ECONOMIC ANALYSIS OF DIFFERENT HYDRAULIC SPRAYERS USED IN SUGARCANE ( Saccharum spp . )

Brazil is the largest sugarcane producer worldwide. In mills, sugarcane crops are sprayed with agrochemicals using self-propelled, tractor-driven, and aerial hydraulic sprayers. As sugarcane spraying is regularly performed over extensive areas, the machinery to be used should be planned and dimensioned to ensure timely operations without overloading and at lower costs. Therefore, the current study aimed to analyze the operational and economic performances of different hydraulic sprayers to perform sugarcane crop control. As long as meeting our goal in field conditions would be difficult, we opted to develop a computational model, named as “TratoCana”, using a spreadsheet and programming language. The model was used to assess economic and operational factors by generating scenarios and analyzing probable routine errors. In short, the results evidenced that initial value and operation speed are factors with the strongest impact on the costs of self-propelled sprayers and tractor-sprayer sets. Yet, the aerial application was mostly affected by fuel costs and crop row lengths. Moreover, the larger the spray tank volume, the lower the costs.


INTRODUCTION
During the 2017/2018 growing season, sugarcane grown area in Brazil was estimated at 8.76 million hectares, with a production estimate of 646.34 million tons (CONAB, 2017).These are extensive areas and have to be regularly sprayed for pest and disease control.However, the planning and management of such operations are essential to ensure that they meet the established a steady and timely pace.Moreover, the planning of spraying routines could also avoid unfavorable weather conditions and improve application efficiency, reducing agrochemical losses and maximizing operational costs, among others.According to Dash & Sirohi (2008), Sichonany et al. (2011), andAkinnuli et al. (2014), agricultural mechanized operations should be planned and managed previously for punctuality, without under-or overloading machinery and having minimal operational costs.
The absence of planning and management in agricultural machinery has resulted in discrepant reductions and increases in operation times.Accordingly, Hansen et al. (2007) studied a row crop harvesting pattern by combines for turning maneuvers, considering four platform sizes (6, 8, 12, and 16-row heads); they observed that the 12-row platform provided the shortest turn time in a row crop.
Agricultural spraying machines can be selected by their operational and economic performances.Basically, operational performance is measured by the following parameters: operating speed, field efficiency, operational field capacity, worked hours, machine hours, and the number of machines required.In turn, economic performance is given by fixed costs per year and per hour, and those with fuel consumption, repair and maintenance services, and operational costs (Mialhe, 1974;Balastreire, 1990;Hunt, 1995).
According to Jokiniemi et al. ( 2012), Rivera et al. (2012), Shamshiri & Ismail (2013), Zaied et al. (2014), andOduma et al. (2015), field efficiency has a direct influence on the operational field capacity of a machine and on mechanized operations themselves.As stated by Araldi et al. (2013), Zhou et al. (2015), and Santos et al. (2018), mechanized operations are performed considering operational times to improve their field efficiency.In this line, Yousif et al. (2013), Ma et al. (2015), and Cervi et al. (2015) claimed that field efficiency is the ratio between the effective time spent in harvesting operations and the total time that a machine remains in the field.
Currently, computer modeling has been adopted to manage agricultural machinery selection.For Santos et al. (2015), such tool facilitates the development of structures and generation of scenarios to be used in a selecting process.Still, according to the same authors, the use of computer models for planning and management of agricultural machines provides acceptable solutions to the arising problems.
Engenharia Agrícola, Jaboticabal, v.39, n.2, p.225-233, mar./apr. 2019Mercante et al. (2010) developed a software, named "PRAPRAG", based on programming language and concluded that self-propelled sprayers have variable costs similar to those of agricultural tractors.Likewise, Faria & Silva (2015) built a model to assess the effects of changes in a maintenance program of mechanized sugarcane harvesting and observed reductions in maintenance rates from 10.0% to 3.5%.Khoub Bakht et al. (2009), Rashidi &Ranjbar (2010), andNiari et al. (2012) also developed some models to analyze the roles of spare parts, operator and mechanic wages, lubricants, oil filter, and fuel consumption in the costs with tractor repairment and maintenance.Similarly, Rohani et al. (2011) elaborated a model to solve cost issues with tractor repairment and maintenance.
In short, agricultural machinery management consists of systematically selecting optimum machines, once performance parameters are interrelated among themselves.Therefore, the aim of the current study was to analyze the operational and economic performances of different hydraulic sprayers used in sugarcane crops.

MATERIAL AND METHODS
A sample scenario was developed for a fictitious sugarcane mill, considering an area of 22,000 ha, a mean sugarcane yield of 80.00 Mg ha -1 , and a price paid to growers for delivered sugarcane of US$ 22.04 Mg -1 , in accordance with UDOP (2016).For this scenario, the spraying of agrochemicals was carried out by three hydraulic sprayers: self-propelled, tractor-driven, and aerial sprayers.Table 1 shows the economic, technical, and operational features of the selected machines.A computational model, named "TratoCana" (version 2.0), was developed aiming at meeting the main characteristics of mechanized spraying for sugarcane cultivation.This model was developed in an Excel ® spreadsheet and was based on the Visual Basic ® programming language.Figure 1 shows the flowchart of the model, which was elaborated according to the symbology proposed by Oakland (2007).
After starting its operation (1)2 , the next steps consisted of inputting (2) crop data such as the area to be sprayed, mean sugarcane yield, and price paid per ton of sugarcane; (3) climate data such as total number of spraying days, worked days, relative air humidity, air temperature, and wind speed; and (4) operational pace based on crop and climate data.
Then, data entry was performed using information about: (5) technical and operational features of terrestrial spraying application such as tip number, tip spacing, tip flow, replenishment time, pump-to-field mean distance, mean crop row length, operation speed, turning maneuver speed, replenishment speed, total tank volume, field efficiency, among others; and (6) technical and operational features of aerial spraying application such as tip number, tip spacing, and tip flow, replenishment time, runway-tofield mean distance, mean crop row length, operation speed, runway-to-field running speed, effective row width, each return curve time, ground time between flights, aerial application rate, total tank volume, field efficiency, among others.
Engenharia Agrícola, Jaboticabal, v.39, n.2, p.225-233, mar./apr.2019 The operational pace associated with the technical and operational features of all spraying systems was used to determine the operational performance of the evaluated machines, sprayer3 , tractor-sprayer set, and aircraft, (7) using available time, operational field capacity, application volume, total application flow rate, total time for displacement and replenishment, total traveled distance, machine hours, and number of required machines.
The results of operational performance were associated to machinery economic data ( 8) such as initial and final values of machines; lifespan (in years and hours); interest per year; lodging, insurance, and taxes (LIT); fuel consumption; repair and maintenance service, among others.These enabled us to calculate their economic performances (9) in terms of costs per hour, area, and liter.
In brief, the results generated by the model (10) allow users to assess both operational and economic performances of mechanized spraying operations, and to infer (11) whether it was feasible (12) or not.As the operation was deemed unfeasible ( 13), new data should be inserted into the model, or another scenario has to be devised.

Agroclimatic factors
The factor 'climate' for the fictitious mill was established based on the number of working days inappropriate for spraying (NWDIS), as proposed by Santos (2017).This approach takes into account agroclimatic parameters such as relative air humidity (RH), wind speed (WS), and air temperature (T).Therefore, for this purpose, mean values of agroclimatic parameters from Rio Largo County -AL (Brazil) in 2014 were used (  Mialhe (1974), the number of working days inappropriate for spraying (NWDIS) was also considered as available time (AT) (Equation 1).Thus, available time (AT) was calculated by adding the number of working days inappropriate for spraying (NWDIS) and the number of sundays and holidays (NSH), discounting the number of days (ND) and then multiplying it by the working hours (WH).

Operational performance
The operational performances of both the sprayer and the tractor-sprayer set were measured as proposed by Mialhe (1974), aiming at defining the number of required machines for the spraying of agrochemicals in the sugarcane field of the assessed mill.
The number of machines (NM) was estimated by the ratio between the operation pace (OP) and operational field capacity (OFC) of the machine (Equation 2). (2) Where, NM = number of machines; OP = operation pace (ha h -1 ), and OFC = operational field capacity (ha h -1 ).
The operational pace (OP) was measured by the ratio between the area to be sprayed (AS) and available time (AT) to perform the agricultural operation.
Operational field capacity (OFC) was estimated by the association of total boom width (TBW), operation speed (OS), and field efficiency (FE).
The volume of spraying solution to be applied was determined in accordance with the guidelines of Matuo et al. (2010) (Equation 3).Application volume (AV) was the ratio between tip flow (TF) and the spacing between tips (SBT), multiplied by operation speed (OS).Total application flow (TAF) was defined by the association between application volume (AV) and operational field capacity (OFC).
Displacement time for replenishment (DTR) corresponds to the time for back and forth to the field, calculated as the ratio between the field-to-pump mean distance (FPMD) and replenishment speed (RS).
In turn, the total time for displacement and replenishment (TTDR) corresponded to the total time spent going to the pump, replenishing, and returning to the field.It was measured by summing displacement time for replenishment (DTR) and replenishment time (RT).
The number of replenishments (NR) was established according to the relationship among application volume (AV), area to be sprayed (AS), and total tank volume (TTV).
Turning maneuver distance (TMD) was established as the turning radius of the machinery by doing a turning (TRMT).
The time to finish each turning (TFT) was estimated as the ratio between turning maneuver distance (TMD) and turning speed (TS).
The number of turning maneuvers (NTM) was defined as the relationship among area to be sprayed (AS), total boom width (TBW), and mean length of rows (MLR).
To define the number of required machines for aircraft spraying, the measurement of the operational performance also followed the proposal of Mialhe (197 4 ), as described for the sprayer and for the tractor-spray set.
The operational field capacity (OFC) of aerial spraying was calculated as adjusted model of Araújo (2009), associating the total volume of sprayer tank (TVST), application volume (AV), distance between runway and the field (DRF), running speed from runway to the field (RSRF), effective row width (ERW), operation speed (OS), time for each return curve (TRC), mean length of rows (MLR), and ground time between flights (GTF).

Sprayer economic performance
The total cost of the sprayer (TCS) was determined by associating the operational cost of the sprayer (OCS) and the area to the sprayed (AS), as follows (Equation 4).
Engenharia Agrícola, Jaboticabal, v.39, n.2, p.225-233, mar./apr.2019 As done for the sprayer and the set, repair and maintenance costs (RMC) of the aircraft was determined according to ASABE (2011), while the repair and maintenance factor (RMF) used data provided by PBA AVIATION (2012).

Mill revenues
Gross and net revenues of the fictitious mill were estimated by an adaptation of the calculations done by Santos et al. (2015).This calculation integrates raw material losses in the gross revenue and mechanized harvesting costs in the net revenue.Thus, the gross revenue of the mill (GRM) encompassed sugarcane ton price (STP) and sugarcane yield (SY), whereas the net revenue of the mill (NRM) considered the difference between gross revenue of the mill (GRM) and total cost (TC) of the machine.

Validation
The computational model "TratoCana" (version 2.0) was validated by setting the simulation results against raw data (primary) from the field and against the bibliographic findings (secondary).Furthermore, model sensitivity and consistency were assessed by the cost analysis.

RESULTS AND DISCUSSION
The mean values of climatic parameters used for the mill refer to the data recorded in Rio Largo County -AL (Brazil) in 2014.Such parameters resulted in a number of working days inappropriate for spraying (NWDIS) of 257 and available time (AT) of 2,583 hours.
According to the results found for the model scenario shown in Figure 2, annual interest (25.99%) (AI) was the most representative factor in sprayer cost, followed by annual depreciation (25.67%) (AD), and lodging, insurance, and taxes (LIT) (6.42%), totaling 58.08% (Figure 2a).As these three items compose the fixed costs, the variable costs represented 41.93% of the total.For the set, the repair and maintenance costs (RMC) had the greatest expression in costs (29.65%), followed by annual depreciation (AD) (25.17%).Fixed costs totaled 52.03% and variable costs accounted for 47.96% (Figure 2b).Fuel cost (FC) was the most representative parameter for the aircraft (67.23%), followed by repair and maintenance costs (RMC) (19.37%), totaling 86.60% of the variable costs and 13.39% of the fixed costs (Figure 2c).
Figure 3 shows that the minimum area for the sprayer, the set, and the aircraft were 261, 228, and 92 ha, and had operational costs of 158.97, 85.04, and 325.51 US$ ha -1 , respectively.Decreases in cost were due to the amortization of fixed costs with machinery, on the basis of worked hours.The larger the area, the longer the worked hours using machines, until the operational field capacity (OFC) was unable to meet the operation pace (OP).This leads to the entrance of a new machine into the field, i.e., two more sprayers, tractor-sprayer sets, and aircraft.Both the sprayer (261 ha) and the set (228 ha) worked 11 hours, while the aircraft (92 ha) operated 12 hours.For the sprayer, the need for a second machine occurred from 52,073 ha sprayed, totaling 2,245 worked hours.For the set, it was from 45,564 ha sprayed, i.e., 2,272 worked hours.And, lastly, there was a need for another aircraft from 18,327 ha (i.e., 2,322 worked hours).
As can be seen in Figure 4, the total cost of machinery use has a share in the revenue of the fictitious mill.According to the used scenario, the sugarcane yield (SY) was 1,760,000 Mg and the gross revenue of the mill (GRM) was US$ 38,791,500.Accordingly, the total cost (TC) of the sprayer reached a share of 0.18%, resulting in a net revenue of the mill (NRM) of 99.82% or US$ 38,720,659.06 (Figure 4a).For the set, total cost (TC) was 0.10% and the net revenue of the mill (NRM) was 99.90% or US$ 38,754,590.63 (Figure 4b).If compared to the sprayer, there was a cost difference of 0.09% (US$ 33,931.88) in the net revenue of the mill (NRM).The aircraft was the one that most impacted the gross revenue of the mill (GRM) since it reached 1.09% (Figure 4c) and, when compared to the sprayer, showed a difference of 0.91% and of 1.00% if compared to the set.

CONCLUSIONS
In the mechanized spraying system, the aircraft had the highest operational costs.
The tractor sprayer set had the least impact on the gross revenue of the mill, followed by the self-propelled sprayer and the aircraft.
time (h); ND = number of days; NSH = number of Sundays and Holidays; NWDIS = number of working days inappropriate for spraying, and WH = working hours (h).

FIGURE 2 .
FIGURE 2. Distribution of costs (%) in the model scenario: a -sprayer, b -tractor-sprayer set, and c -aircraft.AD -Annual depreciation; AI -Annual interest; LIT -Lodging, insurance, and taxes; RMC -Repair and maintenance costs; and FC -Fuel cost.
FIGURE 3. Operational cost and number of machines required as a function of the area to be sprayed.

FIGURE 4 .
FIGURE 4. Distribution (%) of the gross revenue (GRM), net revenue of the mill (NRM), and total cost (TC) of the machinery: a -Sprayer, b -Tractor-sprayer Set, and c -Aircraft.

TABLE 1 .
Economic, technical, and operational variables of the selected machines.

Table 2 )
. These data were obtained from the agrometeorological station located in the Center for Agricultural Sciences of the Federal University of Alagoas (CECA/UFAL).

TABLE 2 .
Mean values of agroclimatic parameters used in the model scenario.