Date of Publication

4-2025

Document Type

Bachelor's Thesis

Degree Name

Bachelor of Science in Manufacturing Eng'g & Mgt w/ Specialization in Mechatronics & Robotics Eng'g

Subject Categories

Biomedical Engineering and Bioengineering

College

Gokongwei College of Engineering

Department/Unit

Manufacturing Engineering and Management

Thesis Advisor

Ira C. Valenzuela

Defense Panel Chair

Renann G. Baldovino

Defense Panel Member

Ronnie S. Concepcion II
Robert Kerwin C. Billones

Abstract (English)

In the Philippines, iceberg lettuce (Lactuca sativa var. capitata) is a priority crop that lacks an efficient postharvest system, with quality grading and sorting still performed manually by farmers. To optimize post-processing time and maintain quality, an automated grading and sorting system for iceberg lettuce was developed. This study aimed to design a vision-based quality grading and sorting system within a mini-production line, integrating computer vision and automation to enhance postharvest handling efficiency. Multi-gene symbolic regression programming (MSRGP) was employed to develop predictive models for chlorophyll-a content, moisture content, and fresh head weight. The resulting models exhibited varying levels of accuracy, with R² values of 0.72 for chlorophyll-a, 0.431 for moisture content, and 0.565 for fresh head weight. Prediction accuracies for these parameters were 89.97%, 85.38%, and 92.80%, respectively, with the highest accuracy observed in moisture content prediction. The three quality parameters were utilized as inputs for fuzzification in a Mamdani-Type 1 fuzzy logic system, which classified iceberg lettuce into three grades: Grade 1 (high quality), Grade 2 (medium quality), and Grade 3 (low quality). The fuzzy logic prediction model demonstrated an average classification accuracy of 90%, correctly grading 27 out of 30 test samples. The automated sorting conveyor system, which accommodates five lettuce heads per batch, exhibited a repeatability rate of 100% at 70% speed and 92% at full operational speed upon five trials. Lastly, the throughput time achieved 8 lettuces per minute based on the system time study. The findings highlight the potential of computer vision and fuzzy logic-based automation in optimizing postharvest systems, enhancing the farm-to-market supply chain for iceberg lettuce in the Philippines.

Abstract Format

html

Abstract (Filipino)

None

Abstract Format

html

Language

English

Format

Electronic

Keywords

Computer vision; Assembly-line methods

Upload Full Text

wf_yes

Embargo Period

4-2026

Download

Share

COinS