed costs alone, the user can insert his forecast costs for capital depreciation, labor, machinery, etc. using the "Another costs" screen.

3.1. Some results generated by the Program

On a specific screen, the Program presents the estimated time needed for each sub-phase t to be completed. If the user has inserted all other costs, the Program will also compute total production costs and present them on this screen (see: Figure 2).

A specific screen that shows the budget for each sub-phase t and the estimated quantity of ration that needs to be formulated every day during each sub-phase can also be accessed (Figure 3). With this information, the dairyman can predict and procure the feedstuffs needed to raise the heifers.

Figure 4 shows the screen that gives the calculated minimum cost ration’s composition and the estimated cost per kg of natural matter for the first sub-phase.

The DSS also presents the nutritional characteristics of each formulated ration, i.e. Ca/P ratio and TDN as a percentage. This information allows an expert in ruminant nutrition to confirm that the formulated ration has the nutritional requirements to cause a heifer to achieve the specified daily weight gain.

4. Simulations, results, and discussion

This section presents seven simulations that replicate conditions found on some southeastern Brazilian dairy farms. The simulations use the feedstuff compositions presented in Table 3.

Table 4 presents the prices of feedstuffs used in each sub-phase. The prices are the monthly average of a historical series from January 1998 to April 2000, deflated to show August 2000 monetary values using the General Price Index calculated by Brazil’s Getúlio Vargas Foundation.

The simulations considered Holstein females weighing 80 kg on the initial date (approximate age of 90 days), 275 kg at puberty, and 340 kg at first breeding (see: Table 1). The estimated time spent raising the cattle is in agreement with the time actually spent raising Holstein heifers on dairy farms in Southeastern Brazil.

As an example, for the simulation which required a heifer daily live weight gain of 650 grams in each of the sub-phases, the Program estimated that the animal will reach breeding age in 490 days (initial age + 400 days), 16.3 months, as shown in Table 5. The feed cost in each sub-phase t and the total feed cost per heifer were calculated by the DSS according to equations (60) and (61). The estimated total feeding cost in this simulation was R$192.37 per heifer, consistent with the feeding costs reported by Yamaguchi et al. (1997) for Holstein females raised in feedlots. Furthermore, all the other simulated live weight gain strategies generated results in agreement with Yamaguchi’s report (Table 5). This live weight gain strategy can be considered a theoretical reference for Holstein heifer raising.

Table 5
- Click to enlarge

The figures in Table 5 show that the production strategy that would assure the least feeding costs is the one producing a continuous heifer weight gain of 800 g/day in the first two sub-phases and 1000 g/day in the last sub-phase. By studying Table 6 below, it can be noticed that the most intensive heifer raising strategy is not the one presenting the least cost per kg of ration. Actually, although more intensive strategies of live weight gain often generate greater costs per kg of ration, they bring the heifers to breeding weight in a shorter time; consequentially, feeding strategies that reduce the time needed to raise heifers can be worthwhile. Furthermore, if the opportunity cost of money over time is considered, raising heifers more intensively may present other favorable factors; especially, as the earlier the heifer begins its reproductive life, the faster it will start calving and producing milk.

Table 6
- Click to enlarge

As noted earlier, the information generated by DSS is available on 9 different output screens. Among other things, these screens present the minimum cost ration’s composition in natural and dry matter, the TDN and crude protein concentration of the dry matter in the minimum cost ration, the expected daily dry matter intake for each sub-phase t, the total feeding cost, the estimated quantity of ration to be formulated in kg/day, and the predicted duration in days for each sub-phase. If the user chooses to insert other costs into the Program, such as the cost of labor, calves, machinery, equipment and facilities, depreciation, and veterinary products, the Program will be able to calculate the total cost of raising a group of confined heifers to breeding age (Figure 2).

Given the presented Program’s interdisciplinary features, its potential to assist agents involved in the dairy farm decision making processes is remarkable. Of course, as users’ suggestions are incorporated, the Program will continue to improve.

References

FNP CONSULTORIA & COMÉRCIO. FNP on line. URL: (http://www.fnp.com.br/foldia/NovoMenu/index.html). 2000.

• CAMPOS, O. F. de; LIZIEIRE, R.S.; DAYRELL, M.S.; OLIVEIRA, J.S. Características e composição de alguns alimentos concentrados utilizados na alimentação de bovinos de leite Coronel Pacheco, MG: EMBRAPA, 1995. 29 p. (Circular Técnica n.38).
• CAMPOS, O.F. de; LIZIEIRE, R.S. Novilhas: elas também merecem sua atenção. Coronel Pacheco, MG: EMBRAPA, 1995. 18 p. (Circular Técnica n.36).
• DERESZ, F. Alimentação e manejo de novilhas na fase de recria. Informe Agropecuário, Belo Horizonte, v. 16, n. 175, p. 37-40, 1992.
• EL-NAJDAWI, M.K.; STYLIANOU, C.A. Expert support system: integrating AI technologies. Communications of the ACM, New York, v. 36, n. 12, p. 55-65, dec. 1993.
• GOMES, S.T. Desenvolvimento da pecuária leiteira em face das políticas governamentais. In: GOMES, S.T. A economia do leite Coronel Pacheco, MG: EMBRAPA-CNPGL, 1996.p. 69-80.
• GOMES, S.T. Transformações na cadeia do leite. Jornal da Produção de Leite, Viçosa, n. 98, abr. 1997.
• NATIONAL RESEARCH COUNCIL – NRC. Nutrient requirements of dairy cattle 5. ed. Washington, D.C.: National Academy Press, 1989. 157 p.
• TURBAN, E. Decision support and expert system: management support systems New York: Macmillan, 1993. 833 p.
• YAMAGUCHI, L.C.T.; DURÃES, M.C.; COSTA, J.L. da et al. Custos de criação de novilhas até o primeiro parto e manutenção de vacas em sistema confinado, com animais da raça holandesa. In: REUNIÃO DA SOCIEDADE BRASILEIRA DE ZOOTECNIA, 34, 1997, Juiz de Fora. Anais.. Viçosa, MG: SBZ, 1997. p. 343-345.

1

Result of the research project funded by FAPEMIG

2

The period between a female weaning up and its age to be technically considered able to start breeding.

Publication Dates

• Publication in this collection
  22 Nov 2005
• Date of issue
  2002