Volume 16, Issue 4, October 2025

Design, Development and Performance Test of Motorized Rotary Cereal Crops Weeder

Beka Adugna Jima, Amana Wako

Agricultural Machinery Engineering, Department of Mechanical Engineering, School of Mechanical, Chemical and Materials Engineering, Adama Science and Technology University, Adama, Ethiopia

Abstract- Weeding with manual labor is labor intensive, time consumption and Chemical Control Weeding method also affect environmental pollution and kill specific species of weeds. The main objective of this study was to design, develop and perform a performance test of motorized rotary weeder to remove weeds from row planted crops at desired depth and spacing. To attain predetermined design, the procedure started from gathering information about weed and continuous withstanding the properties of weeds of the crops, the weeder prototype would have been designed using CATIA. The light weight three row motorized rotary weeder has overall dimension 123cm ×56cm×90cm and the width of coverage of the weeder is 60cm of the three rows in each 20cm space and depth of operation can be adjusted 2.5-3cm. After design and construction, the machine was tested for weed control in seedbed wheat crops at 20 days after sawing and compared with traditional, CCW and in different travel speed with three different treatments to check the performance at field. The data recorded from the experiment have been subjected to the analysis of using origin. The results of field performance evaluation observed 0.224 ha/hr field capacity, 0.086 ha/hr effective field capacity, 94.7% weeding efficiency, 86% field efficiency and 10% decreasing crop damages of the motorized rotary weeder. The power required for operating the weeding in the field was 3.7 hp with wheel slip 6.84%. Also, when it compared with other weeding methods the theoretical field capacity 0.046ha/hr, 2.85ha/h, 0.086 ha/hr, effective field capacity 0.035 ha/hr, 2.56 ha/hr, 0.074 ha/hr and field efficiency 76%, 90%, 86% for manual weeding, CCW and MRW respectively was observed. From the above observed that the machine works satisfactorily for weeding at 20 DAS.

Keywords- Cereal Crops, Design, Production, Performance Test and Row Planted

Download full paper PDF format (Page: 1-10)

Machine Learning for Virtual Personal Assistants

Saad Rehman Babary, Mujtaba Kamal Pasha

Department of Computer Science and Engineering, University of Engineering and Technology, Lahore

Abstract- An Intelligent Virtual Assistant (IVA)—also known as an Intelligent Personal Assistant (IPA)—is a smart software application designed to help people complete tasks or respond to queries using voice or text input. These assistants are often called chatbots, especially when used in messaging platforms or online chats. While some chatbots are built just for entertainment, today’s virtual assistants have become much more advanced. They can understand spoken language and reply using synthesized, human-like voices. Users rely on them for a wide range of everyday tasks—whether it’s asking questions, controlling smart home devices, playing music or videos, or managing emails, calendars, and to-do lists—all through simple, hands-free voice commands.

Keywords- VPA, NLP, Speech to text, Text Analysis and Artificial Intelligence

Download full paper PDF format (Page: 11-15)

AI-Driven Electroencephalography Analysis for Detecting Student Confusion in Digital Learning Environments

Nimra Latif

Department of Data Science, University of Engineering and Technology, Lahore, Pakistan

Abstract- EEG Conformer is a new transforming established to substitute traditional convolutional with instincts of the transformation architectures, which makes it revolutionary in decoding EGG. Transformer architectures are studied in this research and their evolution challenges, guiding them towards the ground-breaking EEG Conformer. The model, a combination of convolutional neural networks (CNNs) and transformers is good both in terms of accuracy as well as interpretability. Some attributes of the EEG Conformer are plotted: visualization techniques, temporal dynamics and real-time applications where this model reveals its flexibility. Settings the stage for a comparative analysis of current EEG models, superior performance is shown by Conformer. Its broader impact is emphasized by transfer learning, ethical considerations and anticipation of future trends. The inclusion of novel dataset would involve EEG data related to students watching MOOC videos with clips introducing confusion. Students’ self-reported level of confusion provides another dimension to the dataset, allowing for broader use of EEG Conformer. The EEG Conformer in its essence is a transformative model, which can be characterized by the seamless fusion of architectural elements and novel datasets that allow it to spearhead EEG decoding and visualization into new horizons.

Keywords- Digital Learning Environment, EEG Conformer, Confusion Detection and Artificial Intelligence (AI)

Download full paper PDF format (Page: 16-23)

Investigation of Mechanical and Physical properties of Polymer Matrix Composite Reinforced with Pineapple Leaf Fiber and Coffee Husk Filler Composite using Response Surface Method

Sisay Diriba Debala, Kinisa Wareso Abesho, Beka Adugna

Department of Mechinical Design and Manufacturing Enigneering, Adama Science and Technology University, Ethiopia
Department of Mechinical Enigneering, Bule Hora University, Ethiopia
Agricultural Machinery Engineering program, Department of Mechanical Engineerin, School of Mechanical, Chemical and Materials Engineering at Adama Science and Technology University, Adama, Ethiopia

Abstract- In the search for structural materials that are strong, light, and cheap, pineapple leaf, which is rich in cellulose and relatively inexpensive, seems to have good potential reinforcement in yarn production. In particular, pineapple leaf fibers (PALF) and coffee husk filler (CHF) can be new sources of raw materials for industries and can be potential for polymer reinforcement. This study utilized Response Surface Methodology experimental design with the most recent Design Expert 13.0 software, and the composite samples were fabricated through hand layup method. The analysis of variance (ANOVA) was used to determine the significance of variables, and the interaction between variables and responses. To assess mechanical characteristics (such as tensile, compression, flexural, and impact strength) and physical characteristics (such as water absorption), regression models were developed and statistically validated. The quadratic model was found to be the best fit for the tensile strength, flexural strength, impact strength and water absorption models, while the two-factor interaction model was determined to be the best fit for the compression strength. The primary significant output parameter for tensile strength, compression strength, flexural strength, impact strength, and water absorption were 28.497% of PALF, 65.41% of PALF, 29.755% of CHF, 84.454% of PALF, and 56.92% of PALF contribution respectively in all responses.

Keywords- Pineapple Leaf Fiber, Coffee Husk Filler, Response Surface Methodology, ANOVA and Quadratic Model

Download full paper PDF format (Page: 24-35)