Azadeh Haghighi

Dr. Azadeh Haghighi

Department of Mechanical and Industrial Engineering, University of Illinois Chicago, 842 W. Taylor Street, Chicago, IL 60607, USA

Research Focus & Metrics

Academic Career Length

6Years (2021-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
8 August 2022 in Sustainability
Verified
Journal Article
Open access
PDF file
MKMohammad Reza Khosravani
Mohammad Reza Khosravani
Azadeh HaghighiAzadeh Haghighi
Azadeh Haghighi
Additive manufacturing has drawn significant attention in both academia and industry due to its capabilities and promising potential in various sectors. However, the adoption of this technology in large-scale construction is still limited due to the numerous existing challenges. In this work, a comprehensive review of large-scale automated additive construction, its challenges, and emerging advances with a focus on robotic solutions and environmental sustainability is presented. The potential interrelations of the two topics are also discussed. A new classification scheme of available and emerging robotic solutions in automated additive construction is presented. Moreover, the vision of environmental sustainability is explored through three lenses: process, material, and printed large-scale structures/buildings. Finally, the current challenges and potential future directions are highlighted. The provided state of the art and challenges can be used as a guideline for future research on large-scale automated additive construction.

19 April 2023 in Journal of Manufacturing Science and Engineering
Verified
Journal Article
Closed access
PDF file
SGSuyog Ghungrad
Suyog Ghungrad
MFMeysam Faegh
Meysam Faegh
BGBenjamin Gould
Benjamin Gould
SWSarah J. Wolff
Sarah J. Wolff
Azadeh HaghighiAzadeh Haghighi
Azadeh Haghighi
Physics-informed deep learning (PIDL) is one of the emerging topics in additive manufacturing (AM). However, the success of previous PIDL approaches is generally significantly dependent on the existence of massive datasets. As the data collection in AM is usually challenging, a novel Architecture-driven PIDL structure named APIDL based on the deep unfolding approach for limited data scenarios has been proposed in the current study for predicting thermal history in the laser powder bed fusion process. The connections in this machine learning architecture are inspired by iterative thermal model equations. In other words, each iteration of the thermal model is mapped to a layer of the neural network. The hyper-parameters of the APIDL model are tuned, and its performance is analyzed. The APIDL for 1000 points with 80:20 split ratio achieves a testing mean absolute percentage error (MAPE) of 2.8% and R2 value of 0.936. The APIDL is compared with the artificial neural network, extra trees regressor (ETR), support vector regressor, and long short-term memory algorithms. It was shown that the proposed APIDL model outperforms the others. The MAPE and R2 of APIDL are 55.7% lower and 15.6% higher than the ETR, which had the best performance among other pure ML models.

1 August 2021 in Manufacturing Letters
Verified
Journal Article
Closed access
SGSuyog Ghungrad
Suyog Ghungrad
BGBenjamin Gould
Benjamin Gould
MSMojtaba Soltanalian
Mojtaba Soltanalian
SWSarah Jeannette Wolff
Sarah Jeannette Wolff
Azadeh HaghighiAzadeh Haghighi
Azadeh Haghighi

27 June 2022
Verified
Conference Paper
Closed access
SGSuyog Ghungrad
Suyog Ghungrad
BGBenjamin Gould
Benjamin Gould
SWSarah Wolff
Sarah Wolff
Azadeh HaghighiAzadeh Haghighi
Azadeh Haghighi
Prediction of the temperature history of printed paths in additive manufacturing is crucial towards establishing the process-structure-property relationship. Traditional approaches for predictions such as physics-based simulations are computationally costly and time-consuming, whereas data driven approaches are highly dependent on huge, labeled datasets. Moreover, these labeled datasets are mostly scarce and costly in additive manufacturing owing to its unique application domain (mass customization) and complicated data-gathering stage. Recently, model-based or physics-informed artificial intelligence approaches have shown promising potential in overcoming the existing limitations and challenges faced by purely analytical or data driven approaches. In this work, a novel physics-informed artificial intelligent structure for scenarios with limited data is presented and its performance for temperature prediction in the selective laser melting additive manufacturing process is compared with one of the state-of-the-art data driven approaches, namely long short-term memory (LSTM) neural networks. Temperature data for training and testing was extracted from infrared images of single-track layer-based experiments for Ti64 material with different combinations of process parameters. Compared to LSTM, the proposed approach has higher computational efficiency and achieves better accuracy in limited data scenarios, making it a potential candidate for real-time closed-loop control of the additive manufacturing process under limited and sparse data scenarios. In other words, the proposed model is capable to learn more efficiently under such scenarios in comparison to LSTM model.

18 October 2023
Verified
Book Chapter
Closed access
PDF file
Azadeh HaghighiAzadeh Haghighi
Azadeh Haghighi

17 December 2025 in Annals of Operations Research
Verified
Journal Article
Open access
PDF file
SGSuyog Ghungrad
Suyog Ghungrad
MMMassimo Molossi
Massimo Molossi
CAClarissa Amico
Clarissa Amico
RCRoberto Cigolini
Roberto Cigolini
Azadeh HaghighiAzadeh Haghighi
Azadeh Haghighi
Remanufacturing-as-a-Service promises scalable, lower-cost repair of metal components by coordinating distributed capabilities through a cloud platform. This paper formulates a quantitative, supply-chain optimization of cloud-enabled remanufacturing built on a hybrid additive–subtractive (ASM) repair cost model and evaluates three delivery scenarios: (i) centralized national hubs with hybrid ASM; (ii) collaborations with tier-one local facilities that own separate AM and SM equipment; and (iii) collaborations with tier-two local facilities that provide only partial capability, requiring multi-facility routing. Using a U.S. tooling and machinery case study, this paper quantitatively fuses stepwise ASM repair economics with strategic facility location under an explicit quality–cost trade-off, closing a gap between process modeling and supply-chain design in remanufacturing. Results show that across plausible freights and capital costs, decentralized collaborations remain within 5% of the cost optimum while improving regional responsiveness; longer planning horizons shift the near-optimal set toward 5–8 hubs, rather than a single national site. Sensitivity tests confirm that (a) higher transportation costs and longer horizons favor additional hubs, while (b) tighter accuracy thresholds steer volume to hybrid hubs or vetted tier-one partners. The model provides clear trigger points—based on accuracy targets, capital expenditures, and freight conditions—for deciding when to build new hubs versus contracting with local partners. It offers managers a robust tool to navigate the critical trade-offs between cost efficiency, service reach, and quality.