- Manufacturing Process and Optimization
- Assembly Line Balancing Optimization
- Industrial Vision Systems and Defect Detection
- Additive Manufacturing and 3D Printing Technologies
- Advanced Statistical Process Monitoring
- Advanced Manufacturing and Logistics Optimization
- Muscle activation and electromyography studies
- Product Development and Customization
- Evacuation and Crowd Dynamics
- Prosthetics and Rehabilitation Robotics
- Human-Automation Interaction and Safety
- demographic modeling and climate adaptation
- Food Supply Chain Traceability
- Advanced Sensor and Control Systems
- Smart Agriculture and AI
- Flexible and Reconfigurable Manufacturing Systems
- Impact of AI and Big Data on Business and Society
- Traffic and Road Safety
- Technology and Data Analysis
- Advanced Decision-Making Techniques
- Remote Sensing in Agriculture
- Advanced Data Processing Techniques
- Intergenerational Family Dynamics and Caregiving
- Additive Manufacturing Materials and Processes
- Identification and Quantification in Food
University of Kigali
2025
Carnegie Mellon University Africa
2019-2024
IBM Research - Thomas J. Watson Research Center
2021
Ulsan National Institute of Science and Technology
2013-2019
Abstract We identify two significant issues that render prosthetics inaccessible. First, obtaining a representation of the residual limb can be difficult to access. Conventional approaches require equipment or expertise often unavailable in resource-constrained communities. Second, it is challenging determine prosthetic design, filament material, and printing process satisfies mechanical functionality requirements because predict properties 3D-printed prosthetics. Therefore, we propose...
Manufacturing systems have evolved to adopt a mixed-model assembly line enabling the production of high product variety. Although system with semi-automation (i.e. human involvement) can offer wide range advantages, becomes very complex as variety increases. Further, while complexity from different options worsen performance, there is lack quantifiable models for manufacturing in literature. Thus, this paper, we propose novel method quantify choice effective management semi-automated line....
Stock trend prediction is an important area of study for researchers and practitioners.In recent years, along with traditional statistical models, machine learning deep techniques have been increasingly adopted in various financial studies.Long Short-Term Memory (LSTM) one the models predicting time-series data.In case vanilla LSTM, shared weights are learned based on all available data; hence, it difficult to accurately learn patterns predict future value from a subset this paper,...
Over the past decades, additive manufacturing has rapidly advanced due to its advantages in enabling diverse material usage and complex design production. Nevertheless, technology limitations terms of quality, as printed products are sometimes different from their desired designs or inconsistent defects. Warping deformation, a defect involving layer shrinkage induced by thermal residual stress generated during processes, is major factor lowering quality raising cost products. This study...
Abstract Customized prosthetics are often inaccessible in resource-constrained communities. The authors of this work identify two significant barriers. First, obtaining a digital representation the residual body is inaccessible. proposed methods overcome issue by achieving via smartphones. Second, determining design and 3D printing process challenging. Although additive manufacturing technology widely adopted to fabricate customized because its high accessibility, it has drawback that...
Despite the recent advances in manufacturing automation, role of human involvement systems is still regarded as a key factor maintaining higher adaptability and flexibility. In general, however, modeling operators system design considers physical resource represented statistical terms. this paper, we propose loop (HIL) approach to investigate operator's choice complexity mixed model assembly line. The HIL simulation allows humans become core component simulation, therefore influencing...
Increasing production variability while maintaining operation efficiency remains a critical issue in many manufacturing industries. While the adoption of mixed-model assembly lines enables high product variety, it also makes system more complex as variety increases. This paper proposes an information entropy-based methodology that quantifies and then minimizes complexity through sequencing. The theory feasibility is demonstrated series simulations to showcase impact sequencing controlling...
Traditional animal identification methods such as ear-tagging, ear notching, and branding have been effective but pose risks to the scalability issues. Electrical offer better tracking monitoring require specialized equipment are susceptible attacks. Biometric using time-immutable dermatoglyphic features muzzle prints iris patterns is a promising solution. This project explores cattle 4923 images collected from 268 beef cattle. Two deep learning classification models implemented - wide...
Abstract Prosthetic devices remain unaffordable to many patients in resource-constrained environments. The design process of personalized prosthetic is sophisticated and requires specialized equipment trained professionals who may have limited availability. situation further exacerbated by the fact that resources are frequently situated away from underserved communities, compounding issue accessibility. Lowering barriers accessing required minimizing involvement designing manufacturing can...
In this paper, an agent-based approach to demographic research is introduced. The simulation expected provide a more open-endedmodeling framework. While microsimulations are widely used in studies, it lacks prescriptive capabilities of analyzing micro- interactions among individuals and environs. presented model dynamic solution for which difficult microsimulations. Keyword: Agent based modeling, data-driven simulation, social microsimulation
Many manufacturers have made extensive efforts to enhance the flexibility of their manufacturing systems, which can produce various products in small volumes with limited resources. For instance, automotive industry, maintaining a high diversity models and options is key objective business guarantee market competitiveness. However, this diversification frequently causes dramatic increase complexity mixed-model assembly system. Nevertheless, quantitative indices for are relatively...
Quality assessment in many production processes typically relies on manual inspections due to a lack of reference data and an effective method classify defects systematic way. Recently, the real-time, automated approach for product quality has been regarded important aspect smart manufacturing applications, such as automotive industry. In this research, we develop implement self-evolving system based adaptive support vector machine (ASVM) model real system. An process is feedback control...
The quality monitoring and control (QMC) has been an essential process in the manufacturing industries to ensure product reliability. With advancements big-data analytics, machine-learning based QMC become more popular various industries, such as automotive electronic companies. At same time, cost effectiveness (CE) of is perceived a main decision criterion that explicitly accounts for inspection efforts direct relationship with capability. In this paper, integrated support vector machine...