Technical-economic optimization in the fattening of turkeys from commercial genetic lines in Temascaltepec, Mexico

Velázquez-Villalva, Héctor Hugo; Posadas-Domínguez, Rodolfo Rogelio; Rebollar-Rebollar, Samuel; Bobadilla-Soto, Encarnación Ernesto; Velázquez-Villalva, Héctor Hugo; Posadas-Domínguez, Rodolfo Rogelio; Rebollar-Rebollar, Samuel; Bobadilla-Soto, Encarnación Ernesto

doi:10.22319/rmcp.v16i3.6876

Services on Demand

Journal

Article

Indicators

Cited by SciELO
Access statistics

Revista mexicana de ciencias pecuarias

On-line version ISSN 2448-6698Print version ISSN 2007-1124

Rev. mex. de cienc. pecuarias vol.16 n.3 Mérida Jul./Sep. 2025 Epub Dec 08, 2025

https://doi.org/10.22319/rmcp.v16i3.6876

Articles

Technical-economic optimization in the fattening of turkeys from commercial genetic lines in Temascaltepec, Mexico

Héctor Hugo Velázquez-Villalva^a

Rodolfo Rogelio Posadas-Domínguez^b^*

Samuel Rebollar-Rebollar^a

Encarnación Ernesto Bobadilla-Soto^c

^{^a} Universidad Autónoma del Estado de México. Centro Universitario UAEM Temascaltepec. Facultad de Ingeniero Agrónomo Zootecnista. Carretera Toluca-Temascaltepec. Colonia Barrio de Santiago. 51300, Estado de México, México.

^{^b} Universidad Autónoma del Estado de Hidalgo. Escuela Superior de Zimapán. Facultad de Contaduría. Hidalgo, México.

^{^c} Universidad Michoacana de San Nicolás de Hidalgo-CONAHCYT. Instituto de Investigaciones Agropecuarias y Forestales. Michoacán, México.

Abstract

The objective of the study was to develop a combined approach of deterministic modeling and Monte Carlo simulation to establish the optimal technical-economic level for fattening commercial genetic lines of turkeys in the municipality of Temascaltepec, Mexico. Data from 42 turkeys of the double-breasted diamond genetic variety, with an average age of three weeks and an initial body weight of 1.077 ± 0.132 kg, were analyzed. In order to model the relationship between feed consumption and final weight of the turkeys, a quadratic production function was fitted using nonlinear regression along with a Monte Carlo simulation analysis to evaluate the risk associated with fluctuations in feed costs and the selling price of turkeys. The deterministic analysis identified that economic profitability is maximized when turkeys consume 26.81 kg of feed and reach a live weight of 11.30 kg. The Monte Carlo simulation confirmed the validity of these results under scenarios of uncertainty in sales prices and feed costs, estimating an average net profit of $379.29 per bird in the economic optimum, compared to $325.39 in the technical optimum, representing a 16.56 % increase in profitability per turkey. These findings confirm that integrating deterministic modeling with stochastic simulation constitutes an effective methodological tool for production decision-making under conditions of variability and uncertainty. It is recommended that this approach be validated in other agricultural contexts to strengthen its practical applicability.

Keywords Economic analysis; Feed consumption; Nonlinear regression; Optimal economic level; Stochastic simulation

Resumen

El objetivo del estudio fue desarrollar un enfoque combinado de modelado determinista y simulación Monte Carlo para determinar el nivel óptimo técnico-económico en la engorda de guajolotes de líneas genéticas comerciales en el municipio de Temascaltepec, México. Se analizaron datos de 42 pavos de la variedad genética diamante doble pechuga, con una edad promedio de tres semanas y un peso corporal inicial de 1.077 ± 0.132 kg. Para modelar la relación entre el consumo de alimento y el peso final de los guajolotes, se empleó una función de producción cuadrática ajustada mediante regresión no lineal, complementada con un análisis de simulación Monte Carlo para evaluar el riesgo asociado con las fluctuaciones en el costo del alimento y el precio de venta de los guajolotes. El análisis determinista identificó que la rentabilidad económica se maximiza cuando los guajolotes consumen 26.81 kg de alimento y alcanzan un peso vivo de 11.30 kg. La simulación Monte Carlo confirmó la validez de estos resultados bajo escenarios de incertidumbre en los precios de venta y el costo del alimento, estimando una utilidad neta promedio de $379.29 por ave en el óptimo económico, frente a $325.39 en el técnico, lo que representa un incremento del 16.56 % en la rentabilidad por guajolote. Estos hallazgos confirman que la integración del modelado determinista con simulación estocástica constituye una herramienta metodológica eficaz para la toma de decisiones productivas en condiciones de variabilidad e incertidumbre. Se recomienda validar este enfoque en otros contextos agropecuarios para fortalecer su aplicabilidad práctica.

Palabras clave Consumo de alimento; Nivel óptimo económico; Simulación estocástica; Análisis económico; Regresión no lineal

Introduction

In Mexico, turkey (Meleagris gallopavo) breeding with commercial genetic lines (meleagriculture) is considered one of the most traditional livestock activities¹. This productive system is developed throughout the national territory, mainly concentrated in indigenous communities², where small production units operate and implement animal husbandry practices adapted to the farmers' conditions³. Turkey production in Mexico is carried out under three well-defined systems: technified, semi-technified, and small producers who raise turkeys in backyard settings in a traditional way⁴, each system with different characteristics and technological levels⁵^,⁶.

Since domestication, several varieties or genetic lines of the wild turkey have been developed, the most commercially important of which include the large white turkey and the double-breasted bronze⁷. These commercial lines are raised mainly in intensive systems and are selected for their rapid growth, feed efficiency, carcass yield and larger breast size⁸. Despite the diversity of turkey production systems, poultry farming in Mexico faces significant challenges to meet the growing domestic demand for turkey meat, which has led to a considerable increase in imports. In 2017, the country ranked as the third largest importer of turkey meat with 7.2 million tons of frozen meat and 85 thousand tons of fresh chilled meat⁹.

A determining factor for improving turkey production is technical and economic efficiency and the optimal use of productive factors (labor, capital, and raw materials) ¹, especially feed, which represents one of the most relevant inputs in this production system ¹^,¹⁰. Animal growth in this context can be accurately modeled by mathematical functions describing the evolution of live weight, allowing the estimation of maximum growth values and optimal ages to maximize productivity and economic benefit ¹¹^,¹². To evaluate the technical efficiency in production Bates and Ocelli ¹³ propose two approaches; the deterministic and the probabilistic. This study focuses on the probabilistic model by applying a quadratic production function with a nonlinear regression approach, complemented with a Monte Carlo simulation analysis.

Although some previous studies in the Mexican agricultural sector have implemented production functions to evaluate the impact of feed on animal weight gain, the results of these studies are not yet available¹⁴^-¹⁷, there is a lack of work focused on meleagriculture with commercial genetic lines to determine the optimal slaughter ages to maximize economic profitability in this production system, particularly considering that feed represents more than 70 % of the total costs in turkey production¹⁰^,¹⁸^,¹⁹. Therefore, the objective of the present study was to develop a combined approach of deterministic modeling and Monte Carlo simulation to determine the optimal technical-economic level in the fattening of commercial genetic lines of turkeys, incorporating dietary variables and the impact of the cost of the diets used in the municipality of Temascaltepec, Mexico.

Material and methods

Study area

The experiment was conducted at the poultry fattening farm of the zootechnical post at the UAEM Temascaltepec University Center. The farm is located between parallels 18°49'58.9'‘ and 19°08’ 37.0'‘ N, and between meridians 99°51'17.9’' and 100°36.0' 46.2'' W. The climate of the zone is semi-warm sub-humid with summer rains. The average annual temperature is 20.3 °C, with precipitation of 1,158 mm²⁰.

Animal handling and facility preparation

The fattening house used in the experiment has dimensions of 9 m in length, 6 m in width, and 3.5 m in height. Individual corrals made of 3/8" rods covered with hexagonal mesh were installed inside, having a length of 100 cm, a width of 50 cm, and a height of 70 cm. These pens were designed to prevent contact between turkeys from different flocks and to ensure adequate ventilation during the fattening cycle. Prior to the start of the experiment, the facilities underwent a cleaning process using soap and water, followed by disinfection with a creoline solution (13.33 ml/L) 48 h before the arrival of the animals.

Each pen was equipped with a lighting system using 40-watt fluorescent bulbs, a 30 cm long feed through with a capacity of 1.2 kg of feed, and a 40-watt fluorescent bulb system. To ensure adequate ventilation control, adjustable plastic curtains were installed throughout the house. During the first four days of the experiment, the curtains remained completely closed to avoid drafts. From the fifth day onwards, progressive adjustments were started, lowering them by 25 cm each day, until, from the seventh day onwards, they remained open between 8:00 and 20:00.

Reception and handling of turkeys

Prior to the arrival of the turkeys, a biosecurity protocol was implemented by placing a sanitary mat at the entrance of the shed, which was soaked with a solution of 2 L of water and 5 ml of creoline to prevent the introduction of external pathogens.

Biological material and feeding

The study used 42 turkeys of the double-breasted diamond genetic variety, with an average age of three weeks and an initial body weight of 1.077 ± 0.132 kg. The turkeys were randomly distributed, one turkey assigned to each pen, and were managed under an ad libitum feeding regime (Figure 1). During the first two hours after their arrival, they received no feed so that they could adapt. Subsequently, the amount of feed to be provided was calculated based on the heaviest turkey, adding 10 % of its body weight to ensure adequate feed intake.

Figure 1 Traditional feeding system in the broiler poultry farm of the zootechnical post of the UAEM Temascaltepec University Center

In this study, three nutritional phases were employed: starting turkeys, growing turkeys, and finishing turkeys, according to the nutritional plan of the commercial feed used in the research, which integrates the growth and development phases into a single stage referred to as growing turkeys. During the initial phase, which lasted from the arrival of the turkeys until d 23, they were fed a diet containing 27 % crude protein. The transition to the growth diet was carried out gradually over 3 d, with decreasing proportions of the initial diet (75-25 %, 50-50 %, and 25-75 %). The turkeys consumed this growth diet between d 27 and 55 (Table 1). Subsequently, they transitioned to the finishing diet, which they were given until the end of the experiment (84 d).

Table 1 Nutritional composition of the diets fed to turkeys at different growth phases (%)

Phases	Crude protein	Crude fat	Crude fiber	Ashes	Moisture	NFE
Starting	27.00	3.50	4.50	7.50	12.00	45. 50
Growing	23.00	4.50	5.00	7.00	12.00	48.50
Finishing	20.00	4.50	5.00	6.00	12.00	52.50

NFE= nitrogen free extract by difference.

Source: Productores Agropecuarios Tepexpan, S.A. de C.V, 2023.

Response variables

The initial live weight of the turkeys was recorded individually at the time of reception, using a scale with a capacity of 40 kg. Live weight was monitored every 8 d, according to the procedure described by Rodriguez-Licea et al¹, who demonstrated that this interval is adequate to capture significant changes in the growth of turkeys under similar conditions. Feed was offered twice daily, at 0900 and 1800 h. At the latter time, the rejected feed was collected from each feeder, stored in bags identified by lot, and weighed to record waste before the next feeding. During the first day after their arrival, the turkeys were given water with a solution of RuViOtic (0.6 g/L) to ensure hydration; thereafter, from the second day until the end of the experiment, water was provided ad libitum.

The daily feed consumption was calculated as the total feed offered minus the feed rejected each day. Based on this information, weekly consumption was estimated by adding the daily values for each turkey. The weekly weight gain was determined by subtracting the final weight recorded in each week from the initial weight at the corresponding phase (start, growth, and finish). Feed conversion was estimated by dividing the total feed consumption by the final weight minus the initial weight of the turkeys. Data were recorded in a field logbook during the entire experimental period. The information was then organized in a Microsoft Excel spreadsheet, which served as the basis for statistical processing and subsequent modeling.

Statistical model

To evaluate the relationship between feed intake and final weight of the turkeys, a production function represented by a nonlinear quadratic regression model was used. In this model, the total feed consumption (A) at each stage of production was used as the independent variable, while the live weight (P) expressed in kilograms was considered as the dependent variable. This nonlinear approach was chosen due to the biological nature of animal growth, where the decreasing marginal return on feed intake has been widely documented in the literature¹⁶^,¹⁷^,²¹. The regression model was determined by evaluating both the statistical significance and the consistency in the biological and economic hypotheses, considering the magnitude and sign of the coefficients in each parameter¹². The model specification was as follows:

P=α+β1A+β2A2+ε

Where P= live weight of each turkey in kilograms; α= intercept of the function; β1 and β2= are the estimated coefficients that capture the linear and quadratic effect of feed consumption; A= is the amount of feed supplied to each turkey (kg); ε= random error.

Analysis of technical and economic efficiency

The technical-economic efficiency was estimated using the methods proposed by various authors¹^,¹²^,¹⁴. These authors calculate the optimum technical level (OTL) from the first partial derivative of the production function with respect to the independent variable “A”. The equalization of the first derivative to zero gives the amount of feed with which the rate of weight gain reaches its maximum value. The value derived from “A” is then substituted into the production function to determine the weight with the maximum biological yield.

To determine the optimum economic level (OEL), the first partial derivative of the production function is used and equals the ratio between the price of the input (feed), divided by the selling price of the live turkey. This approach allows the integration of economic criteria into the model and, by deriving it, the amount of feed and live weight that maximizes economic profitability is obtained¹⁶. During the study period, the cost per kilogram of feed was $11.63, while the selling price per kilogram of live weight was $73 per kilogram.

Monte Carlo simulation

The OTL and OEL were determined through a traditional optimization analysis, which is based on calculations that assume constant conditions. However, to address the uncertainty in the selling price and feed cost, a Monte Carlo simulation analysis was performed; this approach allows modeling the uncertainty of these variables, generating a distribution of potential results that complement the traditional analysis. The Monte Carlo simulation provided a detailed view of the impact of fluctuations in the price of feed and sale of the turkeys on the monetary benefits associated with optimal levels, allowing a more robust assessment of risk and economic gains.

To perform the Monte Carlo simulation analysis, the parameters associated with the price per kilogram of feed and the sales price of turkey were adjusted using 15 yr (2008-2022) of historical data. The historical sales price of turkey was obtained from the statistical portal of the Agrifood and Fisheries Information System (Sistema de Información Agroalimentaria y Pesquera)²², while the historical price per kilogram of feed was determined using an inflation-adjusted methodology, as there is no source of accurate historical price data in Mexico. The adjustment for inflation was made using annual inflation rates consulted on Banco de México's website²³ with a base price of $11.63 per kg of feed for the year 2023. The following formula was used for this adjustment:

adjusted P=current P1+inflation

_ajustedP is the adjusted price of the previous year, _currentP is the base price for the year 2023 ($11.63). This adjustment enabled the creation of a time series of historical feed prices (2008-2022), adjusted to constant values of the base year, ensuring a consistent comparison throughout the study period.

A Monte Carlo simulation was performed with 10,000 iterations to assess the variability in turkey sales prices and feed costs. The initial parameters for the simulation were defined based on historical data on the price per kilo of turkey feed and sales. The feed cost was modeled with a mean = $8.18 and variance of $1.46, while the sales price of turkey included a mean of $57.11 and standard deviation = 14.01 pesos.

For each iteration, total revenue and total cost were calculated at the OTL and the OEL. The total revenue was obtained by multiplying the sales price by the weight of the turkey, while the total cost was calculated by multiplying the feed cost by the consumption. The net profit was determined by subtracting the total cost from the revenue. To ensure the accuracy and robustness of the simulation, a convergence and cross-validation test was performed.

The convergence assessment was carried out through an analysis with different numbers of iterations (1,000, 5,000, and 10,000) to obtain the mean, standard deviation and the 5, 50 and 95 % percentiles of the net gains at each optimal level. Cross validation was performed using five folds to check the consistency of the results. This process involved splitting the simulations into different subsets and combining the results to check whether the estimates in net gain for each optimum were consistent with the data partitions. The optimization analysis and Monte Carlo simulation were carried out using the R software (version 4.4.1)²⁴.

Results

Table 2 shows the weight gain of the turkeys from the third week of age, of 1.08 kg, until wk 12, when they reached an average weight of 12.45 kg. The total feed consumption per turkey was 34.74 kg during the finishing phase, with an FC of 3.05 kg of feed per kg of weight gained.

Table 2 Productive variables in the fattening of double-breasted diamond turkey poults

Variable	Values
Number of turkeys	42.00
Average initial weight per turkey, kg	1.08
Average final weight per turkey, kg	12.45
Total average feed consumption per turkey, kg	34.74
Average feed conversion per turkey	3.05

The F value of the model (1,167.97; P<0.001), indicates that the predictor variables (A and A²) have a significant impact on the weight gain of the turkeys. The coefficient of determination indicates that 99.57 % of the variability in weight is explained by the predictors. The predicted R² (0.9876) confirms that the model has high predictive power and can be applied to practical scenarios under various dietary conditions. The positive coefficient of “A” shows that for each additional kilogram of feed consumed, the average weight gain of the turkeys increases by 0.58 kg, while the negative coefficient of “A²” indicates the presence of decreasing marginal yields. This behavior shows that the increase in feed consumption was not proportional to the weight of the turkeys (Table 3).

Table 3 Regression model results for predicting weight gain in turkey poults

	Estimated parameters	Standard error	t statistic	P -value
Intercept	1.41419	0.15220	9.29	<.0001
Variable A	0.57806	0.02512	23.01	<.0001
Variable A²	-0.00781	0.00074	-10.55	<.0001
R² of the model	0.99570
Predicted R² of the model	0.98769
F of the model	1,167.97			<.0001

Optimal technical level in the production of turkey poults

The OTL was obtained by equalizing the first derivative of the production function to zero, which corresponds to the point where the turkeys reach their maximum biological production or maximum in vivo weight¹²^,²¹.

For the estimated production function: P1.41419+0.57806A-0.00781A², by deriving P with respect to A and equating it to zero, was obtained the following result:

dydx=0.57806-0.01562A=0

Solving for A generates the value of the variable input, which represents the total amount of feed that the turkeys need to reach their maximum production weight or maximum in vivo weight.

-0.01562A=-0.57806

A=-0.57806-0.01562

A= 37.01 kilograms of feed

Turkeys require 37.01 kg of feed to reach their maximum production weight. Substituting this value of A in the production function gives the maximum live weight of the turkeys.

P1.41419+0.57806(37.01)-0.00781(37.01)2= 12.11 kilograms of weight.

Optimal economic level in the production of turkey poults

The OEL was determined by equating the first derivative of the production function to the ratio between the price of one kilogram of feed (Px) and the selling price of one kilogram of turkey (Py). This analysis establishes the optimal amount of feed to be consumed by the turkeys to maximize economic benefit¹²^,²¹.

dPdA=PxPy=0.57806-0.01562A=11.6373

Solving for A:

0.57806-0.01562A=0.15936

-0.01562A=-0.4187

A=-0.4187-0.01562=26.81 kg of feed

Turkeys require 26.81 kg of feed to reach the maximum economic benefit. Substituting this value of A in the production function gives the live weight that maximizes income.

P1.41419+0.57806(26.81)-0.00781(26.81)2= 11.30 kilograms of live weight.

Table 4 shows that the OEL generates a higher profit per turkey ($513.10) than the OTL ($413.60). The reduction in feed consumption for the OEL generates a reduction in the total cost, which translates into an increase in the benefit/cost ratio for each peso invested.

Table 4: Comparison of monetary gains at the optimal technical and the economic level

Parameter	Optimum technical level	Optimal economic level
Feed intake, kg	37.01	26.81
Weight of turkey, kg	12.11	11.30
Total income, $	844.03	824.90
Total cost, $	430.43	311.80
Profit, $	413.60	513.10
Benefit/cost ratio, $	1.96	2.65

Monte Carlo simulation

The Monte Carlo simulation results indicate that the average net gain at the OEL was higher than that at the OTL. As the number of simulations increased, it was observed that both the mean gain and the values at the 5th and 95th percentiles stabilized after 1,000 interactions. Cross-validation confirmed the precision of these estimates with values very close to those obtained by the convergence method. The standard deviation for both optima remained constant, indicating a stable variability in net gain regardless of sample size. The behavior of the percentiles indicated that the central values remained consistent, which reinforces the stability of the estimates and the robustness of the results. However, the extremes of the distribution revealed greater variability in average gain for the OEL, which may be associated with higher risk under adverse conditions compared to the OTL (Table 5).

Table 5 Evaluation of net profit with Monte Carlo simulation for the optimal technical and economic level

	Optimum technical level			Optimal economical level
Number of simulations	Profit ($)	Standard deviation ($)	Percentiles	Profit ($)	Standard deviation ($)	Percentiles
1,000	334.14	182.79	5%: 35.86 50%:338.04 95%: 621.55	387.73	164.98	5%: 119.88 50%: 393.90 95%: 656.86
5,000	330.4	182.75	5%: 32.85 50%:326.64 95%: 629.40	383.82	165.98	5%: 112.41 50%:383.33 95%: 657.83
10,000	325.39	181.51	5%: 30.40 50%:324.09 95%: 624.83	379.29	164.95	5%:111.49, 50%: 378.40 95%: 652.50
Parameter:
Cross-validation	329.25	180.88	5%: 30.32 50%:328.59 95%: 626.42	383.01	164.15	5%: 109.13, 50%: 381.75 95%: 651.61

Discussion

The final weight of the turkeys at 12 wk of life in this study (12.45 kg) was similar to that reported by Santos-Ricalde et al²⁵ in turkeys of the Convertible genetic line. Similar results have been reported in Nigeria, where turkeys fed commercial diets showed weight gains comparable to those in this study²⁶. The feed conversion observed in this study was 3.05 kg, higher than the 2.46 kg recorded for turkeys of the Nicholas 700 line in a similar fattening period²⁷. This difference can be attributed to variations in diet and management used in each investigation, factors that directly impact feed efficiency. On the other hand, a study conducted in Nigeria reported a higher feed conversion ratio of 3.53 kg²⁶, which highlights the influence of management and feeding conditions on the productive performance of turkey poults.

The average final weight of the turkeys in this study (12.45 kg, after 12 wk of fattening) was higher than that reported in backyard and semi-technified production systems, where the average weights at 15 wk were 3.19 and 4.76 kg, respectively¹. The standard error analysis confirmed the validity of this behavior, showing that all assumptions associated with the distribution of variables A and A² are satisfied¹². This ensures that the estimates of the effects of the variables on the production of turkey poults are reliable and not susceptible to variations in the sample data.

The t statistic associated with variables A and A² showed statistical significance in the regression model, confirming their significant influence on turkey production¹. In addition, the value of R² in the model was consistent with that reported by Rodríguez-Licea et a¹, who found that the feed input explained 99.55, 98.76, and 99.34 % of the weight attained by turkeys in technified, semi-technified, and backyard systems, respectively. These results align with other studies, which highlight feed input as a critical factor in turkey production²⁶ and reinforce the importance of diet in feed conversion and growth ²⁷.

According to microeconomic theory²⁸, the negative sign of parameter A² in the production function indicates the presence of a quadratic model with diminishing marginal returns. This behavior implies that as the feed input is progressively increased, the increase in the weight of the turkeys begins to decrease¹. Consequently, there is a decrease in the economic gain at the time of sale.

Maximum live weight production in turkeys does not imply maximum monetary gain¹. This is explained by the fact that, at high levels of feed intake, a proportion of the weight gained is transformed into fat instead of meat, reducing the technical efficiency and increasing production costs¹²^,¹⁴^,²⁹. This study confirmed that the monetary gain was higher in the OEL, a performance that can be attributed, among other variables, to the inverse relationship between the economic optimum and the price of the feed¹⁶, which highlights the importance of optimizing feed costs to maximize the profitability of the production system.

The results of the Monte Carlo simulation reveal clear differences in the net gains obtained for the optimal levels, with OEL showing higher gains in all the simulations performed. This behavior highlights the importance of optimizing production costs to maximize economic benefits. Furthermore, the stabilization of the average earnings and the 5th and 95th percentiles after 1,000 iterations validates the sufficiency of the model in obtaining reliable estimates, which agrees with previous studies that have demonstrated the effectiveness of the Monte Carlo method in estimating economic variables under uncertainty³⁰^,³¹.

A relevant result is the lower standard deviation observed in the OEL compared to the OTL, which suggests greater stability of economic gains when costs are optimized. However, the extreme percentiles (5th and 95th percentiles) show greater variability in the gains associated with OEL, which could imply a higher risk under adverse conditions. This performance is relevant for small producers, who may face greater difficulties in scenarios of economic uncertainty or fluctuations in the market price of feed inputs.

Cross-validation reinforces the robustness of the results obtained by demonstrating consistency with the values derived by the convergence method. This finding indicates that the model used is reliable for projecting economic scenarios in turkey production under conditions similar to those of this research. The stability observed in the central percentiles for both optimum levels is an indicator that the estimates obtained are not only accurate but also highly reproducible across different simulated scenarios.

Conclusions and implications

The deterministic analysis identified that the economic profitability is maximized when turkeys consume 26.81 kg of feed and reach a live weight of 11.30 kg. The Monte Carlo simulation confirmed the validity of these results under scenarios of uncertainty in sales prices and feed cost, estimating an average net profit of $379.29 per bird in the economic optimum, compared to $325.39 in the technical optimum, which represents a 16.56 % increase in profitability per turkey. These findings confirm that integrating deterministic modeling with stochastic simulation constitutes an effective methodological tool for making production decisions under conditions of variability and uncertainty. It is recommended that this approach be validated in other agricultural contexts to strengthen its practical applicability.

Acknowledgments

The authors are grateful to the Centro Universitario UAEM Temascaltepec for facilitating the use of their zootechnical post’s poultry fattening farm and to the students of the School of Agricultural Engineering for participating in the project.

REFERENCES

1. Rodríguez-Licea G, Carrillo-Juárez C, Hernández-Martínez J, Borja BM. Análisis diferencial técnico-económico de los sistemas productivos de guajolotes en el Estado de México. Ciencia Ergo Sum 2017;24(1):25-47. [ Links ]

2. Estrada-Mora A, Alcántara-Carbajal JL, Cadena-Iñiguez J, Tarango-Arámbula LA, Segura-León O, Escalante-Pliego P. La crianza del guajolote (Meleagris gallopavo) en comunidades indígenas de la región centro de México. Agroproductividad 2013;6(6):59-68. [ Links ]

3. Camacho MA, Torres IL, Ramírez CL, López PR, Arcos GL. La avicultura de traspatio en la Costa de Oaxaca, México. Ciencia y Mar 2006;IX(28):3-11. [ Links ]

4. Jerez MP, Herrera JG, Vásquez MA. La gallina criolla en los Valles Centrales de Oaxaca. ITAO-CIGA, Oaxaca, México. 1994. [ Links ]

5. Santos-Ricalde R, Segura-Correa, J. La importancia de la cría de pavos (Meleagricultura) en Yucatán. Bioagrociencias 2020;13(1):1-8. [ Links ]

6. Pérez RG, Rodríguez GB. Huexolotl: pasado y presente en México. Universidad Nacional Autónoma de México, Instituto de Investigaciones Antropológicas. 2020. [ Links ]

7. Cantaro H, Sánchez J, Sepúlveda P. Cría y engorde de pavos. Instituto Nacional de Tecnología Agropecuaria, Argentina. 2010. [ Links ]

8. Segura-Correa JC, Santos-Ricalde R, Sarmiento-Franco L. Origen y genética de las razas comerciales del pavo en México. Bioagrociencias 2022;15(2):84-90. [ Links ]

9. SIAVI. Sistema de Información Comercial Vía Internet. Importaciones de pavos. México. 2019. http://siavi.economia.gob.mx/ . Consultado 26 Ago, 2024. [ Links ]

10. Montoya A, Caicedo S, Montoya IA. Análisis de las oportunidades de aumento de consumo de carne de pavo (Meleagris gallopavo) en Colombia. Suma de Negocios 2015;(6):183-193. [ Links ]

11. Agudelo GD, Cerón MM, Restrepo LL. Modelación de funciones de crecimiento aplicadas a la producción animal. Rev Colom Cienc Pecu 2008;21(1):39-58. [ Links ]

12. Rebollar RS, Hernández J, Rojo R, González F, Mejía D, Cardoso J. Óptimos económicos en corderos Pelibuey engordados en corral. Universidad y Ciencia 2008;24(1):67-73. [ Links ]

13. Battese GE, Coelli T. A model for technical inefficiency effects in a stochastic frontier production function for panel data, Empirical Economics 1995;20:235-332. [ Links ]

14. Rebollar-Rebollar S, Posadas-Domínguez RR, Hernández-Martínez J, Rojo-Rubio R, González-Razo FJ, Guzmán-Soria E. Óptimo técnico y económico en bovinos productores de carne engordados en corral. Trop Subtrop Agroecosystems 2011;14(2):413-420. [ Links ]

15. Bautista MY, Espinosa GJA, Herrera HJG, Martínez CFE, Vaquera HH, Estrada DB et al. Óptimos técnicos para la producción de leche y carne en el sistema bovino de doble propósito del trópico mexicano. Rev Mex Cienc Pecu 2019;10(4):933-950. [ Links ]

16. Ramos-Canché ME, Magaña-Magaña MÁ, Aguilar-Urquizo E, Pech-Zapata A, Piñeiro-Vázquez ÁT, Toledo-López VM, et al. Óptimos económicos en la cría del cerdo pelón mexicano: propuesta de integración para cadena productiva. Ecosistemas Recursos Agropecu 2020;7(1):1-10. [ Links ]

17. Velázquez-Xochimil HG, Portillo-Vázquez M. Determinación del óptimo técnico y económico en maíz (Zea mays L.) modalidad temporal del Estado de México. Agroproductividad 2018;11(1):15-20. [ Links ]

18. Dorantes-Coronado EJ, Posadas-Domínguez RR, Rojo-Rubio R, Mendoza-Méndez RV, López-Benítez H. Evaluación financiera en la engorda rural de guajolote blanco doble pechuga. Invest Cienc Universidad Autónoma de Aguascalientes 2025;33(94):e4969. [ Links ]

19. Arieta-Román RJ, Graillet-Juárez EM, Alvarado-Gómez LC, Martínez-Martínez M, Gómez-Alor JG. Evaluación de ganancia de peso en dos líneas de pollos (Gallus gallus domesticus L.) bajo un manejo en casetas con ambiente natural. Agroproductividad 2018;11(6):79-84. [ Links ]

20. INEGI. Instituto Nacional de Estadística y Geografía. Consulta de indicadores sociodemográficos y económicos por área geográfica. https://www.inegi.org.mx/ . Consultado 15 Sep, 2024. [ Links ]

21. Villavicencio-Gutiérrez MR, Martínez-Castañeda FE, Rogers-Montoya NA, Martínez-Campos AR, Gómez-Tenorio G, Velazquez L, et al. Environmental impacts of medium-scale pig farming at technical and economic optimum production weight in Mexico. Sci Total Environ 2024;(946):174240. [ Links ]

22. SIAP. Servicio de Información Agroalimentaria y Pesquera. Anuario estadístico de la producción ganadera. https://nube.siap.gob.mx/cierre_pecuario/ . Consultado 26 Nov, 2024. [ Links ]

23. Banco de México. Portal de inflación. https://www.banxico.org.mx/tipcamb/main.do?page=inf&idioma=sp . Consultado 28 Nov, 2024. [ Links ]

24. R Core Team. R: A language and environment for statistical computing. Version 4.4.1. R Foundation for Statistical Computing, Vienna, Austria. 2024. [ Links ]

25. Santos-Ricalde R, Segura-Correa J, Palma-Ávila I, Sandoval-Castro C, Sarmiento-Franco L. Evaluación de dos programas de alimentación sobre el comportamiento productivo y lesiones en patas de pavos comerciales. Nova Scientia 2017;9(2):37-50. [ Links ]

26. Makinde OJ, Inuwa M. The use of agro industrial by-products in the diet of grower turkeys. Trop Subtrop Agroecosystems 2015;18(3):371-378. [ Links ]

27. Valarezo-Ulloa M, Valarezo-García J, Vacacela-Ajila W, Ortega RR. Evaluación productiva y económica del engorde de pavos de la estirpe Nicholas 700. Rev Dirección de Investigación 2017;6(1):91-99. [ Links ]

28. Leroy MR, Meiners RE. Microeconomía. Tercera ed., Estado de México, México: Mc Graw Hill; 1990. [ Links ]

29. Rebollar-Rebollar S, Gómez-Tenorio G, Callejas-Juárez N, Guzmán-Soria E, Hernández-Martínez J. Óptimos técnicos y económicos en cortes de carne de cerdo en dos regiones de México. Agronom Mesoamericana 2014;25(1):161-168. [ Links ]

30. Tayyab M, Malik AI, Khan I, Ullah M. A simulation-optimization approach for economic assessment of an imperfect serial production management under uncertainty. J Industrial Prod Engineering 2024;41(7):618-635. [ Links ]

31. Amorim FR, Guimarães CC, Afonso P, Tobias MSG. Forecasting cost risks of corn and soybean crops through Monte Carlo Simulation. Appl Sci 2024;14(17):8030. [ Links ]

Received: March 04, 2025; Accepted: June 10, 2025

^{Conflict of interest}

The authors of this paper declare that they have no conflict of interest.

Este es un artículo publicado en acceso abierto bajo una licencia Creative Commons