About

Professor Nii O. Attoh-Okine is the Chair of Civil and Environmental Engineering at the University of Maryland's A. James Clark School of Engineering. His research expertise is in resilient infrastructure, with a strong focus on cybersecurity and digital technologies applied to transportation and civil infrastructure. He is a well-recognized expert in railway engineering and leads efforts in digital twins, cybersecurity, and blockchain technologies applied to civil engineering. Professor Attoh-Okine is a member of the National Academy of Science, Engineering and Medicine committee on Safe Transportation of Liquefied Natural Gas by Railroad Tank Car. He holds a Ph.D. from the University of Kansas and a Diplom Ingenieur from Rostov Institute of Civil Engineering. His background includes a joint appointment in electrical and computer engineering at the University of Delaware, where he was also the Interim Cybersecurity Initiative Academic Director. He is a member of the State of Delaware Cybersecurity Council and has served as the lead representative from the university to the Digital Innovation Hub, a US-Japan University Consortium promoting research in digital innovation across multidisciplinary topics. In recognition of his leadership, he was selected as one of only 18 international speakers to attend the G7 Ministerial Meeting held in Japan in 2019. Professor Attoh-Okine is a founding associate editor of the ASCE/ASME Journal of Risk and Uncertainty Management in Engineering Systems and has served as an associate editor for several other prominent journals. He has authored two books, 'Resilience Engineering: Models and Analysis and Big Data' and 'Differential Privacy in Railway Track Engineering.' His research encompasses railway track analysis, resilience engineering, cyber resilience, and the application of digital technologies such as blockchain, digital twins, and topological data analysis to civil infrastructure. His work aims to enhance the safety, resilience, and digital transformation of transportation systems and civil infrastructure.

Research topics

• Artificial Intelligence
• Computer Science
• Machine Learning
• Statistics
• Knowledge management
• Chemistry
• Food science
• Medicine
• Data science
• Biochemistry
• Mathematics
• Mechanical engineering
• Engineering

Selected publications

• Improving Feasibility in Quantum Approximate Optimization Algorithm for Vehicle Routing via Constraint-Aware Initialization and Hybrid XY-X Mixing

  arXiv (Cornell University) · 2026-04-08

  articleOpen accessSenior author

  The Quantum Approximate Optimization Algorithm (QAOA) is a leading framework for quantum combinatorial optimization. The Vehicle Routing Problem (VRP), a core problem in logistics and transportation, is a natural application target, but it poses a major feasibility challenge for standard QAOA because feasible solutions occupy only a tiny fraction of the search space, and the conventional Pauli-$X$ mixer can disrupt partial solution structures that satisfy key local constraints. To address this issue, we propose a constraint-aware QAOA framework with two complementary components. First, we design a lightweight initialization strategy that encodes a selected subset of simple yet informative local one-hot constraints into the initial state, thereby reducing the initial superposition space and increasing the probability mass on states with important local structure. Second, we introduce a hybrid XY-$X$ mixer that preserves the constraint structure imposed at initialization while retaining exploratory flexibility over the remaining unconstrained degrees of freedom during QAOA evolution. We evaluate the proposed framework against standard QAOA under three progressively more realistic regimes: ideal statevector simulation, finite-shot sampling, and noisy finite-shot sampling. Across all regimes, the proposed method consistently achieves lower average energy and higher feasible-solution ratios than standard QAOA, indicating more effective guidance toward structurally valid, lower-cost VRP solutions. However, the performance gap narrows in the noisy regime. Because this setting adopts a hardware-inspired error model based on near-best-reported laboratory-level qubit gate and readout fidelities, the observed attenuation suggests that the practical advantage of the more structured mixer is likely to grow as quantum hardware improves and error rates decline.

  PublisherOA PDF

• Improving Feasibility in Quantum Approximate Optimization Algorithm for Vehicle Routing via Constraint-Aware Initialization and Hybrid XY-X Mixing

  arXiv (Cornell University) · 2026-04-08

  preprintOpen accessSenior author

  The Quantum Approximate Optimization Algorithm (QAOA) is a leading framework for quantum combinatorial optimization. The Vehicle Routing Problem (VRP), a core problem in logistics and transportation, is a natural application target, but it poses a major feasibility challenge for standard QAOA because feasible solutions occupy only a tiny fraction of the search space, and the conventional Pauli-$X$ mixer can disrupt partial solution structures that satisfy key local constraints. To address this issue, we propose a constraint-aware QAOA framework with two complementary components. First, we design a lightweight initialization strategy that encodes a selected subset of simple yet informative local one-hot constraints into the initial state, thereby reducing the initial superposition space and increasing the probability mass on states with important local structure. Second, we introduce a hybrid XY-$X$ mixer that preserves the constraint structure imposed at initialization while retaining exploratory flexibility over the remaining unconstrained degrees of freedom during QAOA evolution. We evaluate the proposed framework against standard QAOA under three progressively more realistic regimes: ideal statevector simulation, finite-shot sampling, and noisy finite-shot sampling. Across all regimes, the proposed method consistently achieves lower average energy and higher feasible-solution ratios than standard QAOA, indicating more effective guidance toward structurally valid, lower-cost VRP solutions. However, the performance gap narrows in the noisy regime. Because this setting adopts a hardware-inspired error model based on near-best-reported laboratory-level qubit gate and readout fidelities, the observed attenuation suggests that the practical advantage of the more structured mixer is likely to grow as quantum hardware improves and error rates decline.

  PublisherDOI

• Quantum VAE–Driven Predictive Framework with RAG–LLM Reasoning for Rail Track Geometry Maintenance Decisions

  SSRN Electronic Journal · 2025-01-01

  preprintOpen accessSenior author
  PublisherDOI

• Topological Data Analysis for Railway Track Geometry Safety and Maintenance

  Journal of Computing in Civil Engineering · 2025-07-11

  articleSenior author

  This paper explores the use of topological data analysis (TDA) to improve railway track geometry, focusing on enhancing safety and maintenance. It uses methods such as Betti numbers, homology, and the Mapper algorithm to understand the shape and connections within the geometry data. The focus of this paper is on persistent homology and the Mapper algorithm, which reveal consistent patterns and gaps in data. The research emphasizes TDA’s practical implications for railway maintenance and monitoring, advocating for its integration into railway engineering practices to address challenges and improve system performance. It points to future possibilities for using TDA in railway engineering, providing some suggestions and better tools needed to take full advantage of TDA’s benefits for railway safety.

  PublisherDOI

• Quantum Bayesian Inference for Rail Infrastructure Condition Assessment

  2025-08-30

  articleSenior author

  Bayesian networks (BNs) are increasingly used for modeling complex infrastructure systems under uncertainty because of their ability to enable probabilistic reasoning. In this work, the potential of Quantum Bayesian networks (QBNs) to infer infrastructure deterioration in a rail system was explored. QBNs are quantum versions of classical BNs that utilize quantum circuits and algorithms to encode prior and conditional dependencies among variables. In this case study, a hierarchical rail condition system comprising substructure and superstructure components such as ballast, subgrade, rails, and ties, along with environmental factors like temperature and moisture, is described using a BN. In particular, conditional relationships are defined using classical conditional probability tables and mapped onto a parameterized quantum circuit with the aid of quantum gates. The QBN was executed on IBM's quantum simulator, and inference is performed through repeated quantum sampling. The QBN was evaluated by comparing its marginal, joint, and conditional probability estimates against those obtained from an equivalent classical BN implemented using exact inference. Our results, visualized through probability distributions and bitstring statistics, show a high degree of consistency between quantum and classical models, with low variance across multiple runs. This study demonstrates the feasibility of using QBNs for interpretable and scalable infrastructure assessment.

  PublisherDOI

• Guest Editorial Introduction to the Special Issue on Cyber and Digital Information in Railway Engineering and Operation

  IEEE Transactions on Intelligent Transportation Systems · 2025-08-01

  article1st authorCorresponding
  PublisherDOI

• Hybrid Quantum Neural Network and Shapely Additive Explanations in Railway Track Geometry Modeling

  ASCE-ASME Journal of Risk and Uncertainty in Engineering Systems Part B Mechanical Engineering · 2024-12-24

  articleSenior author

  Abstract This research investigates the application of explainable quantum machine learning (QML) for predictive maintenance in the railroad industry. By utilizing ground-penetrating radar (GPR) data to characterize subsurface track conditions (ballast fouling index, ballast thickness index, layer roughness index, and moisture likelihood index), a quantum neural network (QNN) model was developed to predict track geometry (profile and alignment) defects in a Class 3 railroad track. Shapley additive explanations (SHAP) were employed to analyze the feature importance and the model’s decision-making processes to ensure model interpretability. The QNN model correctly predicted 42 out of 55 test data points. SHAP analysis identified the ballast fouling index and layer roughness index as the most important parameters, aligning with engineering expectations.

  PublisherDOI

• Multiway Analytics Applied to Railway Track Geometry and Ballast Conditions

  ASCE-ASME Journal of Risk and Uncertainty in Engineering Systems Part A Civil Engineering · 2024-10-29 · 1 citations

  articleSenior author

  Railroad systems generate large amounts of data, which, when effectively analyzed, can significantly enhance maintenance decisions to improve safety and system performance. Tensor decomposition, as an advanced multidimensional data analysis tool, offers unique advantages over traditional two-way matrix factorizations, such as the uniqueness of the optimal solution and component identification, even with substantial data missing. This paper introduces the basic concepts of tensor decomposition and specifically demonstrates its application in analyzing railway track geometry and subsurface conditions. By applying tensor analysis to multidimensional data sets, the study identifies critical patterns in track geometry and ballast conditions. Key findings indicate that tensor-based models can effectively predict track deformations and align maintenance schedules more accurately, thus optimizing repair operations and extending the lifespan of railway infrastructure.

  PublisherDOI

• Special Section on Digital Twins: A New Frontier in Critical Infrastructure Protection and Resilience

  ASCE-ASME Journal of Risk and Uncertainty in Engineering Systems Part B Mechanical Engineering · 2024-01-25

  articleOpen access1st authorCorresponding

  A digital twin (DT) is a computational model (or set of coupled) that evolves over time to persistently represent the critical structure, its components, system, or process. Digital twin underpins intelligent automation by supporting data-driven decision-making and enabling asset-specific analysis and system behavior. Within the context of critical infrastructure systems, the digital twins represent the flow of information among connected platforms. In the future, as many agencies turn to digital twin capabilities, they have to migrate toward continuous real-time performance models and calibrate by pairing data from real-time sensors, meters, weather, and other data. The digital twin can be used to run “what-if” scenarios, predict and prevent failures, provide early alerts of anomalies, and conduct predictive analysis. The strength of a digital twin is the interconnectivity of data and models. The main characteristics of a digital twin are:It is worth noting that digital twin technology and simulation are not the same. The DT technology is more dynamic and performs real-time updates of the virtual models. Simulation, which is more “static,” cannot perform any real-time updates of the virtual model. Once the input data are defined, there is no room for real-time updates. Therefore, DT technology provides more accurate behavior, including the system's performance over time. It is worth noting that a digital twin without a physical twin is a traditional model.With all the current research in critical infrastructure systems, a major missing element is the appropriate selection and use of graphical models, which govern the information and data exchange between the physical and the virtual model. The graphical model encodes the two connections between the physical and the virtual. Much research is needed in this critical area of DT implementation.This Special Section contains four papers. Badiru et al. discuss the modeling for critical infrastructure. The outcome of this model is connected to a climate variable. The work relies heavily on the work of COP26. The work has strong National Defense implications, and the authors did a fine work of proposing a new model that can serve as a blueprint in both systems engineering and address critical fracture behavior with climate as a driving factor.Yanik et al. applied DT technology as a verification and validation tool in rotating machinery. This approach has an extensive application in various industrial machines and equipment. Issues like fault diagnosis and prognosis, which are essential in industrial medicine, can be addressed using DT technology. The author's approach and idea have a wide range of applications.da Silva et al. address the performance of a classifier of a mechanical system using transfer learning. The authors used the Gaussian mixture model. The paper is application-oriented but demonstrates the effective application of the domain adaptation model.de-Carvalho Michalski et al. used the DT technology as a possible alternative to detect faults in a physical space. The key advantage of this approach is its capability to defects without prior knowledge of system operating conditions. The approach presented by the authors involves the use of system equations and sensor readings. The work is more on the theoretical side.General research on digital twin technology will only be enormous with time. Researchers must not mix the DT technology with simulation. Also, the graphical model encodes physical and virtual systems that need urgent application. In terms of the latter, if care is not taken, DT technology's decision-making will be flawed and meaningless.

  PublisherOA PDFDOI

• Optimal Transport Theory for Railway Track Safety Applications

  ASCE-ASME Journal of Risk and Uncertainty in Engineering Systems Part B Mechanical Engineering · 2024-02-20

  article1st authorCorresponding

  Abstract Optimal mass transport (OMT) is an optimization problem that determines a method of moving mass or value from one function to another in the most efficiently and optimally. Using the ideas of OMT, this technical note aims to formulate and solve an OMT model for railway track safety. The inputs for the analysis include (a) geometry data, (b) rail defect, (c) subsurface condition, and (d) tonnage. The technical note only presents the steps in achieving a complete OMT analysis. Therefore, the idea of the note is to serve as an illustration of the effectiveness of OMT in railway engineering applications.

  PublisherDOI

Frequent coauthors

• Albert Y. Ayenu-Prah

  University of New Orleans

  16 shared

• Ahmed Lasisi

  University of Manitoba

  14 shared

• Offei Adarkwa

  10 shared

• Stephen Mensah

  9 shared

• Emmanuel Nii Martey

  University of Professional Studies

  9 shared

• Yaw Adu‐Gyamfi

  University of Missouri

  8 shared

• Allan M. Zarembski

  University of Delaware

  8 shared

• Leslie Mills

  University of Delaware

  7 shared

Education

• PhD, Civil Eng

  University of Kansas

  1993

Awards & honors

• Member of the National Academy of Science, Engineering and M…