Analysis of the efficiency of production factors in shallot farming using True Seed of Shallot (TSS) in Adipala District, Cilacap Regency

 Corresponding author: Shafiq Arfianto Copyright © 2025 Author(s) retain the copyright of this article. This article is published under the terms of the Creative Commons Attribution License 4.0. Analysis of the efficiency of production factors in shallot farming using True Seed of Shallot (TSS) in Adipala District, Cilacap Regency Shafiq Arfianto *, Edy Prasetyo and Migie Handayani Agribusiness, Faculty of Animal and Agricultural Sciences, Diponegoro University, Indonesia. World Journal of Advanced Research and Reviews, 2025, 26(02), 3819-3834 Publication history: Received on 20 April 2025; revised on 25 May 2025; accepted on 27 May 2025 Article DOI: https://doi.org/10.30574/wjarr.2025.26.2.2077 Abstract Red onion is one of the horticultural commodities with significant economic value. Cilacap Regency is a major red onionproducing area in Central Java, one of which is Adipala District. In 2023, the red onion production in Cilacap Regency reached 951.12 tons with a productivity of 8.38 tons/ha. The productivity of Cilacap Regency is still low compared to Central Java Province, which is 10.23 tons/ha. The productivity of red onions in Adipala Subdistrict is 9.16 tons/ha. This study was conducted using the census method. Primary data collection was carried out through interviews with red onion farmers as respondents using a questionnaire. The study was conducted in Adipala District with 65 red onion farmers as respondents. The primary data collected included: respondent characteristics, land area, seeds, labor, NPK fertilizer, organic fertilizer, dolomite, pesticides, age, education level, experience, extension services, and land status. Secondary data were obtained from the Ministry of Agriculture, the Central Statistics Agency, and the Provincial/District Agriculture Office. Data were analyzed using the Frontier 4.1 application, followed by allocative and economic efficiency analysis using the Cobb-Douglas production function dual cost equation. The results showed that land, seeds, labor, NPK fertilizer, and dolomite had a positive effect on production, meaning they increased production, while organic fertilizer and pesticides did not have a significant effect. The analysis indicates that the average technical and allocative efficiency of red onion farmers in TSS is sufficient, but economic efficiency is still lacking. The average technical, allocative, and economic efficiencies are 0.86, 0.70, and 0.60, respectively. Efforts to reduce inefficiency include optimizing extension services. To obtain more diverse data, further comprehensive research is needed on the efficiency of red onion farming during the same planting season. Keywords: Horticulture; Onion; Production; Allocative Efficiency; Cobb-Douglas 1. Introduction Shallots are a high economic value horticultural commodity in Indonesia. Regulation of the Minister of Agriculture Number 46 of 2019 establishes shallots as a national strategic commodity with the consideration that the commodity can affect the value of inflation, is needed in large quantities, cannot be substituted with other horticultural commodities and its production involves many farmers with a large development area. Domestic shallot production over the past three decades has shown positive growth, in line with exports experiencing quite high growth, as well as significant growth in imports. Based on FAO (Food and Agriculture Organization) data, in 2017-2021 Indonesia in the ASEAN scope became the fourth-ranked exporting country exporting shallots with an average of 6.53 million USD as well as the second-ranked importing country importing shallots with an average of 51.76 million USD (Center for Agricultural Data and Information Systems, 2023). The national shallot production centers in 2018-2022 are spread across ten provinces. The three provinces with contributions above 10% to national production are Central Java, East Java, and West Nusa Tenggara, each with World Journal of Advanced Research and Reviews, 2025, 26(02), 3819-3834 3820 contributions of 29.93% (average production of 531.88 thousand tons), 24.86% (average production of 441.78 thousand tons), and 11.41% (average production of 202.73 thousand tons). Central Java is the highest production center province in Indonesia. Two districts contribute 78.12% to shallot production in Central Java, namely Brebes with a contribution of 68.94% (production of 383.68 thousand tons) and Demak with a contribution of 9.18% (production of 51.08 thousand tons). While other districts contributed a total of 21.88%. (Center for Agricultural Data and Information System, 2023). Cilacap Regency is one of the potential shallot development areas in Central Java Province. In 2023 Cilacap Regency had a planting area of 120.58 hectares, a harvest area of 113.53 hectares, a production of 951.12 tons and an average productivity of 8.38 tons/hectare (Central Bureau of Statistics, 2023). The potential for shallot development in Cilacap Regency is supported by agro-climatic conditions, namely climate, temperature and irradiation that are suitable for shallot cultivation. Shallot cultivation in Cilacap District develops on lowland/coastal land with sandy soil texture, among others in Adipala Sub-district, Nusawungu Sub-district and surrounding areas. Sandy soil is quite potential for farming because it has a crumbly soil structure, medium texture, and good drainage and aeration that support shallot cultivation, In addition, sandy land is relatively safer from disease (Iriani, 2013). Adipala Sub-district is one of the shallot production centers in Cilacap Regency. Initially, shallot farmers in Adipala subdistrict practiced bulb-based shallot farming. Starting in 2021, shallot farmers in Adipala Sub-district have been introduced to shallot farming from seed (True Seed of Shallot/TSS). The government continues to promote the use of TSS as an alternative seed to replace bulb seeds whose prices fluctuate as well as an effort to increase production and farmers' income. Previous research (Monica et al, 2021) which conducted research in 2020 stated that shallot bulb farming in Adipala District was not technically efficient. Onion farming in Adipala District can still be increased in productivity through the efficient use of production factors, as stated by Nurjati et. Al. (2018) which states that the strategy that can be applied to increase competitiveness through optimizing shallot production inputs is through the use of botanical seed technology/true seed of shallot. Kusnadi et al. (2011) also mentioned that increasing productivity through the application of technical efficiency is important because it can be used as an effort to increase production in addition to agricultural extensification, considering the availability of agricultural land is decreasing along with the conversion of agricultural land and various other causes. In addition to increasing efficiency in terms of inputs, it should be noted that the social conditions of farmers such as age, education level, experience in farming and frequency in seeding create variations in allocating factors of production between one farmer and another. Various studies on shallot production efficiency have been conducted. Among them is the research of Nurjati et al. (2018) which revealed that shallot farmers in Pati Regency are technically efficient, but not economically and allocatively efficient. Similar findings were also revealed by Mustiarasari et al. (2019) who stated that the average shallot farmer in Majalengka Regency is technically efficient. This is in line with Ismiasih et al. (2024) who analyzed farm businesses based on 11,206 shallot farmers collected by BPS in the 2013 Agricultural Census stated that the technical efficiency level of shallot farming in Indonesia was quite efficient. In contrast to previous studies, this study analyzes the production efficiency of TSS shallot farming in Adipala Subdistrict, which is one of the prospective areas for shallot development in Cilacap Regency. 2. Material and methods The research was conducted in October-November 2024. The timing of the study considered when farmers finished harvesting shallots in the third planting season from June to August 2024. The research location was taken purposively, namely in Adipala Sub-district, Cilacap Regency. The reason for choosing the location is because Cilacap Regency is one of the districts in Central Java that is prospective for the development of shallot farming and Adipala Sub-district is the center of shallot production in Cilacap Regency where many farmers have begun to apply TSS shallot cultivation. The number of research samples was 65 farmers. Because the number of research samples was less than 100, the census method was used. Census research involves collecting data from all respondents to provide a comprehensive picture of the condition of an area. This type of research is quantitative descriptive research. Quantitative method is a method whose research data is in the form of numbers and analysis using statistics. The data collected in this study were analyzed using Frontier 4.1 software and the results of the analysis were presented descriptively quantitatively. The Stochastic Frontier method is one of the methods used in estimating the production frontier and also measuring the level of production efficiency. According to Gujarati (2009), the Cobb-Douglas production function is one example of a log linear multiple regression model, a multiple linear regression model is a linear regression model with more than one explanatory variable. The formula is as follows: World Journal of Advanced Research and Reviews, 2025, 26(02), 3819-3834 3821 Y = aX(1) (b1), X(2) (b2), X(3) (b3), X(4) (b4), X(5) (b5), X(6) (b6), X(7) (b7) ……………….(3. )1 In the production function, the factors that are thought to affect production are land, seeds, NPK fertilizer, organic fertilizer, soil conditioner, pesticides and labor. To determine the efficiency of the use of production factors on the production of TSS system farms, frontier analysis is used, or the stochastic production frontier method. The stochastic frontier production function model for TSS shallot farming is as follows: Ln Y = αo + β1ln X1+ β2ln X2+ β3ln X(3)+ β4ln X4+ β5ln X(5)+ β6ln X(6)+ β7ln X(7)+ (vi-ui) ……………..(3. )2 Description: • Y = TSS shallot production per growing season (kg) • α = intercept • β = regression coefficient (estimated parameter coefficient) (i=1 to 7) • X1= land area used for farming (ha) • X2= seed (stem) • X3= labor (HOK) • X4= NPK fertilizer (kg) • X5= organic fertilizer (kg) • X6= dolomite (kg) • X7= pesticide (ml) • vi - ui = (vi) confounding error, (ui ) technical inefficiency effect in the model. Completion of the stochastic frontier production function using frontier 4.1 software with the Maximum Likelihood Estimation (MLE) method. The expected value of the regression coefficient is β1 - β7 > 0, which means that the estimation of the stochastic frontier production function gives a positive value of the estimated parameter. A positive coefficient means that an increase in input use is expected to increase TSS shallot production. The coefficient value of each independent variable can be tested for its significant value using the t-count (t-ratio) value with the t-table value. If the t-count value is greater than the t-table, it can be said that it is significant to the dependent variable and vice versa if the t-count value is smaller than the t-table, it can be said that it is not significant to the dependent variable. At this stage, a classical assumptions test is conducted to measure the regression function model to be used whether there are no violations of classical assumptions related to errors or independent variables. In addition, this method also serves to see whether the function model used is consistent and fulfills the assumptions of the Cobb-Douglas production function. Analysis of the efficiency of the use of production factors is used to determine the extent to which the efficiency of the use of production factors (inputs) that can affect production (output). The analysis of the efficiency of the use of production factors in this study uses a stochastic frontier production function in the form of the Cobb-Douglas production function. Estimation of the production function is done in two ways, namely estimation using the Ordinary Least Squares (OLS) method and estimation using the Maximum Likelihood (MLE) method. The MLE method is useful for estimating the overall production factor parameters, intercepts, and variances of both error components vi and ui. The variation of output from the frontier due to technical inefficiency can be represented by the gamma parameter (γ) as follows (Battese and Coelli 2005). The Maximum Likelihood Method (MLE), describes the relationship between the maximum production that can be achieved at the level of use of factors of production and existing technology. This analysis will determine the technical efficiency of the sample farmers, as well as the factors that influence technical inefficiency. Technical efficiency analysis was calculated using the following formula (Coelli et al. 1998): .................(3. )3 World Journal of Advanced Research and Reviews, 2025, 26(02), 3819-3834 3822 Where TE is the technical efficiency of the i-th farmer, exp(-E[ui|εi]) is the expected value (mean) of uiconditional on εi, so 0 ≤ TEi ≤ 1. The value of technical efficiency is inversely related to the value of the technical inefficiency effect and is only used for functions that have a certain number of outputs and inputs (cross section data). Determination of the extent of the efficiency level is made by referring to Ojo's research, (2006) by dividing the distribution of efficiency levels as follows; highly efficient if TE ≥ 0.90, moderately efficient if 0.70 ≤TE ≤ 0.90 and not yet efficient if TE < 0.70. In measuring allocative and economic efficiency, the dual cost function of the homogeneous Cobb-Douglas production function is first derived (Debertin 1986). The assumption used is the form of Cobb-Douglas production function using two inputs as follows: Y = β(0) X(1) (β)(1) X(2) (β)(2) ......................(3. )4 And the input cost function is C = P(1) X(1)+ P(2) X(2) ................(3. )5 The form of the dual cost function can be derived by using the assumption of cost minimization with output constraints Y = Y0. To obtain the dual cost function, the expansion path value must be obtained, which can be obtained through the Langrange function as follows: L = P(1) X(1)+ P(2) X2 + λ (Yβ(0) X(1) (β)(1) X(2) (β) (2)) ........(3. )6 To obtain the values of x1 and x2 expansion path the Langrange function is derived in the first-order condition as follows: dL dX2=P1−λβ0β1X1β1−1X2β2=0 …….(3. )7 dL dL2=P1−λβ0β2X1β1X2β2−1 =0 ……..(3. )8 dL dλ =Y−β0X1β1X2β2=0 ……..(3. )9 From equations (3.8) and (3.9) we obtain X1 and X2 expansion path values are : X1= (P2 P1)(β1 2)X2 …………(3. )10 X2= (P1 P2)(β2 β1)X1 …………………(3. )11 After that, equation (3.10) is substituted into equation (3.11) to become : Y=β0𝑋1β1[P1 P2β1 β2X1]β2 .........(3. )12 Y=β0P1β2P2−β2β2β2β1β2 ……….. (3. )13 X1β1+β2=Y β0P1β2P2−β2β2β2β1β2. ……..(3. )14 From equation (3.12), the input demand function for X1 and X2 can be summarized as follows is determined to be : 𝑋1=[ Y β0P1β2P2−β2β2β2β1−β2]1 β1+β2 .........(3. )15 World Journal of Advanced Research and Reviews, 2025, 26(02), 3819-3834 3823 𝑋2=[ Y β0P1β2P2−β2β2β2β1−β2]1 β1+β2 …….(3. )16 To obtain the dual frontier cost function, the equations X1and X2are used. is substituted into the cost equation (3.4) i.e. : C*= P1[Y β0P1β2P2−β2β2β2β1−β2]1 β1+β2+ P2[Y β0P2β1P1−β1β1β1β2−β1]1 β1+β2 .........(3. )17 According to Jondrow et al. (1982), economic efficiency (EE) is defined as the ratio between the minimum observed total cost of production (C*) and the actual total cost (C) as shown in the following equation: 𝐸𝐸 =𝐶∗ 𝐶 = 𝐸(𝐶𝑖|𝑢𝑖=0,𝑌𝑖,𝑃 𝐸(𝐶𝑖|𝑢𝑖,𝑌𝑖,𝑃𝑖) = E[exp.( Ui/ε)] ................(3. )18 Economic efficiency (EE) is a combination of technical efficiency and allocative efficiency so that allocative efficiency (AE) can be obtained by equation : 𝐴𝐸 =𝐸𝐸 𝑇𝐸 ……………(3. )19 where EA is 0≤ EA ≤1 and EE is 0≤ EE ≤1. The value of technical efficiency is inversely related to the value of technical inefficiency effect. The inefficiency effect model used in this study refers to the technical inefficiency effect model developed by Battese and Coelli (2005). The use of software Frontier 4.1 in addition to producing regression analysis also produces an analysis of the effects of technical inefficiency in the form of the alleged value of the parameter ui. The variable ui, which is used to measure the effect of technical inefficiency, is assumed to be independent and normally truncated distribution with N(u(i,) (σ) (2) ). According to Elly (2014), to determine the value of the distribution parameter (μi) of the technical inefficiency effect can use the following formula: Ui = δo + δ1z1 + δ(2)z2 + δ(3)z3 + δ(4)z4 + ω(1)d1 .................... (3. )20 Where: • Ui = technical inefficiency effect • z1= farmer age (years) • z2= farmer's formal education level (years) • z3= shallot farming experience (years) • z4= frequency of attending counseling (times/month) • d1= land ownership dummy (d1=0 if owned, d1=1 if rented) • δ = regression coefficient (estimated parameter coefficient) (i=1 to 4) The expected coefficient of the inefficiency estimation parameter (δ) is δ1 - δ4, ω1<0, which means that the estimation of the stochastic frontier production function gives a negative value of the estimated parameter. In order to be consistent, the estimation of production function parameters and inefficiency function (equations 3.2 and 3.4) was done simultaneously with the FRONTIER 4.1 program (Coelli, 1996). Testing the stochastic frontier parameters and technical inefficiency effects was done in two stages. The first stage is the estimation of parameter β1 using the OLS method. The second stage is the estimation of all parameters β0, β1, the variation of Ui and vi using the maximum likelihood method (MLE). The confidence level δ is 5% and 10%, while the test criterion used is the one-way generalized ratio test, with the following test equation: World Journal of Advanced Research and Reviews, 2025, 26(02), 3819-3834 3824 …………..(3. )21 Where L (H0) and L (H1) are the likelihood function values of the null hypothesis and alternative hypothesis, respectively. Test criteria: LR of one-sided error > x2 (restriction) (Kodde Palm table) then reject H0 LR one-sided error < x2(restriction) (Kodde Palm table) then accept H0 If H0: γ = δ0 = δ(1)............... δ5= 0 states that the technical inefficiency effect does not exist in the production function model. If this hypothesis is accepted, then the production function model on average adequately represents the empirical data. Processing results of the FRONTIER 4.1 program. according to Aigner et al., (1977), Jondrow et al. (1982) or Greene (2011), will provide an estimated value of variance in the form of parameterization as follows: 𝜎2= 𝜎𝑣2+ (2) ……………………(3.22 ) …………………. (3.23 ) The parameters of this variance can find a value , hence the value 0: γ:1. The value of the parameter γ is the contribution of technical efficiency in the total residuals. 3. Results and discussion 3.1. Overview of TSS Shallot Farming Based on information from respondent farmers and the results of field observations, the general description of TSS shallot farming in the third planting season from June to August 2024 is shown through the results of descriptive analysis displayed in Table 1 Table 1 Descriptive Analysis of TSS Shallot Farming Respondents in Adipala Subdistrict, Cilacap Regency Variables Minimum Median Average Maximum Production (kg) 350 1.100 1.834 16.000 Productivity (tons/ha) 8,00 14,28 13,44 18,85 Cultivated area (ha) 0,04 0,07 0,13 1 Seedling Quantity (btg) 10.000 22.000 36.237 320.000 Seedling Usage (btg/ha) 142.857 266.733 320.000 Labor (HOK) 7 19 31 270 TK utilization (HOK/ha) 142 232 328 Amount of NPK Fertilizer (kg) 25 40 81 800 NPK Fertilizer Dose (kg/ha) 357 563 875 Amount of Organic Fertilizer (kg) 500 1.000 1.791 16.000 Dose of Organic Fertilizer (ton/ha) 8,6 13,1 20,0 Dolomite (kg) 15 50 87 1.000 World Journal of Advanced Research and Reviews, 2025, 26(02), 3819-3834 3825 Dolomite dosage (kg/ha) 214 608 1.429 Pesticide (ml) 200 1.800 1.656 3.500 Pesticide Dosage (ml/ha) 214 608 1.429 Source: Primary Data Analysis 2025 3.1.1. Production and Productivity The production of shallots in the research area during the third planting season from June to August 2024 was 350 kg to 16,000 kg, depending on the area of land cultivated. Thus, the productivity ranges from 8–18 tons per hectare, with an average of 13.44 tons/ha. This result is higher compared to the average productivity of Central Java Province, which was 10.23 tons/ha, and Cilacap Regency, which was 8.38 tons/ha in 2023 (BPS, 2024). 3.1.2. Land Area Mubyarto (1989) states that land is one of the production factors that serves as the “factory” for agricultural products, contributing significantly to agricultural activities. The scale of agricultural production is influenced by the size of the land used. However, Soekartawi (1993) notes that a larger agricultural land area does not necessarily mean higher land efficiency. On the contrary, with relatively narrow land, supervision of the use of production factors is better, the use of labor is sufficient, and the capital required is not too large. The land area cultivated by the respondents for TSS shallot cultivation ranged from 0.04 to 1.00 ha with an average of 0.13 ha. 3.1.3. Seedling Seed quality determines the superiority of a commodity. The use of high-quality seeds can produce products with good quality. Red onion farmers in Adipala Subdistrict began implementing red onion farming using seed-based seedlings (True Seed of Shallot/TSS) in 2021. The red onion seed varieties (TSS) used in the study area include Maserari, Sanren, and Lokananta. These varieties are certified high-quality seeds released by the Ministry of Agriculture. The seeds must be sown first for 30–45 days before planting. Sowing is done using trays in a nursery. The seed requirement for TSS is 4–5 kg per hectare. If the plant population is 200,000 plants per hectare (planting distance 18 cm × 18 cm, reduced by 30% for drainage ditches), then 1,000 trays are needed for germination, assuming the use of trays with 200 holes per tray. To improve seedling efficiency, some farmers use trays with 220 holes and fill each hole with 2–3 seeds, resulting in more seedlings per tray. Based on interviews with the farmers, the number of seeds used by the farmers ranged from 10,000 to 320,000 stems depending on the land area, with an average of 266,733 stems per hectare per planting season. 3.1.4. Labor Labor is an important production factor that must be considered in the production process in sufficient quantities, not only in terms of availability but also quality and type of labor (Soekartawi, 2003). Labor can be measured in man-days (MD), including land preparation, seedling production, planting, maintenance, fertilization, pest and disease control, and harvesting. The number of laborers used by the surveyed farmers ranged from 7 to 270, depending on the land area, with an average of 232 MD per hectare per growing season. 3.1.5. Fertilization Applying fertilizer with the right composition can produce high-quality products. The most widely used fertilizer is NPK fertilizer, both subsidized and non-subsidized. Other fertilizers are adjusted to the needs of the land and farmers' habits, such as MAP, KNO3, KCl, and others. The first fertilization is carried out when the plants are 10–15 days old after planting. The second application is carried out when the plants are 1 month old after planting. The amount of NPK fertilizer applied by the respondents ranges from 25–800 kg, depending on the land area, with an average of 563 kg per hectare per growing season. The doses of other fertilizers were not studied in this research. Organic fertilizers (such as compost, manure, or green manure) play an important role as base fertilizers in red onion cultivation. Their purpose is to improve soil fertility, provide macro and micro nutrients, enhance soil microorganism activity, and improve soil physical and chemical properties. The organic fertilizers used include cattle manure at a recommended rate of 10–20 tons per hectare or chicken manure at a rate of 5–6 tons per hectare. The organic fertilizers applied by the surveyed farmers ranged from 500 to 16,000 kg, depending on the land area, with an average of 13.1 tons per hectare per growing season. The application of dolomite during soil preparation in red onion cultivation aims to neutralize soil acidity and improve soil structure to support plant growth. Dolomite contains calcium (Ca) and magnesium (Mg), which are important for World Journal of Advanced Research and Reviews, 2025, 26(02), 3819-3834 3826 strengthening plant tissues and enhancing nutrient absorption. Dolomite is typically applied at a rate of 1–2 tons per hectare, spread evenly over the field, and mixed into the soil during initial soil preparation before planting. With optimal soil pH (around 6.0–6.8), red onion growth becomes healthier, bulb yield increases, and the risk of root disease can be reduced. Dolomite applied by farmers in the study area ranged from 15 to 1,000 kg, depending on the land area, with an average of 600 kg per hectare per growing season. 3.1.6. Pest and Disease Control Pest and disease control is carried out to keep plants healthy and produce optimal bulbs. Pesticides are used selectively according to the type of pest or disease attacking, such as insecticides for caterpillar or thrips pests, and fungicides for fusarium wilt or leaf spot diseases. Pesticide application should be based on the pest or disease threshold, with doses and intervals following recommendations to ensure effectiveness and prevent resistance. Excessive use of insecticides can cause losses for farmers, as the chemical compounds in insecticides can lead to environmental contamination and reduced crop yields. Pesticide application rates on red onion crops vary depending on the type of pesticide, target pest or disease, and the active ingredient used. The pesticides applied by the respondents in MT III were generally insecticides with a usage amount of 200–3,500 ml depending on the land area, with an average usage of 1,656 kg per planting season. 3.1.7. Harvesting and Post-Harvest Farmers in the study area harvested red onions when the plants were 60–70 days after planting (DAP), characterized by yellowing leaves, wilting, dark red bulbs emerging from the soil surface, and the characteristic aroma of red onions. Harvested red onions are tied to their stems for easier handling. The bulbs are then sun-dried until sufficiently dry (1– 2 weeks) under direct sunlight, followed by grading according to bulb size. 3.2. Cobb Douglas Production Function Estimation Results The stochastic frontier production function model used in this analysis is a Cobb-Douglas production function consisting of seven explanatory variables, namely land area, seeds, labor, NPK fertilizer, organic fertilizer, dolomite and pesticides. Based on Table 2 and Table 3, it is known that the MLE Log-likelihood value of 65.78 is much higher than OLS of 50.98, so the Likelihood-Ratio test rejects H₀ (OLS). This proves that the MLE method captures the real inefficiency structure. The log likelihood difference of 14.8 shows that the MLE model provides a much better fit to the data than OLS. MLE with its flexibility to use a more suitable distribution, captures the data pattern better, resulting in a higher log likelihood. Table 2 Estimation of Frontier Production Function with OLS method on TSS Shallot Farming in Adipala District, Cilacap Regency Variables Parameters Coefficient Standard error t-count Constant α0 1.33699 0.80576 1.65930 ns Planting area (X1) β1 0.07418 0.10764 0.68916 ns Seedlings (X2) β2 0.13164 0.10361 1.27048 ns Labor (X3) β3 0.31865 0.07713 4.13154 *** NPK fertilizer (X4) β4 -0.02991 0.11647 -0.25683 ns Organic Fertilizer (X5) β5 0.48720 0.13187 3.69456 *** Dolomite/Lime (X6) β6 0.06151 0.06126 1.00405 ns Pesticides (X7) β7 0.00930 0.02054 0.45252 ns sigma-square 0.01391 gamma 0.62000 OLS log likelihood 50.98 Notes: *** significant at α 0.01 level (2.67), ** significant at α 0.05 level (2.01), ; * Significant at α level 0.10 (1.67) World Journal of Advanced Research and Reviews, 2025, 26(02), 3819-3834 3827 Based on Table 3, it is known that the LR Test value in the MLE method is 29.60. This value is greater than 13.40 (Kodde palm table df = 7, α = 0.05) then H Ois rejected and H 1is accepted so it is concluded that there is a case of inefficiency in TSS shallot farming in Adipala District. The estimation results with the MLE method obtained a gamma value of 0.99 and a real effect at the level of α = 0.10. This result shows that 99 percent of the variation in TSS shallot production of respondent farmers is caused by technical efficiency. While the remaining 1 percent is influenced by external influences / stochastic effects that cannot be controlled by farmers such as pests, diseases, land fertility, temperature, climate and so on as well as the influence of random error (vi) or error in modeling. Table 3 Estimation of Frontier Production Function with the MLE method on TSS Shallot Farming in Adipala Subdistrict, Cilacap Regency Variables Parameters Coefficient Standard error t-count Constant α0 3.45008 0.24734 13.94870 *** Planting area (X1) β1 0.30147 0.05664 5.32272 *** Seedlings (X2) β2 0.28895 0.09280 3.11380 *** Labor (X3) β3 0.08965 0.04923 1.82094 * NPK fertilizer (X4) β4 1.03458 0.06001 1.72390 * Organic Fertilizer (X5) β5 0.09958 1.13496 0.87738 ns Dolomite/Lime (X6) β6 0.05754 0.02633 2.18550 ** Pesticides (X7) β7 0.00123 0.01044 0.11808 ns sigma-square 0.03978 5.65618 *** gamma 0.99883 173.189 *** LR-test: 29.60 Log likelihood MLE 65.78 Notes: *** significant at α 0.01 level (2.67), ** significant at α 0.05 level (2.01), ; * significant at α level 0.10 (1.67) The significant value of γ = 0.99 (α = 0.10) confirms that the frontier model is very appropriate in capturing the technical inefficiency of TSSb shallot farming in the study location. Several studies on shallot farming in Indonesia report very high γ (gamma) values, ranging from 0.648 to 0.999, indicating that almost all deviations from the production frontier are due to technical inefficiency. Ismiasih & Jamhari (2024) at the national level (2013 Agricultural Census) found γ = 0.98 in the SFA MLE model, significant at 1% α. Bachtiar & Tamami (2024) in Pacet sub-district of Mojokerto district reported γ = 0.999, indicating the technical inefficiency component almost absolutely dominates the error variance. Mutiarasari et al. (2019) in Majalengka district recorded γ = 0.648 (α 5%), indicating technical inefficiency is still the main driver of output deviation. To determine the effect of each variable on output, a significance test was conducted. Partial testing (t-test) of the production function, shows that the production factors Land area (X1) , Seed (X2) , Labor (X3) , NPK Fertilizer (X4) , Dolomite (X6) , Pesticide (X7) , affect the production of TSS shallots. Table 4.10 shows that all parameter signs in the TSS shallot production function with the MLE method are positive as expected. The parameter estimation value on the stochastic frontier production function can show the elasticity value of the inputs used. The input variables that significantly affect the production of TSS shallots are land area, seeds, labor, NPK fertilizer and dolomite. Meanwhile, the variables of organic fertilizer and pesticide have no significant effect. Estimation of the Land Area variable (X1). The coefficient or elasticity value of the land variable has a real effect on TSS shallot production at the α = 0.01 level with a value of 0.301. This figure shows that the addition of land area can still increase the production of TSS shallots where other inputs remain. A 1% increase in land area increases shallot production by 0.301%, significant at 1%. The results of this finding are in accordance with the research of Muhaimin (2017), Aziza et al. (2022) Aziza et al. (2021); which states that land area has a positive and real effect on shallot production. The land area variable is quite responsive to shallot production, so that TSS shallot cultivation land extensification/expansion activities can be carried out as an effort to increase shallot production in Cilacap Regency. Local governments can support with policies that optimize land use including increased land access, better land management, and mitigation of natural disaster risks that can affect the available land area. World Journal of Advanced Research and Reviews, 2025, 26(02), 3819-3834 3834 [12] Kiromah, S. (2021). Optimization of Input Allocation in Red Onion (Allium Ascalonicum) Farming in Tawangargo Village, Karangploso District, Malang Regency. JU-ke (Journal of Food Security), 4(2), 41-49. [13] Kusnadi, N., N. Tinaprilla, S. H. Susilowati & A. Purwoto. (2011). Analysis of Rice Farming Efficiency in Several Rice-Producing Areas in Indonesia. Journal of Agro-Economics, 29(1):25-48. [14] Minarsih, I., & Waluyati, R. L. (2019). Production Efficiency in Red Onion Farming in Madiun District. Journal of Agricultural Economics and Agribusiness, 3(1), 128-137. [15] Moekani, D. M., Darwanto, D. H., Waluyati, L. R., & Shantosi, A. (2023). Technical Efficiency of Red Onion Farming by Millennial Farmers in Bantul District. Journal of Agro Economics, 41(2), 115-127. [16] Monica, E., Hartati, A., & Wijayanti, I. K. E., (2021). Technical Efficiency of Red Onion Farming on Sandy Soil in Adipala Subdistrict, Cilacap District. Journal of Agricultural Science Vol. 23 No.1, January 2021: 134-147. [17] Muchlisin, M. (2020). Growth and Production Response of Red Onion (Allium ascalonicum L.) to Dolomite Lime Application (Doctoral dissertation, 021008 Tridinanti University Palembang). [18] Muhaimin, M. Wahid. A. (2017). Efficiency of Production Factors in Red Onion Farming in Indonesia. Russian Journal of Agricultural and Socio-Economic Sciences, 2017-05.33. [19] Mutiarasari, N. R., Fariyanti, A., & Tinaprilla, N. (2019). Analysis of Technical Efficiency of Red Onion Commodities in Majalengka District, West Java. Jurnal Agristan, 1(1). [20] Muzazin, N. (2022). Analysis of Red Onion Farming Production (Allium cepa L.) in Bendo Village, Pare District, Kediri Regency. Journal of Agricultural Sociology and Agribusiness, 4(2), 14-24. [21] Nugraha, W. (2022). Factors Influencing Red Onion Production Among Farmers in Enrekang District (Doctoral dissertation). Muhammadiyah University of Palopo. [22] Nurjati, Eka., Fahmi, I., & Jahroh, S. (2018). Analysis of Red Onion Production Efficiency in Pati District Using the Stochastic Cobb-Douglas Production Frontier Function. Journal of Agro Economics, Vol. 36 No. 1, May 2018:1529. [23] Ogundari, K., & Ojo, S. O. (2006). An examination of technical, economic, and allocative efficiency of small farms: The case study of cassava farmers in Osun State of Nigeria. Journal of Central European Agriculture, 7(3), 423432. [24] Agricultural Data and Information Center (2023). Analysis of Agricultural Sector GDP in 2023. Ministry of Agriculture. [25] Agricultural Data and Information Center (2023). Outlook for Agricultural Commodities in the Horticulture Subsector: Red Onions. Ministry of Agriculture. ISSN 1907-1507. [26] Sarlan, M. (2020). Analysis of the Efficiency of Production Factor Use in Red Onion Farming in Pringgabaya District, East Lombok Regency. Rinjani Scientific Journal (JIR). Vol. 8. No. 2. [27] Satrio AD, Witjaksono R, Harsoyo. 2017. The influence of internet use on the adoption of red onion cultivation technology in sandy coastal areas of Bantul Regency. Agridevina, 6(2): 121-133 [28] Simatupang, J. T. , Hutapea, K. P. ., & Aguaninta, D. S. . (2021). Analysis of the Influence of Production Factors on Production and Income of Red Onion Farming: Case Study: Hinalang Village, Purba District, Simalungun Regency, North Sumatra Province. Journal of Agricultural Science Research, 19(2), 37–45. [29] Soekartawi. (2002). Basic Principles of Agricultural Economics. Jakarta: Raja Grafindo. [30] Wijaya, W., Dwirayani, D., Savitri, M. I., Wahana, S., & Astuti, L. C. (2023). Efficiency and Risks of Red Onion Farming in Gebang Subdistrict, Cirebon Regency, West Java. Indonesian Agribusiness Journal (Journal of Indonesian Agribusiness), 11(2), 408-421.