Industrial Engineering Commons^™

Diffog: Differentiable Policy Trajectory Optimization With Generalizability, Zhengtong Xu, Zichen Miao, Qiang Qiu, Zhe Zhang, Yu She

School of Industrial Engineering Faculty Publications

Imitation-learning-based visuomotor policies excel at manipulation tasks but often produce suboptimal action trajectories compared to model-based methods. Directly mapping camera data to actions via neural networks can result in jerky motions and difficulties in meeting critical constraints, compromising safety and robustness in real-world deployment. For tasks that require high robustness or strict adherence to constraints, ensuring trajectory quality is crucial. However, the lack of interpretability in neural networks makes it challenging to generate constraint-compliant actions in a controlled manner. This article introduces differentiable policy trajectory optimization with generalizability (DiffOG), a learning-based trajectory optimization framework designed to enhance visuomotor policies. By …

Exploring The Effectiveness Of Virtual Reality Learning Through Use Of Visual Eye-Tracking Analytics (Veta) And Biological Measurements, Mckinley Anne Sherman

Master's Theses

Virtual Reality (VR) offers an immersive and interactive platform for experiential learning. The purpose of this thesis was to evaluate the relationship between physiological responses and cognitive workload within a VR learning environment and to explore VR as an effective instructional tool. This research compared participant engagement, stress, and learning performance within a 6th-grade science module developed in VR by incorporating biometric data collected via Polar H10 heart rate monitor and Varjo Areo VR headset eye-tracking. Thirty-three participants completed a pre-lesson demographic survey, post-lesson survey, VR sickness questionnaire, and the NASA Task Load Index (NASA-TLX). While completing the lesson, the …

Intelligent Real-Time Flotation Froth Stability Monitoring Using Embedded Computer Vision: Lead (Pb) Processing Case Study, Ahmed Bendaouia, Taha Ismaili, Ismail Najib, Oussama Hasidi, Intissar Benzakour, Jianzhi Li, El Hassan Abdelwahed

Manufacturing & Industrial Engineering Faculty Publications

Froth stability in flotation refers to the ability of the froth layer to remain intact and avoid collapsing during the flotation process. A stable froth layer is crucial to maximizing mineral recovery and ensuring the efficient operation of flotation cells. Several factors influence the stability of the froth, including the depth of the froth, the crowding, the size of the bubbles, and the rate of movement of the froth. In this study, our main objective is to analyze and evaluate froth stability in real time, since it plays a critical role in process optimization for flotation-based mineral processing. We propose …

Use Of Polymer Fiber To Improve Mechanical Properties Of Hma Containing Recycled Asphalt Pavement (Rap), Ashraf Rahim, Shadi Saadeh, Hani Al Zaraiee

Mineta Transportation Institute

A great percentage of highways and roads in California are constructed with Hot Mix Asphalt (HMA), and, as California infrastructure ages, these highways and roads must be maintained and rehabilitated. Reclaimed Asphalt Pavement (RAP) is considered an excellent alternative to virgin (raw, unprocessed) materials because it reduces the use of virgin aggregate and binder. Also, the use of RAP decreases the amount of construction waste placed into landfills. This laboratory study investigated the effect of two different commercial polymer fibers on the mechanical properties of HMA with RAP. Three different HMA with RAP mixes were used in the study. One …

Process-Driven Manufacturability Constraints In Design For Wire Arc Additive Manufacturing, Monsuru Ramoni, Sampson Gholston, Albert E. Patterson

Manufacturing & Industrial Engineering Faculty Publications

Wire arc additive manufacturing (WAAM) has emerged as a useful option for large-scale metal additive manufacturing. It has gained widespread use in the aerospace industry and other applications that require large and complex custom parts. The process is a combination of a precise control system and a welding process, typically GTAW or MIG welding. It is a member of the directed energy deposition (DED) family of AM processes. Compared with many metal AM processes, it is relatively simple and cost-effective but almost always requires a significant amount of postprocessing before parts can be used. The AM-based nature of the process …

Tacscope: A Miniaturized Vision-Based Tactile Sensor For Surgical Applications, Md Rakibul Islam Prince, Sheeraz Athar, Pokuang Zhou, Yu She

School of Industrial Engineering Faculty Publications

The lack of tactile feedback in robot-assisted minimally invasive surgery (RMIS) limits surgeons’ ability to palpate tissues, a critical technique for locating abnormalities such as tumors. To address this challenge, we introduce TacScope, a novel, vision-based tactile sensor leveraging the magnification properties of a spherical-surface elastomer to provide tactile feedback for advanced clinical applications. TacScope features a robust, low-cost, and easyly fabricate design, enabling seamless integration into surgical robotic setups. It reconstructs high-resolution 3D geometry from variations in particle-density distribution across its elastomer surface, requiring only a single image for calibration. The curved elastomer membrane alters particle-density distribution under …

Multi-Material Printer Development, Jessica Hu, Phoebe Eplett

College of Engineering Summer Undergraduate Research Program

Surgeons require pre-surgical models to practice procedures and improve success rates. However, the current process of creating physical models for practice using 3D printing is limited to producing homogeneous mimics of the human body, while the human body is a heterogeneous structure composed of multiple materials with varying properties, including bones, muscles, and skin. Additive manufacturing methods cannot easily print multiple materials simultaneously, as current AM machines only print homogeneous material, unlike the human body. In this proposal, we aim to develop a multi-material additive manufacturing system, from software to hardware, to manufacture pre-surgical models. We will investigate multi-material AM …

Evaluating The Accuracy Of Gen Ai Detecting Misinformation, Alan Sebastian

College of Engineering Summer Undergraduate Research Program

This research project will investigate the ability of advanced Large Language Models (LLMs) to identify and assess misinformation across diverse forms of media, including text, images, and video. In an age where misleading content spreads rapidly across digital platforms, evaluating the reliability and integrity of AI systems tasked with fact-checking is critical. We will develop a comprehensive dataset composed of factual and misleading examples drawn from various well-known and reliable fact-checking organizations. Each item will be independently reviewed and transparently labeled to ensure reproducibility. We will then prompt a curated group of state-of-the-art LLMs—including GPT-4, Claude, Gemini, Perplexity, Grok, and …

3d Printing Scoliosis Braces, Zak Neddo, Jaylan Mo

College of Engineering Summer Undergraduate Research Program

As 3D printing technology continues to transform the healthcare industry, the need for FDA-compliant additive manufacturing facilities in public universities has become increasingly important. While private companies and medical research institutions have successfully integrated 3D printing for medical device prototyping and surgical planning, public universities—including Cal Poly—lack the dedicated infrastructure needed to support FDA-regulated medical manufacturing and research. This project aims to bridge that gap by researching, designing, and implementing a dedicated FDA-compliant section within Cal Poly’s multi-purpose 3D Printing Facility. Over the course of eight weeks, student researchers will study FDA regulations, design an optimized facility layout, and implement …

Deepseek, Chatgpt, Or Gemini? A Multi-Method Investigation Of Neural And Behavioral User Experience (Ux), Keziah Gopalla

College of Engineering Summer Undergraduate Research Program

As artificial intelligence tools become integral to everyday tasks, understanding how users interact with these systems is essential for improving user experience and system design. This research aims to investigate and compare the interface usability and emotional responses elicited by three prominent AI tools—ChatGPT, DeepSeek, and Google Gemini—using the Emotiv Insight EEG headset. By combining usability testing with emotional biometrics, this study offers a novel approach to evaluating conversational AI systems. The study will capture both subjective usability metrics and objective emotional markers such as arousal, valence, and engagement. Participants will complete standardized tasks using each AI tool, while their …

Resilience Engineering Via Bifurcation And Ecological Network Analysis: Demonstrated In An Electric Power Case Study, Rogelio Gracia Otalvaro

Doctoral Dissertations and Master's Theses

Modern systems are increasingly complex, interconnected cyber-physical systems that combine digital controls with physical infrastructure. This integration, along with the constant introduction of new technologies and actors into the network, enables reliable operation but introduces vulnerabilities to unexpected and varied disruptions and cascading failures, making resilience a critical concern. Traditional risk management and resilience assessment methods often struggle with the nonlinearity and dynamic behavior of these systems. This dissertation proposes a novel approach combining Bifurcation Analysis (BA) and Ecological Network Analysis (ENA) to enhance the understanding and improvement of system resilience. BA, a mathematical method from dynamical systems theory, is …

Robotic System With Tactile-Enabled Leaf Tracking For High-Resolution Hyperspectral Imaging Device For Autonomous Corn Leaf Phenotyping In Controlled Environments, Xuan Li, Ziling Chen, Raghava Uppuluri, Pokuang Zhou, Tianzhang Zhao, Darrell Zachary Good, Yu She, Jian Jin

School of Industrial Engineering Faculty Publications

Hyperspectral imaging of individual corn leaves provides valuable data for analyzing nutrient content and diagnosing diseases. However, existing leaf-level imaging techniques face challenges such as low spatial resolution and labor-intensive processes. To address these limitations, this study developed a robotic system integrated with a high-resolution line-scanning hyperspectral imaging device to autonomously scan a corn leaf. The hyperspectral imaging device used a vision-based tactile sensor for active leaf tracking throughout the scanning process, ensuring high image quality. Additionally, the device incorporated an in-hand leaf manipulation mechanism that ensured the leaf was properly positioned on the tactile sensing area at the start …

Transparent Eeg Analysis: Leveraging Autoencoders, Bi-Lstms, And Shap For Improved Neurodegenerative Diseases Detection, Badr Mouazen, Ahmed Bendaouia, Omaima Bellakhdar, Khaoula Laghdaf, Aya Ennair, El Hassan Abdelwahed, Giovanni De Marco

Manufacturing & Industrial Engineering Faculty Publications

Highlights

• Novel hybrid architecture: Combined autoencoders with bidirectional LSTM networks for enhanced EEG signal classification, achieving 98% accuracy in distinguishing AD, FTD, and healthy controls.

• Explainable AI integration: Implemented SHAP (SHapley Additive exPlanations) framework to enhance model transparency and identify entropy as the most influential feature for neurodegenerative disease detection.

• Optimal temporal segmentation: Demonstrated that 5-s EEG windows with 50% overlap provide the best balance between classification accuracy and computational efficiency.

• Comprehensive feature extraction: Utilized Power Spectral Density (PSD) analysis across standard frequency bands (Delta, Theta, Alpha, Beta, Gamma) following autoencoder-based dimensionality reduction.

• Superior performance validation: Outperformed traditional machine learning …

  ---

  Go to article

Using Statistical Clustering Of Trajectory Data To Support Analysis Of Subject Movement In A Virtual Environment, Martin Galicia Avila, Douglas Timmer, Alley C. Butler

Manufacturing & Industrial Engineering Faculty Publications

Gracia de Luna conducted experiments with an HMD virtual environment in which human subjects were presented with surprise distractions. His collected data for head, dominant hand, and non-dominant hand included 6 DOF human subject trajectories. This paper examines this data from 57 human subject responses to those surprise virtual environment distractions using statistical trajectory clustering algorithms. The data is organized and processed with a Dynamic Time Warping (DTW) algorithm and then analyzed using the Density Based Spatial Clustering (DBSCAN) algorithm. The K-means method was used to determine the appropriate number of clusters. Chi Squared goodness of fit was used to …

Enhanced Global Oil Spill Dataset From 1967 To 2023 Based On Text-Form Incident Information, Yiming Liu, Zhuoli Yin, Hua Cai

The School of Sustainability Engineering and Environmental Engineering (SEE) Faculty Publications

Knowing how much oil was released into the environment in each incident is critical to studying the environmental, ecological, and economic impacts of oil spills. However, the release amounts (RAs) for numerous oil spill incidents remain unavailable in a structured format for large-scale analysis. The most extensive global oil spill database, managed by NOAA’s Office of Response and Restoration, only documents the worst-case scenario estimations in the machine-readable dataset, while more accurate values are embedded within unstructured incident descriptions and subsequent updates. To enhance the dataset with more accurate RAs, we developed a framework to extract the actual RAs from …

Scalable And Adaptive Agile Framework For Semiconductor Foundry: Advanced Packaging And Heterogeneous Integration Perspective, Pravin Thorat

Harrisburg University Dissertations and Theses

This research addressed the critical requirement for a scalable and adaptive agile framework specifically designed for the unique demands of semiconductor foundries specializing in advanced packaging and heterogeneous integration (HI). The semiconductor industry was encountering growing pressure to innovate and respond quickly to rapidly evolving demands, yet traditional manufacturing processes often struggled to adapt. Existing agile frameworks, mainly developed for the software industry, lacked the necessary adaptations to address the complexities of semiconductor manufacturing, including extended lead times, high capital investment, rigorous quality requirements, and the integration of various technologies. This research gap hindered the ability of semiconductor foundries to …

Transforming Movement Assessment In Physical Therapy Through Subaquatic Data Collection, Kaitlyn Mcdonald

Theses and Dissertations

This study explored the interest and perceived barriers to integrating subaquatic diagnostic technologies (SDTs) into hydrotherapy among licensed physical therapists. Seventeen semi-structured interviews were conducted using a mixed-methods design, with 15 interviews included in the final analysis. Quantitative data were analyzed using chi-square tests, while qualitative responses were coded thematically. Results indicated no statistically significant relationships between SDT interest and career stage or hydrotherapy access, though qualitative data highlighted concerns about cost, limited access, and usability. Despite mixed interest in adoption, participants identified several potential benefits of SDTs, including improved treatment tailoring, increased patient buy-in, and enhanced outcome monitoring. Functional …

Neural Architecture Search-Driven Unsupervised Domain Adaptation For Enhanced Wood Chip Quality Evaluation In Forest Industries, Abdur Rahman

Theses and Dissertations

Reliable and efficient measurement of wood chip moisture content is crucial for forest-reliant industries, including biofuel production, pulp and paper manufacturing, and bio-refineries, as it directly influences product quality and energy output. Traditional methods like the oven-drying technique, despite their widespread use, are time-consuming and impractical for real-time applications, while modern alternatives such as NIR spectroscopy, electrical capacitance, and microwave analysis are often costly, less portable, and require specialized expertise. This dissertation addresses these limitations by leveraging deep learning and machine vision to develop a scalable, accurate, and portable method for moisture content measurement using RGB images of wood chips. …

Optimization And Simulation Models For Integrating Social Determinants In Assessing And Optimizing The Accessibility And Efficiency Of Community Pharmacy Networks., Md Morshedul Alam

Electronic Theses and Dissertations

Community pharmacies are essential components of the healthcare system, serving not only as points of medication access but also as providers of health consultations and preventive care. However, disparities in pharmacy access persist due to a combination of geographic, socioeconomic, and behavioral factors. This dissertation develops a comprehensive methodological framework to evaluate and improve accessibility to community pharmacies, integrating optimization, choice modeling, and discrete event simulation. First, we develop an Optimal Community-Pharmacy Assignment (OCPA) model to quantify geographic accessibility across a regional population network. The model minimizes risk-adjusted travel distances between population centers and pharmacies, incorporating social determinants of health …

Multi-Stage Stochastic Programming For Disaster Relief Logistics Under Forecast Uncertainty, Sudhan Bhattarai

All Dissertations

Hurricanes are among the most destructive annual disasters in the United States, presenting interdependent challenges for evacuation planning and relief-supply logistics. Coordinating evacuation and relief operations is crucial to ensure the timely and effective movement of at-risk populations and the delivery of essential supplies. This dissertation develops and evaluates three progressively advanced multi-stage stochastic programming (MSSP) frameworks that integrate evacuation and relief-item pre-positioning while explicitly accounting for uncertainty in hurricane forecasts.

Chapter 1 introduces a fully adaptive MSSP model for the integrated hurricane relief logistics and evacuation planning (IHRLEP) problem. The model simultaneously optimizes evacuation flows and inventory pre-positioning over …