Next to banana and pineapple, mango is the third most important fruit crop of the Philippines based on export volume and value. The country’s export variety, the “Carabao mango,” is one of the best varieties in the world. In fact, the Guinness Book of World Records listed the said variety as the sweetest mango in the world.
Unfortunately, “the Philippine mango industry has been on a continuous decline in all indicators of industry performance which include production volume, productive area, as well as yield per unit area and yield per tree,” said Philippine Mango Industry Roadmap 2021-2015.
From 2000 to 2009, the area planted to mangoes was 163,106 hectares with 925,249 metric tons production. The yield per unit area was 5.7 metric tons per hectare. This was based on the record provided by the Department of Agriculture.
Although the hectare of area planted to mango went up (187,530 hectares) from 2010 to 2020, the production went down (793,296 metric tons). The yield per unit was 4.2 metric tons.
“Despite this, the industry remains to be the third highly exported fruit crop in the Philippines, recording a gross value added of P35.520 billion and 1.95% contribution to the major industry in 2020,” the roadmap said.
The aim of the roadmap is to have “a sustainable and resilient Philippine mango industry” which would offer “competitive and world class mangoes through innovation and inclusivity.”
In recent years, the Department of Science and Technology (DOST) has come up with some solutions to the problems that beset the mango industry.
For instance, the manual classification of mangoes has long been a bottleneck in the mango supply chain, characterized by time-consuming efforts and subjective judgment. The University of the Philippines Cebu (UP Cebu) has harnessed the power of artificial intelligence (AI) and brought automation to the labor-intensive task of sorting carabao mangoes for the fresh export market.
UP Cebu Professor Jonnifer Sinogaya headed a team that conducted the “Mango Automated Neural Net Generic Grade Assignor (MANGGA),” which the DOST’s Philippine Council for Agriculture, Aquatic and Natural Resources Research and Development (PCAARRD) is funding.
The team’s systematic approach to data acquisition has led to an extensive data set of 10,440 images captured from various angles and orientations and corresponding ethylene concentrations collected from 870 individual mangoes, which served as the cornerstone for training a cutting-edge AI model for sorting Carabao mangoes.
The MANGGA project team has coded the Convolutional Neural Network (CNN) from scratch and also created an image data acquisition system. Their preliminary training of a single-input CNN model exhibited an impressive 94% accuracy in determining whether mangoes are suitable for export based on their overall visual characteristics.
Using the Philippine National Standard for quality metrics, the refinement of the CNN and Computer Vision System (CVS) promises a more efficient way to grade export-quality Carabao mangoes.
“The MANGGA project encourages the adoption of a smart postharvest system within the local mango industry,” wrote Thea Mariel Valdeavilla of the Science and Technology Information Institute. “With the premise of creating a conveyor system designed to sort mangoes based on their marketability, this initiative stands poised to revolutionize mango grading, offering efficiency and safety to the fresh export market.”
Another problem that hampered the mango industry is fruit fly infestation. According to the Department of Agriculture, the flies destroy up to 80% of mango fruits, with those affected showing brown scab-like spots. These pests usually start infesting mango trees shortly after flowering.
To reduce fruit fly infestation and improve the yield and quality of carabao mango, the PCAARRD collaborated with the Australian Center for International Agricultural Research last year in coming up with innovative area-wide management (AWM) approaches.
“With the integration of AWM, pest and disease control strategies, and best management practices,” farmers can produce export-quality carabao mangoes. The said approach was initiated in project sites located in Island Garden City of Samal, Davao del Norte and Davao City.
One of the major contributors of the success of the study is the recommended fruit bagging material used by the farmer cooperators, which provided visible differences in the fruit quality as compared to the use of imported newspapers.
“The use of the said fruit bagging material may also potentially reduce pesticide application,” said Dr. Emma Ruth V. Bayogan, of the University of the Philippines Mindanao, who led the project team.
Aside from exploring the optimal number of uses for the fruit bagging material, the team is also working on improving postharvest handling systems, such as optimizing hot water treatments to prolong the shelf-quality mango.
Dr. Celia D. Medina, an entomologist from the University of the Philippines at Los Baños, is doing the fruit fly monitoring study to continuously assess fruit fly injury levels on the mango fruits produced.
“Significant outcomes of the project are set to increase the profits of mango growers, expand market access, and boost the mango industry, not only in the Philippines but the whole Asia-Pacific region,” wrote STII’s Danica Louise C. Sembrano.
Meanwhile, the Philippine Center for Postharvest Development and Mechanization (PhilMech) has successfully developed a mango sorting machine that enables farmers to cut production costs and increase product value.
The PhilMech automated mango sorting machine uses a computer vision system to sort and classify green mango fruit. The machine can detect external defects, weight, and sizes. It works by feeding the mango fruits into the machine through a conveyor belt and passing through an imaging chamber before being sorted directly into the classification bins. Use of the machine will cut the labor requirements from 20 workers down to two workers.
The machine was able to record a 94.44% accuracy and precision in sorting and grading mango fruits following international standards for exports. It can also process up to 720 to 800 pieces of mango per hour.
PhilMech expects that the sorting machine will be available to farmers at an affordable price. Compared to similar imported products, which may cost up to P1 million, producing a unit of PhilMech mango sorting machine only costs around P168,000.
Tackling the concern of unemployment due to the automation of industries, lead developer Engineer Arlene Joaquin expresses that the Automated Mango Sorting Machine can be seen as an opportunity rather than a threat. She shared that among the concerns of some mango farms is the lack of manpower especially during peak harvest season.
“The problem with the mango industry is that when the harvest is simultaneous, no one is available. And (the Mango Sorting Machine) is a very good intervention because they can still hire laborers from neighboring towns,” said Engr. Joaquin.
