AHFE Open AccessHuman Factors in Simulation, Software and Systems Engineering
Table of Contents
Designing a Generative AI–Supported Modular Quotation Process for SMEs: A Design Science Research Approach
Small and medium-sized enterprises (SMEs) remain structurally constrained by limited resources, insufficient digital maturity, and a strong dependence on tacit knowledge embedded in individual employees. These constraints become particularly visible in quotation and offer creation, which is often manual, weakly structured, and characterized by information gaps between customer requirements and internal assessment capabilities. This study explores how generative artificial intelligence (GenAI) can support the redesign of SME quotation processes through modularity principles to improve transparency, flexibility, and cognitive ergonomics during early requirements work. Following Design Science Research Methodology, the study applies the stages of problem identification, objective specification, and artifact development. The resulting artifact is a modular quotation framework operationalized via a dual-role GenAI assistant: (1) a customer-facing chatbot that elicits and structures requirements through natural-language dialogue and (2) an internal analytical assistant that decomposes requests into reusable modules and supports initial feasibility and estimation work under human oversight. The artifact was iteratively evaluated through structured self-testing and practitioner feedback from roles in software development and IT project management. Feedback indicates improved requirement clarity and more systematic decomposition of complex requests, while also highlighting limitations related to file processing, budget realism, and interaction guardrails. Overall, the paper contributes a replicable blueprint for AI-enabled modular quotation support in SMEs and derives design implications for human-AI collaboration in software and systems engineering contexts.
Matthias Vogel, Tobias Schmallenbach, Giuseppe Strina
Open Access
Article
Conference Proceedings
Supporting Inspection of Structured Qualitative Team Task Analytic Data
One barrier to translating an effective team-based care process between clinical sites is that there are no comprehensive software systems to support process specification and comparison. As an incremental step to fill this gap, this work describes software requirements and design concepts to support the inspection and comparison of qualitative team process data across sites. The analysis can define labels for process attributes and valid values for them. The analyst can then encode data for one or more sites. Our custom macro-enabled Microsoft® Excel workbook presents reports with the values for each attribute as well as values by site. The reports support the analyst in fixing any differences created by how the data are encoded, as well as identifying true differences.
Ellen Bass, Tauseef Mamun, Benson Zhang, Junkai Ge, Meghan Lane-Fall
Open Access
Article
Conference Proceedings
Integration of MBSE Elements and Automation with System Development Processes for Advanced Performance & Efficiency
This research presents the culmination of a progressive study, detailing the results of integrating advanced Model Based Systems Engineering (MBSE) elements with system automation to enhance stakeholder selection processes. The importance of precise system selection remains paramount for optimal user safety and comprehension. By exploring the robust capabilities of Artificial Intelligence (AI), Machine Learning (ML), and automation tools, this study demonstrates significant improvements in developmental outputs over time. MBSE serves as a revolutionary methodology possessing complex capabilities that fundamentally elevate system performance and development. This methodology is successfully implemented through rigorous requirements writing, the formulation of architectural patterns, the establishment of comprehensive pattern libraries, and stringent verification processes. The strategic combination of these MBSE components, systems thinking, and automated intelligence functions in parallel to systematically improve selection processes for diverse users and stakeholders. Consequently, this final paper focuses on how these compounded systems engineering elements operate cohesively to guarantee that users select the most vital, beneficial systems tailored strictly to their preferences and operational needs. Furthermore, this study illustrates how the deployment of architectural patterns and pattern libraries seamlessly verifies requirements to output exceptionally performant architectures. Because modern architectures typically function as systems of systems requiring both high-level and low-level decomposition, this methodology efficiently promotes enhanced operational efficiency. While applicable across any diverse domain, this research specifically applies the refined framework to home security systems, definitively demonstrating the enhanced deliverables produced by merging MBSE, artificial intelligence, and advanced automation elements.
Daijha Hilliard, Bhushan Lohar, John Wade, Saeed Latif, Robert Cloutier
Open Access
Article
Conference Proceedings
Analysis for an End-to-End MBSE Operational Architecture
Model-Based Systems Engineering (MBSE) has shifted systems engineering toward digitally integrated architectures. However, methodological fragmentation among frameworks like UAF, ARCADIA, and MOFLT persists, characterized by ontological gaps and restricted interoperability. These constraints impede seamless End-to-End (E2E) lifecycle continuity across the SE “Vee” model. This research addresses these limitations by extending the E2E MBSE Framework, specifically formalizing the Operational Architecture (Ops-Arch) phase. Using a rigorous decomposition and reconstruction methodology, the Ops-Arch is synthesized by leveraging NASA standards and extracting high-fidelity artifacts from legacy frameworks. The development employs comparative analysis, reverse engineering of architectural patterns, and meta-model engineering to identify functional gaps and circularities. The resulting synthesis produces a comprehensive reference architecture that harmonizes disparate modeling blocks into a unified construct. This approach reduces practitioner cognitive load and facilitates the transition from abstract intent to concrete structural models. Key contributions include a cross-framework meta-model alignment, a semantic gap map, and a formalized E2E MBSE construct package anchored in a reference ontology. By bridging methodological heterogeneity, this framework ensures consistency in large-scale system design and enhances cross-disciplinary collaboration. This work demonstrates that dissimilar operational views can be systematically harmonized into a singular, precise architecture, improving both technical accuracy and human comprehension. Future research will evolve these constructs into executable architectures applied to a System of Interest (SoI) case study for integrated Verification and Validation (V&V).
Joshua Adelabu, Bhushan Lohar, John Wade, Sean Walker, Carlos Montalvo
Open Access
Article
Conference Proceedings
A method of increasing the electromagnetic immunity of the Spectrum Monitoring Sensor to UAV’s by using a shielded casing
The article presents a method for increasing the electromagnetic immunity of a spectrum monitoring sensor installed on an unmanned aerial vehicle (UAV) through the use of a specially designed shielded enclosure. The paper describes the design and implementation process of the shielded casing, focusing on limiting the influence of external electromagnetic disturbances on the sensor’s sensitive electronic components. The design solutions included the selection of materials with high shielding effectiveness over a wide frequency range, the application of filters and feedthroughs to suppress conducted interference, and the use of conductive gaskets to ensure electromagnetic continuity of the enclosure. The effectiveness of the proposed solutions was verified through laboratory tests of radiated and conducted emissions conducted in accordance with the MIL-STD-461G standard, including RE102 and CE102 measurements. Comparative tests performed for the sensor operating with and without the shielded enclosure demonstrated a significant reduction in emission levels in critical frequency ranges. The results confirm an improvement in electromagnetic compatibility and increased immunity of the sensor to interference generated by UAV onboard systems. The study shows that a properly designed shielded enclosure is an effective and practical approach to enhancing electromagnetic immunity of sensors used on UAV platforms and provides valuable guidelines for protecting other electronic devices operating in high-interference environments.
Rafał Przesmycki, Marek Bugaj, Jarosław Bugaj, Paweł Skokowski, Krzysztof Malon, Michał Kryk
Open Access
Article
Conference Proceedings
Annoyance Modeling in Cooperative Personnel Scheduling
Cooperative algorithmic systems that depend on human input seldom model users' cognitive or affective states, even though these states can influence data input quality and, consequently, computational results. This paper demonstrates how annoyance – a prototypical user state elicited when systems repeatedly request input – can be modeled in a scheduling context and integrated into an algorithmic optimization process. We consider an industrial job scheduling scenario in which a cooperative scheduling system coordinates employee access to a shared machine. Rather than requesting all availability information upfront, the system iteratively improves an initial suboptimal schedule by querying users across multiple interaction rounds, subject to individual availability constraints and the goal of minimizing operational costs. Such repeated interactions can cause annoyance to accumulate, potentially resulting in careless responding or cooperation breakdown. To investigate the implications of modeling annoyance in this context, we simulated interactions between users and the scheduling system. Simulated users followed behavioral rules based on individual availability profiles, with annoyance building up across interaction rounds until an individual threshold was exceeded, beyond which users stopped responding truthfully. Results from 45,360 simulation runs showed that integrating annoyance modeling makes the algorithm more robust to parameter variations and yields better performance under suboptimal parameter choices than optimizing for cost reduction alone. These findings demonstrate that engineering psychology and algorithm design can mutually benefit one another: interactive scheduling offers a domain for applied user experience research, while engineering psychology provides algorithm designers with concepts to better account for users’ willingness to cooperate with algorithmic systems.
Christiane Attig, Johannes Varga, Tim Schrills, Tobias Rodemann, Günther Raidl
Open Access
Article
Conference Proceedings
From Manual Patrols to Automated Detection: Leveraging Aerial Imagery, Computer Vision and Large Language Models for Wildfire Risk Mitigation
Wildfire risk around electrical transmission and distribution infrastructure has grown significantly because of climate change, vegetation encroachment, prolonged drought cycles, and extreme weather events. Electric utilities are required to inspect their electric assets on regular basis. The team traditionally rely on ground patrols, helicopter inspections, and manual review of aerial photographs to evaluate asset condition and detect burn indicators. While effective, these methods are time-consuming, labor-consuming, costly, and limited by human capacity, making frequent monitoring impractical at scale. This paper presents an integrated framework that uses aerial imagery, convolutional neural networks (CNNs), computer vision (CV) segmentation, and multimodal large language models (LLMs) to automate the detection of charring, scorch marks, vegetation encroachment, and other wildfire risk factors. The approach reduces manual inspection burdens, increases monitoring frequency, saves cost, and enables proactive wildfire mitigation.
Haranath Varanasi, Zining Yang
Open Access
Article
Conference Proceedings
Safety Predictive Model with Machine Learning and Its Application in DART Analysis in Utility
Workplace safety is critical for utilities and field operations, where complex work orders pose significant risks to employees and operations. Safety Predictive Model (SPM) is a data-driven solution that proactively identifies high-risk work orders and supports mitigation strategies. SPM uses enterprise data such as material group codes, work order types, location attributes, seasonal factors, and circuit details. Injury and Serious Injury/Fatality (SIF) records are linked to work orders to strengthen model accuracy. Historical data informs modelling, while new data validates performance. Over 90 variables are engineered for machine learning, with predictive strength assessed via Information Value. Algorithms tested include logistic regression, decision trees, SVM, and gradient boosting, with ensemble methods selected using ROC AUC and KS statistics. A composite risk score flags top deciles as high risk, applying district-specific thresholds. Beyond assessing upcoming work orders, SPM reveals key risk drivers, enabling utilities to anticipate and mitigate safety challenges effectively.
Alec Zhixiao Lin, Isaac Chen Fu
Open Access
Article
Conference Proceedings
Scalable Threat Detection in Customer Interactions Using LLMs and LLM-as-Judge Framework
This paper introduces a Customer Threat Detection Model leveraging a pre-trained large language model (LLM) on a major cloud platform to analyze customer service call transcripts and social media posts for potential security threats. The solution was developed in response to a critical need by the corporate security team to proactively identify threats during high-risk periods—such as the Southern California wildfires in January—when call volumes to the Customer Contact Center surged and employees and property faced elevated safety risks. Historically, manual identification of threats was slow and inconsistent, creating potential exposure for the organization. Operating in batch mode, the system processes daily calls and assigns each interaction a threat score (0–100), mapped to five ordinal bins from Low to High. The model combines expert-defined keywords with semantic embedding techniques to expand its threat lexicon, enabling detection of evolving language and context. Each transcript is transformed into a structured prompt and evaluated by the LLM to produce a threat score and category.Manual review sampled calls showed ~93% accuracy but proved resource-intensive and impractical for ongoing monitoring. To address scalability, we applied an “LLM-as-a-Judge” framework, where LLMs act as surrogate evaluators of model outputs. For 10K sampled calls, two summaries per call, overall and threat-focused, were generated and independently assessed by a second LLM to assign ordinal threat categories. Agreement metrics (accuracy, Cohen’s kappa, mean absolute difference), triadic consistency, and keyword sensitivity were computed. A small Keyword Influence Delta indicated strong contextual detection and guided keyword refinement.Results indicate good agreement between the deployed model and independent LLM judges, demonstrating scalability and reduced analyst workload in safety‑critical monitoring contexts.
Jonathan Presto
Open Access
Article
Conference Proceedings
Human Performance Modeling in Virtual Factories: A Simulation-Driven Ergonomics Approach
Manual workstation operations in manufacturing are traditionally evaluated based on cycle time and task completion, while ergonomic risks and detailed execution behaviors are rarely assessed within the same study. During early workstation planning, this disconnect leaves factors such as sequence deviations, tool use rhythm, shoulder loading, and excessive reach unquantified, hindering the alignment between engineering objectives and human factors requirements. This study proposes a multi view, vision based framework that captures both performance and ergonomic indicators during manual workstation tasks. A precision assembly workstation equipped with three synchronized cameras—one overhead and two lateral cameras—was used to observe three representative tasks: (T1) part picking and placement, (T2) tool-assisted fastening, and (T3) visual inspection. Video data from 18 operators across 324 trials (approximately 10 hours) were processed using RT-DETRv2 for object detection, OCSort for identity tracking, and a pose estimation module for upper-body kinematics. An ROI-based classifier was used to enhance fine-grained component recognition, while multi-view consistency enabled robust event-log generation. The proposed pipeline achieved stable event extraction with an event-level F1 score of approximately 0.88 at near-real-time processing speed. Derived indicators included sequence compliance, cycle time, tool-use rhythm, reach distance, and posture exposure. The results revealed distinct task characteristics: T1 exhibited the highest reach demand, T2 showed the highest shoulder-loading exposure, and T3 involved extended decision-making during inspection. Mixed-effects models confirmed significant task effects on both time-based and posture-related metrics (p < 0.01). Furthermore, a workstation redesign reduced excessive reach by 28%, arm elevation exposure by 19%, and mean cycle time by approximately 9%, demonstrating the value of multi-view vision sensing for ergonomics-informed workstation design.
Chunshih Cheng, Chia Chen Kuo, Chien-hsin YANG, YUJIE TSAI
Open Access
Article
Conference Proceedings
Optimization of Motion Capture Technology for a Human Digital Twin with Reduced Sensor Setups
In recent years, ergonomic research has gained increasing prominence, highlighting the importance of analysing and improving workstation design to better support human workers and enhance task performance. Motion Capture technologies are widely used for ergonomic assessment and Human Digital Twin development; however, their practical deployment in industrial environments is often limited by complex setups, high costs, and lengthy calibration procedures. This study proposes a joint-level optimization methodology to reduce MoCap sensor requirements keeping motion reconstruction accuracy and the reliability of ergonomic evaluations. The approach leverages the posture prediction capabilities of the Digital Human Modelling IPS IMMA platform and validated through both controlled benchmark experiments and a real industrial use case.
Manuela Vargas Gonzalez, Valerio Cibrario, Denise Tumiotto, Annalisa Bertoli, Cesare Fantuzzi
Open Access
Article
Conference Proceedings
Application of 3D Neck Modeling in Ergonomic Product Design
Neck pain is a common health problem and has had a significant impact worldwide. The products designed for the neck aim to provide healthcare functions. To fully realize their potential, they typically need to fit snugly against the body, and user comfort must be considered to ensure ergonomic design. Therefore, obtaining accurate anthropometric data is crucial for ergonomic product design. Traditional anthropometric methods are time consuming, labor intensive, and prone to error. With the continuous development of image processing and computer aided design and modeling technologies, people have been able to construct high precision and reliable 3D neck models. 3D modeling has become a key technology for developing human centered designs for the neck. As a versatile and essential domain, 3D neck modeling provides the fundamental basis for innovation in ergonomic design, healthcare, and safety engineering. This review follows the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) framework for systematic reviews and meta-analyses, analyzing research on 3D neck modeling published up to August 2025. This paper presents a systematic review of existing approaches to developing 3D neck models. It also examines the techniques used for processing and analyzing 3D data, including their limitations, and discusses the applications of these models in ergonomic product design. The results show that 3D neck modeling is increasingly used in product design, and its applications extend far beyond ergonomic product design. These models have become essential tools in many other professional fields. However, no single technique can fully capture the complexity of the neck; therefore, future work should combine multiple modeling techniques.
Xinyi Pan, Wenjing Yang
Open Access
Article
Conference Proceedings
Simulating Force-Posture Co-evolution in Horizontal Pushing Task using a Digital Human Model
Ergonomic analysis of a manual material handling task is essential to evaluate the musculoskeletal disorder risk involved. Digital Human Model (DHM) simulation is one of the techniques used to identify potential hazards. In the current method, the user provides force and posture, and subsequently, the simulation computes the joint stresses. In most simulations, the force and the posture are independent; therefore, they are mutually irresponsive. The work presented here argues that the responsive co-evolution of force and posture is essential for realistic performance assessment. It is known that musculoskeletal loading is affected by the force direction. Therefore, a wrong estimation of applied force direction could lead to an inaccurate assessment of joint stresses. In this work, first, we performed experiments to identify the unknown variables influencing the applied direction of force. Subsequently, based on the obtained data, a mathematical force model is developed that correlates the applied direction of the force with the position of the point of application of force and magnitude. The force model is then integrated into an existing DHM; this removes the need to provide the force direction manually. Using a physics-based object response model for a linear spring and box on a table experiencing static and dynamic friction, two illustrative task simulations that do not require any force specification are presented. The necessary exertions and associated biomechanical efforts can be derived from the result.
Dibakar Sen, Mayank Badola
Open Access
Article
Conference Proceedings
Enhancing the prediction accuracy of EKASTOS with individual parameter tuning
Whole-body vibration strongly influences perceived ride comfort, particularly in automated vehicles, where limited visual cues and unpredictable movements introduce a “surprise factor” that challenges postural stabilization. Conventional seat-to-head transmissibility assessments rely on simplified, linear assumptions that insufficiently represent multi-axis biomechanical responses and neglect inter-individual variability. More detailed and individualized modelling approaches are therefore required. This study, employs Ekastos, a computationally efficient 3D full-body dynamic model developed in Simscape MATLAB, to evaluate its ability to reproduce experimental anterior-posterior whole-body vibration responses across three distinct anthropometries with regards to body size. A hierarchical multi-objective evolutionary optimization framework is implemented to identify model postural control parameters and benchmarked against a gradient-based method. Average (response and anthropometry) seat-to-head frequency-response functions are first compared between experimental data and model responses obtained using the two optimization methods. Model performance is assessed using metrics for head, trunk, and pelvis motion in both X-translation and pitch, comparing model responses against (i) average and (ii) across minimum-maximum range of individual experimental responses. Afterwards, individualization is examined for two subjects by comparing (a) average-optimized postural control parameters with individualized anthropometry and (b) subject-specific postural control parameters with individualized anthropometry. Under average-response conditions, the multi-objective approach reduced objective metrics by 9-31%. Applying average anthropometry and parameters to individuals increased errors by 18–650% (>1000% in extreme cases), whereas anthropometry adaptation and subject-specific tuning reduced errors by up to 47%. These results highlight the necessity of robust optimization and individualization for accurate prediction of seated human dynamic responses under whole-body vibration.
Chrysovalanto Messiou, Riender Happee, Georgios Papaioannou
Open Access
Article
Conference Proceedings
Body Types of Chinese Adult Males
To understand the body characteristics of Chinese adult males, a sample of males aged 18–60 in China was selected. Using factor analysis, two common factors influencing male body types were extracted from 22 measurement items. Based on correlation analysis, five characteristic variables were selected. The k-means clustering method was then applied to classify Chinese adult male body types into six categories. This study on the analysis and classification of Chinese adult male body characteristics can provide a foundation for constructing different types of human body models.
Linghua Ran, Xin Zhang, Zhao Chaoyi, He Zhao, Zhongting Wang, Yue He
Open Access
Article
Conference Proceedings