Scholarly Works, Industrial and Systems Engineering

Permanent URI for this collection

https://hdl.handle.net/10919/24292

Research articles, presentations, and other scholarship

Browse

Now showing 1 - 20 of 455

High-Quality Dataset-Sharing and Trade Based on A Performance-Oriented Directed Graph Neural Network
Zeng, Yingyan; Zhou, Xiaona; Chilukuri, Premith; Lourentzou, Ismini; Jin, Ran (2025)
The advancement of Artificial Intelligence (AI) models heavily relies on large high-quality datasets. However, in advanced manufacturing, collecting such data is time-consuming and labor-intensive for a single enterprise. Hence, it is important to establish a context-aware and privacy-preserving data sharing system to share small-but-high-quality datasets between trusted stakeholders. Existing data sharing approaches have explored privacy-preserving data distillation methods and focused on valuating individual samples tied to a specific AI model, limiting their flexibility across data modalities, AI tasks, and dataset ownership. In this work, we propose a performance-oriented representation learning (PORL) framework in a Directed Graph Neural Network (DiGNN). PORL distills raw datasets into privacy-preserving proxy datasets for sharing and learns compact meta data representations for each stakeholder locally. The meta data will then be used in DiGNN to forecast the AI model performance and guide the sharing via graph-level supervised learning. The effectiveness of the PORL-DiGNN is validated by two case studies: data sharing in the semiconducting manufacturing network between similar processes to create similar quality defect models; and data sharing in the design and manufacturing network of Microbial Fuel Cell anodes between upstream (design) and downstream (Additive Manufacturing) stages to create distinct but related AI models.
Data-Driven Sample Average Approximation with Covariate Information
Kannan, Rohit; Bayraksan, Guezin; Luedtke, James R. (INFORMS, 2025-01-06)
We study optimization for data-driven decision-making when we have observations of the uncertain parameters within an optimization model together with concurrent observations of covariates. The goal is to choose a decision that minimizes the expected cost conditioned on a new covariate observation. We investigate two data-driven frameworks that integrate a machine learning prediction model within a stochastic programming sample average approximation (SAA) for approximating the solution to this problem. One SAA framework is new and uses leave-one-out residuals for scenario generation. The frameworks we investigate are flexible and accommodate parametric, nonparametric, and semiparametric regression techniques. We derive conditions on the data generation process, the prediction model, and the stochastic program under which solutions of these data-driven SAAs are consistent and asymptotically optimal, and also derive finite sample guarantees. Computational experiments validate our theoretical results, demonstrate examples where our datadriven formulations have advantages over existing approaches (even if the prediction model is misspecified), and illustrate the benefits of our data-driven formulations in the limited data regime.
Residuals-based distributionally robust optimization with covariate information
Kannan, Rohit; Bayraksan, Guezin; Luedtke, James R. (Springer, 2023-09-26)
We consider data-driven approaches that integrate a machine learning prediction model within distributionally robust optimization (DRO) given limited joint observations of uncertain parameters and covariates. Our framework is flexible in the sense that it can accommodate a variety of regression setups and DRO ambiguity sets. We investigate asymptotic and finite sample properties of solutions obtained using Wasserstein, sample robust optimization, and phi-divergence-based ambiguity sets within our DRO formulations, and explore cross-validation approaches for sizing these ambiguity sets. Through numerical experiments, we validate our theoretical results, study the effectiveness of our approaches for sizing ambiguity sets, and illustrate the benefits of our DRO formulations in the limited data regime even when the prediction model is misspecified.
Optimization under uncertainty of a hybrid waste tire and natural gas feedstock flexible polygeneration system using a decomposition algorithm
Subramanian, Avinash S. R.; Kannan, Rohit; Holtorf, Flemming; Adams II, Thomas A.; Gundersen, Truls; Barton, Paul I. (Pergamon-Elsevier, 2023-12-01)
Market uncertainties motivate the development of flexible polygeneration systems that are able to adjust operating conditions to favor production of the most profitable product portfolio. However, this operational flexibility comes at the cost of higher capital expenditure. A scenario-based two-stage stochastic nonconvex Mixed-Integer Nonlinear Programming (MINLP) approach lends itself naturally to optimizing these trade-offs. This work studies the optimal design and operation under uncertainty of a hybrid feedstock flexible polygeneration system producing electricity, methanol, dimethyl ether, olefins or liquefied (synthetic) natural gas. A recently developed C++ based software framework (named GOSSIP) is used for modeling the optimization problem as well as its efficient solution using the Nonconvex Generalized Benders Decomposition (NGBD) algorithm. Two different cases are studied: The first uses estimates of the means and variances of the uncertain parameters from historical data, whereas the second assesses the impact of increased uncertain parameter volatility. The value of implementing flexible designs characterized by the value of the stochastic solution (VSS) is in the range of 260–405 M$ for a scale of approximately 893 MW of thermal input. Increased price volatility around the same mean results in higher expected net present value and VSS as operational flexibility allows for asymmetric exploitation of price peaks.
Design of PID controllers using semi-infinite programming
Turan, Evren Mert; Kannan, Rohit; Jäschke, Johannes (Elsevier, 2022)
The PID controller is widely used, and several methods have been proposed for choosing the controller parameters to achieve good performance. The controller tuning problem is set up as a semi-infinite program (SIP), with the integrated squared error (ISE) or the H∞ norm of the frequency domain error function (|𝐸(𝑠)|∞) as the objective function, and H∞ constraints for robustness and noise attenuation. Previous authors considered discrete points to enforce the H∞ constraints, however this is an outer approximation that does not guarantee a feasible point. When a feasible point can be found, it may require multiple iterations with a finer and finer discretisation. Here, the SIP is solved using a global optimisation algorithm. Several numerical experiments show that the proposed formulation converges quickly (<10 seconds) and gives sensible controller tuning values without the need to apply expert knowledge to the tuning problem. These results suggest that this is an attractive method for automated controller tuning.
A stochastic approximation method for approximating the efficient frontier of chance-constrained nonlinear programs
Kannan, Rohit; Luedtke, James R. (Springer, 2021-01-23)
We propose a stochastic approximation method for approximating the efficient frontier of chance-constrained nonlinear programs. Our approach is based on a bi-objective viewpoint of chance-constrained programs that seeks solutions on the efficient frontier of optimal objective value versus risk of constraints violation. To this end, we construct a reformulated problem whose objective is to minimize the probability of constraints violation subject to deterministic convex constraints (which includes a bound on the objective function value). We adapt existing smoothing-based approaches for chance-constrained problems to derive a convergent sequence of smooth approximations of our reformulated problem, and apply a projected stochastic subgradient algorithm to solve it. In contrast with exterior sampling-based approaches (such as sample average approximation) that approximate the original chance-constrained program with one having finite support, our proposal converges to stationary solutions of a smooth approximation of the original problem, thereby avoiding poor local solutions that may be an artefact of a fixed sample. Our proposal also includes a tailored implementation of the smoothing-based approach that chooses key algorithmic parameters based on problem data. Computational results on four test problems from the literature indicate that our proposed approach can efficiently determine good approximations of the efficient frontier.
Stochastic DC optimal power flow with reserve saturation
Kannan, Rohit; Luedtke, James R.; Roald, Line A. (Elsevier, 2020-12)
We propose an optimization framework for stochastic optimal power flow with uncertain loads and renewable generator capacity. Our model follows previous work in assuming that generator outputs respond to load imbalances according to an affine control policy, but introduces a model of saturation of generator reserves by assuming that when a generator's target level hits its limit, it abandons the affine policy and produces at that limit. This is a particularly interesting feature in models where wind power plants, which have uncertain upper generation limits, are scheduled to provide reserves to balance load fluctuations. The resulting model is a nonsmooth nonconvex two-stage stochastic program, and we use a stochastic approximation method to find stationary solutions to a smooth approximation. Computational results on 6-bus and 118-bus test instances demonstrate that by considering the effects of saturation, our model can yield solutions with lower expected generation costs (at the same target line violation probability level) than those obtained from a model that enforces the affine policy to stay within generator limits with high probability.
Convergence-order analysis of branch-and-bound algorithms for constrained problems
Kannan, Rohit; Barton, Paul I. (Springer, 2017-06-01)
The performance of branch-and-bound algorithms for deterministic global optimization is strongly dependent on the ability to construct tight and rapidly convergent schemes of lower bounds. One metric of the efficiency of a branch-and-bound algorithm is the convergence order of its bounding scheme. This article develops a notion of convergence order for lower bounding schemes for constrained problems, and defines the convergence order of convex relaxation-based and Lagrangian dual-based lower bounding schemes. It is shown that full-space convex relaxation-based lower bounding schemes can achieve first-order convergence under mild assumptions. Furthermore, such schemes can achieve second-order convergence at KKT points, at Slater points, and at infeasible points when second-order pointwise convergent schemes of relaxations are used. Lagrangian dual-based full-space lower bounding schemes are shown to have at least as high a convergence order as convex relaxation-based full-space lower bounding schemes. Additionally, it is shown that Lagrangian dual-based full-space lower bounding schemes achieve first-order convergence even when the dual problem is not solved to optimality. The convergence order of some widely-applicable reduced-space lower bounding schemes is also analyzed, and it is shown that such schemes can achieve first-order convergence under suitable assumptions. Furthermore, such schemes can achieve second-order convergence at KKT points, at unconstrained points in the reduced-space, and at infeasible points under suitable assumptions when the problem exhibits a specific separable structure. The importance of constraint propagation techniques in boosting the convergence order of reduced-space lower bounding schemes (and helping mitigate clustering in the process) for problems which do not possess such a structure is demonstrated.
The cluster problem in constrained global optimization
Kannan, Rohit; Barton, Paul I. (Springer, 2017-05-11)
Deterministic branch-and-bound algorithms for continuous global optimization often visit a large number of boxes in the neighborhood of a global minimizer, resulting in the so-called cluster problem (Du and Kearfott in J Glob Optim 5(3):253–265, 1994). This article extends previous analyses of the cluster problem in unconstrained global optimization (Du and Kearfott 1994; Wechsung et al. in J Glob Optim 58(3):429–438, 2014) to the constrained setting based on a recently-developed notion of convergence order for convex relaxation-based lower bounding schemes. It is shown that clustering can occur both on nearly-optimal and nearly-feasible regions in the vicinity of a global minimizer. In contrast to the case of unconstrained optimization, where at least second-order convergent schemes of relaxations are required to mitigate the cluster problem when the minimizer sits at a point of differentiability of the objective function, it is shown that first-order convergent lower bounding schemes for constrained problems may mitigate the cluster problem under certain conditions. Additionally, conditions under which second-order convergent lower bounding schemes are sufficient to mitigate the cluster problem around a global minimizer are developed. Conditions on the convergence order prefactor that are sufficient to altogether eliminate the cluster problem are also provided. This analysis reduces to previous analyses of the cluster problem for unconstrained optimization under suitable assumptions.
DynamicPrint: A physics-guided feedforward model predictive process control approach for defect mitigation in laser powder bed fusion additive manufacturing
Riensche, Alex; Bevans, Benjamin; Carrington Jr, Antonio; Deshmukh, Kaustubh; Shephard, Kamden; Sions, John; Synder, Kyle; Plotnikov, Yuri; Cole, Kevin; Rao, Prahalada (Elsevier, 2025-01-05)
In this work, we developed and applied a physics-guided autonomous feedforward model predictive process control approach called DynamicPrint to mitigate part defects in laser powder bed fusion (LPBF) additive manufacturing. Currently, the processing parameters for LPBF of a specific material are optimized through empirical testing of simple-shaped coupons. These optimized parameters are typically maintained constant when printing complex parts. However, using constant parameters often causes uneven temperature distribution in complex parts, leading to such defects as inhomogeneous microstructure, poor surface finish, thermal-induced distortion, and build failures. By contrast, DynamicPrint autonomously adjusts the processing parameters layer-by-layer before an LPBF part is printed to prevent non-uniform temperature distribution and mitigate thermal-induced defects. The a priori process parameter adjustments in DynamicPrint are guided by rapid physics-based thermal simulations. Through experiments with complex stainless steel 316 L LPBF parts, we demonstrate the following beneficial outcomes of DynamicPrint: (i) homogenous grain sizes and consistent properties (microhardness); (ii) improved geometric accuracy and surface integrity of hard-to-access internal features; and (iii) avoidance of recoater crashes and elimination of supports in parts with prominent overhang features. DynamicPrint can greatly accelerate the time-to-market for LPBF parts by offering a rapid, physics-based method for process qualification, unlike the current cumbersome and expensive empirical build-and-test approach.
Black Representation and District Compactness in Southern Congressional Districts
Goedert, Nicholas; Hildebrand, Robert; Pierson, Matthew; Travis, Laurel; Fravel, Jamie (2024-04-01)
This paper explores the assumed trade-off between district compactness and Black representation in legislative districts in the American South. We perform analysis both on heuristically generated districts using current US demographics, and on historical congressional maps since the 1970s. Computations are performed using an iterative heuristic to find feasible solutions guided by multiple objectives. We find that while the trade-off has been strongly observed historically, it is possible to effectively address both goals simultaneously in most cases. We are able to demonstrate maps substantially superior to the present enacted maps on both dimensions in at least seven of nine states analyzed. Nevertheless, the trade-off appears more necessary in states with larger and/or more heavily rural Black populations than in more urbanized states, where the drawing of compact Blackinfluence districts is easier.
Analyzing multiple-source water usage patterns and affordability in rural central Appalachia
Dudzinski, Emerald; Ellis, Kimberly P.; Krometis, Leigh-Anne H.; Albi, Kate; Cohen, Alasdair (2024-07-18)
Nearly 500,000 American households lack complete plumbing, and more than 21 million Americans are reliant on public drinking water systems with at least one annual health-based drinking water violation. Rural, low-income, and minority communities are significantly more likely to be burdened with unavailable or unsafe in-home drinking water. Lack of access and distrust of the perceived quality of municipally supplied water are leading an increasing number of Americans to rely instead on less regulated, more expensive, and potentially environmentally detrimental water sources, such as roadside springs and bottled water. Previous research studies have stressed the importance of considering the economic burden of all water related expenditures including financial and non-financial water related costs; however, past examinations of water costs have primarily focused on municipal water supplies. We propose an economic model to consider the full economic burden associated with multiple-source water use by incorporating both direct costs (e.g., utility bills, well maintenance, bottled water purchase, payments for water hauling/delivery) and indirect water-related expenditures (e.g., transportation costs to gather water, productivity lost due to time spent collecting). Using data gathered from household surveys along with the economic model, this study estimates the economic burden from two case studies in rural Central Appalachia with persistent water quality concerns: (1) McDowell County, WV (n=15) and (2) Letcher and Harlan Counties, KY (n=9). All surveyed households (n=24) rely on multiple-source water to meet their needs, frequently citing their perception of unsafe in-home tap water. Bottled water was the most common choice for drinking water in both settings (92%, n=24), though roadside spring use was also prevalent in McDowell County, WV (53%, n=15). The results show that multiple-water source use is associated with a large economic burden. Households reliant primarily on bottled water as their drinking water source spent 12.3% (McDowell County, WV) and 5.6% (Letcher and Harlan Counties, KY) of their respective county’s median household income (MHI) on water related expenditures. Households reliant primarily on roadside springs as their drinking water source spent 11.8% (McDowell County, WV) of MHI on water related expenditures. Hence, the vast majority of participating households (92%, n=24) spend above the US water affordability threshold of 2% MHI. The application of this economic model highlights major water affordability concerns in water insecure Appalachian communities and provides a foundation for future studies and enhancements.
Love at First Sight: Mere Exposure to Robot Appearance Leaves Impressions Similar to Interactions with Physical Robots
Hosseini, S. Maryam Fakhr; Hilliger, Samantha; Barnes, Jaclyn; Jeon, Myounghoon; Park, Chung Hyuk; Howard, Ayanna M. (IEEE, 2017-01-01)
As the technology needed to make robots robust and affordable draws ever nearer, human-robot interaction (HRI) research to make robots more useful and accessible to the general population becomes more crucial. In this study, 59 college students filled out an online survey soliciting their judgments regarding seven social robots based solely on appearance. Results suggest that participants prefer robots that resemble animals or humans over those that are intended to represent an imaginary creature or do not resemble a creature at all. Results are discussed based on social robot application and design features.
Head-up Displays Improve Drivers' Performance and Subjective Perceptions with the In-Vehicle Gesture Interaction System
Cao, Yusheng; Li, Lingyu; Yuan, Jiehao; Jeon, Myounghoon (Taylor & Francis, 2024-01-01)
In-vehicle infotainment systems can cause various distractions, increasing the risk of car accidents. To address this problem, mid-air gesture systems have been introduced. This study investigated the potential of a novel interface that integrates a Head-Up Display (HUD) with auditory displays (spearcons: compressed speech) in a gesture-based menu navigation system to minimize visual distraction and improve driving and secondary task performance. The experiment involved 24 participants who navigated through 12 menu items using mid-air gestures while driving on a simulated road under four conditions: HUD (with, without spearcons) and Head-Down Display (HDD) (with, without spearcons). Results showed that the HUD condition significantly outperformed the HDD condition in participants’ level 1 situation awareness, perceived workload, menu navigation performance, and system usability. However, there were trade-offs on visual fixation duration on the menu, and lane deviation. These findings will guide future research in developing safer and more effective HUD-supported in-vehicle gesture interaction systems.
An Overview of the 3rd International Workshop on eXtended Reality for Industrial and Occupational Supports (XRIOS)
Cho, Isaac; Kim, Kangsoo; Han, Dongyun; Bayro, Allison; Jeong, Heejin; Kim, Hyungil; Moon, Hayoun; Jeon, Myounghoon (IEEE, 2024)
The 3rd International Workshop on the eXtended Reality for Industrial and Occupational Supports (XRIOS) focuses on identifying the present advancements in XR research, particularly in the realms of human factors and ergonomics, as they apply to industrial and occupational tasks. The workshop also aims to explore potential future research directions. XRIOS was held for the first time at IEEE VR 2022, where it served as the first venue for building an interdisciplinary research community that bridges XR developers/practitioners and human factors and ergonomics researchers interested in industrial and occupational XR applications. XRIOS 2024, marking the first in-person workshop, follows the successes of XRIOS 2022 and 2023 in response to society's growing needs by expanding the XRIOS community and enhancing opportunities for engagement and collaboration.
Promoting STEAM education and AI/robot ethics in a child- robot theater afterschool program
Dong, Jia; Mitchell, Jennifer; Yu, Shuqi; Harmon, Madison; Holstein, Alethia; Shim, Joon Hyun; Choi, Koeun; Zhu, Qin; Jeon, Myounghoon (ACM, 2024)
A nine-week robot theater afterschool program was conducted in an elementary school to promote Science, Technology, Engineering, Arts, and Mathematics (STEAM) education using various Artificial Intelligence (AI) tools and social robots. In particular, the program aims to explore children's perceptions of the ethical implications of AI/robots in education. As a result, children showed excitement towards learning and interacting with AI/robots. Children's responses to ethical topics were also valuable as they expressed empathy towards robots and thought the developer should be responsible for negative situations, such as bullying. The present program will expand the knowledge of AI/robot ethics in early education.
The Effects of Whole-Hand Interactions with One Fingertip Vibrotactile Feedback on Cooperative VR Game Experience and Performance
Moon, Hye Sung; Moon, Hayoun; Orr, Grady; Jeon, Myounghoon (MIT Press, 2023-12-01)
New technologies have recently advanced user experiences in virtual reality (VR), whereas full sensation of diverse modalities has been not achieved yet. If any, haptic feedback has been delivered via bulky gloves. We have developed a novel thimble device that can deliver vibrotactile feedback via one fingertip. With this device, in the present study we investigated the effects of interaction methods and vibrotactile feedback on users’ social presence, presence, engagement, workload, and performance in a cooperative VR game. Twenty-six participants wearing a VR headset played a cooperative VR game with the experimenter under four conditions: (1) controllers with no vibrotactile feedback, (2) controllers with vibrotactile feedback, (3) hand tracking with no vibrotactile feedback, and (4) hand tracking with vibrotactile feedback. Results showed that hand tracking improved participants’ presence, engagement, and perceived workload compared to the traditional VR controllers. Also, vibrotactile feedback enhanced presence. However, the VR controllers outperformed the hand tracking interactions in completion time. The usability of hand interactions with vibrotactile feedback shows a promising result. We discuss the trade-offs between user experience and performance of the interaction methods and the potential of vibrotactile feedback in the VR environment.
Neurodivergence in Sound: Sonification as a Tool for Mental Health Awareness
Nadri, Chihab; Al Matar, Hamza; Morrison, Spencer; Tiemann, Allison; Song, Inuk; Lee, Tae Ho; Jeon, Myounghoon (International Community for Auditory Display, 2023-06)
The need to build greater mental health awareness as an important factor in decreasing stigma surrounding individuals with neurodivergent conditions has led to the development of programs and activities that seek to increase mental health awareness. Using a sonification approach with neural activity can effectively convey an individual’s psychological and mental characteristics in a simple and intuitive manner. In this study, we developed a sonification algorithm that alters existing music clips according to fMRI data corresponding to the salience network activity from neurotypical and neurodivergent individuals with schizophrenia. We conducted an evaluation of these sonifications with 24 participants. Results indicate that participants were able to differentiate between sound clips stemming from different neurological conditions and that participants gained increased awareness of schizophrenia through this brief intervention. Findings indicate sonification could be an effective tool in raising mental health awareness and relate neurodivergence to a neurotypical audience.
Child-robot musical theater for STEAM education
Choi, Koeun; Yu, Shuqi; Dong, Jia; Kim, Jisun; Lee, Yeaji; Devanshu, Vajir; Haines, Chelsea; Newbill, Phyllis; Upthegrove, Tanner; Wyatt, Andrea; Jeon, Myounghoon (2022-04-21)
Increasing Driving Safety and In-Vehicle Gesture-Based Menu Navigation Accuracy with a Heads-up Display
Cao, Yusheng; Li, Lingyu; Yuan, Jiehao; Jeon, Myounghoon (ACM, 2022)
More and more novel functions are being integrated into the vehicle infotainment system to allow individuals to perform secondary tasks with high accuracy and low accident risks. Mid-air gesture interactions are one of them. The current paper will present novel designs to solve a specific issue with this method of interaction: visual distraction within the car. In this study, a Heads-up display (HUD) will be integrated with a gesture-based menu navigation system to allow drivers to see menu selections without looking away from the road. An experiment will be conducted to investigate the potential of this system in improving drivers' overall safety and gesture interaction accuracy. The experiment will recruit 24 participants to test the system. Participants will provide subjective feedback about the directions for conducting future research and improving the overall experience, as well as objective performance data.

Browse

Recent Submissions