Mechanized tomato harvesting

The tomato crop is one of the most sensitive in the harvesting process, requiring care from operators and well-performed calibrations.

The mechanized harvesting operation is one of the most important tools in the entire production process of tomato cultivation for industrial processing. In this context, the smaller the losses resulting from harvesting, there will be a greater increase in productivity, reflected in greater total production in the future.

Tomato is a very sensitive product that requires several cultural treatments, which generates high costs in labor and machine time. Therefore, there is concern on the part of producers in measuring harvest losses. The harvesters currently used have a track system that highlights the fruit branches through the forced vibration of a rotor, this system being used in large crops, such as coffee.

The track system must have precise regulation, as the better this is, the better the performance and efficiency will be at the time of operation. In more traditional crops, such as soybeans and corn, studies carried out on the trail system are very common, however, the same does not occur with tomato crops.

Statistical Process Control is a technique used in production processes that helps detect problems in the process of a product, aiming to reduce waste and rework, as well as increase productivity.

The use of statistical process control in this agricultural operation can prove to be fundamental, as it can present us with a view of how the process is occurring, indicating possible failures and possible improvements for the next operations.

In view of the above, assuming that there is a certain variability in fruit losses in the mechanized harvesting of industrial tomatoes, the objective of this work was to evaluate the quality of the operation of the mechanized harvesting of industrial tomatoes, due to losses in the trail system, through of statistical process control.

The present study was carried out in the municipality of Morrinhos, Goiás, using the tomato cultivar HEINZ 9553 transplanted in a direct planting system. The harvesting process took place 125 days after transplanting, using a GUARESI self-propelled harvester to assess losses., model G-89/93 MS 40”, with FIAT-Iveco 129 kW engine. During the work, the accelerator was used with the engine rotating at 1900 rpm and an average operating speed of 1,14 m/s. At the time of harvest, where the data were collected, the soil had an average water content of 18,4 %.

The experimental design adopted was completely randomized with four replications, consisting of three harvesters with different ages (M1: 7.255 h; M2: 2.984 h; M3: 6.512 h). Losses in the trail system were characterized by fruits that did not detach from the branches after the harvester passed through the area.

To quantify losses, a template with an area of ​​1 m was used.²(1x1 m), where after the harvester passed, the necessary materials for loss assessments were collected at each point. After the material was collected, it was bagged, identified and weighed on a digital scale, model AD2000 with an accuracy of 0,01g. After weighing the collected material, the values ​​were extrapolated to kg/ha.

The process control charts were prepared based on the statistical parameters of the loss process in the threshing system, considering as the central line the average loss in the threshing system in each machine and the average amplitude composed of the upper control limit ( LSC) and the lower control limit (LIC). The LSC and LIC control limits were calculated based on the standard deviation of the variables corresponding to 3σ and -3σ, respectively.

Figure 1 shows the results of losses in the trail system for machine 1. The general average in this occurrence was 8263 kg/ha, with the upper and lower limits being 19.699 and -3174 kg/ha, respectively. In this context, it was observed that all observations are within control.

The range of observation values ​​presented an average of 4300 kg/ha and obtained lower values ​​between observations 2 and 3 where the values ​​found were close to 0 (LIC). In general, the amplitude between observations does not exceed the upper and lower control limits, indicating that the process was not interfered by special causes. Wear on internal mechanisms interferes with trail efficiency, and even with adequate adjustment, the mechanism is compromised by the current condition of its elements.

Figure 2 shows the results of losses in the trail system for machine 2. The general average in this occurrence was 5.250 kg/ha, with the upper and lower limits being 12.298 and -1.798 kg/ha, respectively. 

The amplitude of the observation values ​​presented an average of 2.650 kg/ha and continued to vary between the average of the differences, mainly between observations 2 and 4. The amplitude between the observations did not exceed the upper and lower control limits, indicating that the process was not interfered with by special causes.

This machine showed lower losses, and this can be explained by the shorter time of use and consequently less wear on the track mechanisms. Adjusting the operation of a track system with shorter usage time also has greater precision and, as a result, there is greater work efficiency of its internal mechanisms.

Figure 3 shows the results of losses in the trail system for machine 3. The general average in this occurrence was 7.613 kg/ha, with the upper and lower limits being 13109 and 2.116 kg/ha, respectively. This machine presented lower losses than machine 1 and greater than machine 2, and this behavior can be explained by the number of hours worked on the machines (intermediate value between machines 1 and 2).

The range of observation values ​​presented an average of 2.067 kg/ha and continued to vary between the average differences, mainly between observations 2 and 3. The observations did not exceed the upper and lower control limits, indicating that the process did not suffer interference from special causes.

Losses in the threshing system by the harvesters tested were within the control limits and standards established in this study, with machines with a lower number of hours worked showing lower losses in the threshing system.

Túlio de Almeida Machado, Haroldo Carlos Fernandes, Federal University of Viçosa; Cristiane Fernandes Lisboa, Daniel Antônio da Cunha; State University of Goiás, João Paulo Barreto Cunha, Federal Rural University of Rio de Janeiro

Article published in issue 162 of Cultivar Máquinas.