Simulation-based Method Research Articles

The cosmological analysis of X-ray cluster surveys. VI. Inference based on analytically simulated observable diagrams

The number density of galaxy clusters across mass and redshift has been established as a powerful cosmological probe, yielding important information on the matter components of the Universe. Cosmological analyses with galaxy clusters traditionally employ scaling relations, which are empirical relationships between cluster masses and their observable properties. However, many challenges arise from this approach as the scaling relations are highly scattered, maybe ill-calibrated, depend on the cosmology, and contain many nuisance parameters with low physical significance. For this paper, we used a simulation-based inference method utilizing artificial neural networks to optimally extract cosmological information from a shallow X-ray survey, solely using count rates, hardness ratios, and redshifts. This procedure enabled us to conduct likelihood-free inference of cosmological parameters $ m $ and $ To achieve this, we analytically generated several datasets of 70\,000 cluster samples with totally random combinations of cosmological and scaling relation parameters. Each sample in our simulation is represented by its galaxy cluster distribution in a count rate (CR) and hardness ratio (HR) space in multiple redshift bins. We trained convolutional neural networks (CNNs) to retrieve the cosmological parameters from these distributions. We then used neural density estimation (NDE) neural networks to predict the posterior probability distribution of $ m $ and $ given an input galaxy cluster sample. Using the survey area as a proxy for the number of clusters detected for fixed cosmological and astrophysical parameters, and hence of the Poissonian noise, we analyze various survey sizes. The 1 sigma errors of our density estimator on one of the target testing simulations are 1\,000\,deg$^2$, 15.2<!PCT!> for $ m $ and 10.0<!PCT!> for $ and 10\,000\,deg$^2$, 9.6<!PCT!> for $ m $ and 5.6<!PCT!> for $ We also compare our results with a traditional Fisher analysis and explore the effect of an additional constraint on the redshift distribution of the simulated samples. We demonstrate, as a proof of concept, that it is possible to calculate cosmological predictions of $ m $ and $ from a galaxy cluster population without explicitly computing cluster masses and even the scaling relation coefficients, thus avoiding potential biases resulting from such a procedure.

Optimal Number of Replications for Obtaining Stable Dynamic Fit Index Cutoffs

Factor analysis is commonly used in behavioral sciences to measure latent constructs, and researchers routinely consider approximate fit indices to ensure adequate model fit and to provide important validity evidence. Due to a lack of generalizable fit index cutoffs, methodologists suggest simulation-based methods to create customized cutoffs that allow researchers to assess model fit more accurately. However, simulation-based methods are computationally intensive. An open question is: How many simulation replications are needed for these custom cutoffs to stabilize? This Monte Carlo simulation study focuses on one such simulation-based method—dynamic fit index (DFI) cutoffs—to determine the optimal number of replications for obtaining stable cutoffs. Results indicated that the DFI approach generates stable cutoffs with 500 replications (the currently recommended number), but the process can be more efficient with fewer replications, especially in simulations with categorical data. Using fewer replications significantly reduces the computational time for determining cutoff values with minimal impact on the results. For one-factor or three-factor models, results suggested that in most conditions 200 DFI replications were optimal for balancing fit index cutoff stability and computational efficiency.

Quantifying the Tip Leakage Vortex Wandering in a Low-Speed Axial Flow Fan via URANS Simulation

The tip leakage vortex is a dominant noise and aerodynamic loss source of low-speed axial flow fans. The tip leakage vortex often has an unsteady behavior, like the periodic motion of the vortex core: the so-called vortex wandering. This periodic vortex wandering results in additional noise, aerodynamic loss, and an oscillation in the moment acting on the blading, which causes an unfavorable vibration of the rotor. In the research campaigns focusing on vortex wandering, complicated measurement techniques (PIV) or high-fidelity but computationally costly simulations are commonly used. The present paper aims to introduce a relatively cost-effective unsteady Reynolds-averaged Navier-Stokes simulation-based investigation method of vortex wandering. The capabilities of the proposed technique are presented through a low-speed axial flow fan case study.

Simulation and Optimization of Automated Guided Vehicle Charging Strategy for U-Shaped Automated Container Terminal Based on Improved Proximal Policy Optimization

As the decarbonization strategies of automated container terminals (ACTs) continue to advance, electrically powered Battery-Automated Guided Vehicles (B-AGVs) are being widely adopted in ACTs. The U-shaped ACT, as a novel layout, faces higher AGV energy consumption due to its deep yard characteristics. A key issue is how to adopt charging strategies suited to varying conditions to reduce the operational capacity loss caused by charging. This paper proposes a simulation-based optimization method for AGV charging strategies in U-shaped ACTs based on an improved Proximal Policy Optimization (PPO) algorithm. Firstly, Gated Recurrent Unit (GRU) structures are incorporated into the PPO to capture temporal correlations in state information. To effectively limit policy update magnitudes in the PPO, we improve the clipping function. Secondly, a simulation model is established by mimicking the operational process of the U-shaped ACTs. Lastly, iterative training of the proposed method is conducted based on the simulation model. The experimental results indicate that the proposed method converges faster than standard PPO and Deep Q-network (DQN). When comparing the proposed method-based charging threshold with a fixed charging threshold strategy across six different scenarios with varying charging rates, the proposed charging strategy demonstrates better adaptability to terminal condition variations in two-thirds of the scenarios.

Discrete-Event Simulation Integrates an Improved NEH Algorithm for Practical Flowshop Scheduling Problems in the Satellite Industry

The production of multiple types of satellites based on a common manufacturing platform represents a permutation flowshop scheduling problem (PFSP) with complex constraints. This is a highly complex scheduling problem, yet there is still a gap between theoretical research and practical application, particularly in the satellite industry. Therefore, we propose a more practical method that integrates discrete-event simulation modelling and an improved NEH algorithm to solve a more realistic PFSP. The discrete-event simulation-based method includes the following three main components: a flexible PFSP simulation modelling approach, an improved NEH algorithm, and an interaction mechanism between the simulation model and the optimisation algorithm. The proposed method allows automatic and flexible simulation modelling according to the characteristics of the actual satellite manufacturing workshop, which determines the practical nature of the approach proposed in this paper and then achieves excellent scheduling results based on the special interaction mechanism. The computational results demonstrate that this is a 9.18% improvement over the initial NEH algorithm and a 1.40% improvement over the best current improved NEH algorithm.

Comparison of discussion-based and simulation-based learning methods using the Gas Man® on knowledge of the uptake and distribution of inhalation anesthetics of anesthesia resident: A randomized controlled trial

Abstract Background and Aims: Inhalation anesthetics is basic knowledge that must be mastered by an anesthesiologist. Lack of competencies can result in higher morbidity and mortality in anesthesiology practice. Various learning methods were developed to improve understanding and retention. With the advancement of technology, the utilization of screen-based simulation (SBS) using applications is common, such as the Gas Man® application, to help students understand physiology, pathophysiology, and kinetic inhalation anesthetic agents. The primary objective of this study was to compare the knowledge improvement of anesthesia residents between the simulation-based and discussion-based learning methods on the uptake and distribution of anesthetic inhalation. The secondary objective was to compare the satisfaction levels of participants in both learning methods. Material and Methods: This study is a randomized controlled trial involving anesthesiology and intensive care residents from the Faculty of Medicine, Universitas Indonesia (FKUI) 2022–2023. The participants were randomized into two groups (simulation and discussion groups). A senior attending anesthesiologist led the discussion. The participants answered a pretest, followed by a 1-hour lecture about inhalation gas. Lastly, participants answered a posttest and filled in the satisfaction survey. Statistical analyses used the Mann–Whitney test to compare the two groups. The Wilcoxon test was used to compare the knowledge improvement between the two groups. Results: Thirty-eight residents underwent simulation-based (n = 19) and discussion-based (n = 19) learning methods. The simulation group demonstrated a median (interquartile range) posttest score of 80 (76.67–83.33) compared with the discussion group with a score of 50 (40–66.67) (P value <0.05). More than half of the participants in both groups answered “satisfied.” Conclusion: The simulation-based learning method using the Gas Man® application performed better than the discussion-based learning method on increasing knowledge of the uptake and distribution of inhalation anesthetics in anesthesia residents. Key Message: Simulation-based learning method may improve understanding and retention. This study supported the conclusion that the simulation-based learning method with Gas Man is better than the discussion-based learning method for improving knowledge of the uptake and distribution of inhalation anesthetics.

Optimal Market Completion through Financial Derivatives with Applications to Volatility Risk

This paper investigates the optimal choices of financial derivatives to complete a financial market in the framework of stochastic volatility (SV) models. We first introduce an efficient and accurate simulation-based method applicable to generalized diffusion models to approximate the optimal derivatives-based portfolio strategy. We build upon a double optimization approach, i.e., expected utility maximization and risk exposure minimization, already proposed in the literature, demonstrating that strangle options are the best choices for market completion within equity options. They lead to lower investors’ risk exposure for a wide range of strikes compared to the lesser flexibility of calls, puts, and strangles. Furthermore, we explore the benefit of using volatility index derivatives and conclude that they could be more convenient substitutes when short-term maturity equity options are not available.

A Survey of Obstetric Anaesthesia Services and Practices in the United Kingdom.

Background Variability in obstetric anaesthetia practice and care delivered within the UK is under-explored. The ObsQoR study explored structures, processes, and outcomes of obstetric anaesthesia in 107 hospitals within the UK's National Health Service, and the results of the hospital-level survey are reported here. Methods Hospitals were surveyed to assess obstetric anaesthesia provision, practice, and care delivery. Questions explored staffing, service provision and training, facilities present, clinical practices, outcome measurement, and key indicators of quality in obstetric anaesthesia. Results We received responses from 106 participating hospitals, representing 69% of all UK obstetric units. One hundred (94%) hospitals had a dedicated consultant obstetric anaesthetist within working hours, with 27 (25%) of hospitals' duty anaesthetists having additional clinical responsibilities outside the care of obstetric patients outside of working hours. Around 102 hospitals (98%) offer multidisciplinary team training, of which 95 (93%) use a simulation-based method. Dedicated high-risk antenatal clinics were present in 50 (47%) hospitals. The majority of hospitals provide written patient information in multiple languages for discussing obstetric anaesthesia options (77, 82%). Seventy-three hospitals (69%) use point-of-care testing to estimate haemoglobin concentration. Labour epidural analgesia is most commonly delivered via patient-controlled epidural analgesia in 80 (76%) hospitals, and the incidence of post-dural puncture headaches was recorded by 80 (76%) hospitals. Conclusions These results demonstrate variation in the provision of staffing, facilities, clinical practices, and outcome measurements across the UK. To deliver safe and equitable care across the UK, there needs to be standardisation of anaesthetic peripartum care based on national recommendations and the benchmarking and measurement of appropriate markers of quality.

Model-implied simulation-based power estimation for correctly specified and distributionally misspecified models: Applications to nonlinear and linear structural equation models.

Closed-form (asymptotic) analytical power estimation is only available for limited classes of models, requiring correct model specification for most applications. Simulation-based power estimation can be applied in almost all scenarios where data following the model can be estimated. However, a general framework for calculating the required sample sizes for given power rates is still lacking. We propose a new model-implied simulation-based power estimation (MSPE) method for the z-test that makes use of the asymptotic normality property of estimates of a wide class of estimators, the M-estimators, and give theoretical justification for the approach. M-estimators include maximum-likelihood, least squares estimates and limited information estimators, but also estimators used for misspecified models, hence, the new simulation-based power modeling method is widely applicable. The MSPE employs a parametric model to describe the relationship between power and sample size, which can then be used to determine the required sample size for a specified power rate. We highlight its performance in linear and nonlinear structural equation models (SEM) for correctly specified models and models under distributional misspecification. Simulation results suggest that the new power modeling method is unbiased and shows good performance with regard to root mean squared error and type I error rates for the predicted required sample sizes and predicted power rates, outperforming alternative approaches, such as the naïve approach of selecting a discrete selection of sample sizes with linear interpolation of power or simple logistic regression approaches. The MSPE appears to be a valuable tool to estimate power for models without an (asymptotic) analytical power estimation.

Cyanobacterial blooms management: A simulation-based optimization method

Controlling cyanobacterial blooms is not only an engineering and technical issue but also an optimization problem in environment management. Under budget constraints, a novel simulation-based optimization model for cyanobacterial control is constructed in this study. The simulation model is used for simulating cyanobacteria growth and diffusion processes. The optimization model is utilized to determine the optimal search and treatment path. Through the interactive coupling of simulation modeling and resource allocation optimization, this research provides decision-makers with new operational guidelines for cyanobacterial control. Our test results demonstrate that the initial invasion frequency has a greater economic impact than invasion abundance. The nearby cells to the initial invasion are affected first, and then the influence radiates outward in a diffusion pattern. Using a slow search speed and a treatment frequency of every 10 days can achieve the lowest possible economic losses in most test scenarios. Moreover, we also find that the optimal search and treatment paths revolve around the initial invasion location. This study is a typical interdisciplinary research, which can assist water resource managers in making more accurate decisions regarding cyanobacteria removal paths, removal frequencies, and treatment speeds.

Experimental Design of Steel Surface Defect Detection Based on MSFE-YOLO—An Improved YOLOV5 Algorithm with Multi-Scale Feature Extraction

Integrating artificial intelligence (AI) technology into student training programs is strategically crucial for developing future professionals with both forward-thinking capabilities and practical skills. This paper uses steel surface defect detection as a case study to propose a simulation-based teaching method grounded in deep learning. The method encompasses the entire process from data preprocessing and model training to validation analysis and innovation optimization with the goal of deepening students’ understanding of AI technology and enhancing their ability to apply it to real-world scenarios. We have designed an experimental framework that incorporates the Efficient Multi-Scale Attention (EMA) mechanism into the Backbone network. This approach helps students understand the principles of feature extraction and the core functions of attention mechanisms. Additionally, we introduced a novel architecture—Convolution 3 Dilated Convolution X (C3DX)—into the Neck network. This architecture effectively expands the network’s receptive field, improves its ability to capture multi-scale information, and thus enhances defect detection accuracy. Furthermore, the implementation of the Efficient Intersection over Union (EIoU) loss function optimizes the bounding box predictions, further increasing the model’s accuracy and robustness. Overall, the teaching design not only ensures that the content remains at the cutting edge of technology but also emphasizes its practicality and operability. This approach enables students to effectively apply theoretical knowledge to real-world engineering projects.

Drift Control of High-Dimensional Reflected Brownian Motion: A Computational Method Based on Neural Networks

Motivated by applications in queueing theory, we consider a stochastic control problem whose state space is the d-dimensional positive orthant. The controlled process Z evolves as a reflected Brownian motion whose covariance matrix is exogenously specified, as are its directions of reflection from the orthant’s boundary surfaces. A system manager chooses a drift vector [Formula: see text] at each time t based on the history of Z, and the cost rate at time t depends on both [Formula: see text] and [Formula: see text]. In our initial problem formulation, the objective is to minimize expected discounted cost over an infinite planning horizon, after which we treat the corresponding ergodic control problem. Extending the earlier work by Han et al. [Han J, Jentzen A, Weinan E (2018) Solving high-dimensional partial differential equations using deep learning. Proc. Natl. Acad. Sci. USA 115(34):8505–8510], we develop and illustrate a simulation-based computational method that relies heavily on deep neural network technology. For the test problems studied thus far, our method is accurate to within a fraction of 1% and is computationally feasible in dimensions up to at least [Formula: see text].

Stress-induced warpage estimation of advanced semiconductor copper interconnect processes

The growth of the semiconductor industry is driven by the demand for electronic products and high transistor density. However, complex manufacturing processes generate residual stress and result in wafer warpage. Therefore, mastering wafer warpage has become a crucial challenge. This study proposes a process-oriented simulation methodology with simulation-based equivalent material method to overcome the difficulty of finite element modeling and the substantial amount of computation time. Three different methodologies, including volume percentage, representative volume element, and Timoshenko bi-material approach, are discussed due to the estimation of residual stress for equivalent material. In addition, each methodology is validated through process-oriented simulations and comparison with measurement data. The Timoshenko bi-material approach is efficient in predicting warpage in the back end of line (BEOL) interconnects and provides a comprehensive understanding of the warpage variation that occurs during different stages of BEOL.

Catastrophe-related disruptions’ preparedness and emergency management in Morocco: a proactive risks and resilience digital twin-based analysis

Purpose The purpose of this study is to analyze the operations and performance dynamics of a supply chain (SC) subject to disruptions. The preparedness of Moroccan responders in handling emergencies could be enhanced significantly, by devising digital twin-based decision support systems (DSSs). Design/methodology/approach The authors create a discrete-event simulation model to investigate proactively risks and resilience of a Moroccan basic-items SC (BISC). In this study, the authors analyze the effects of catastrophe-related disruptions (CRDs) on the Moroccan BISC, by the use of a simulation-based decision-supporting quantitative method. Findings In the disruption-free simulation experiment, the outcome was a satisfactory 100% coverage. By implementing CRDs, inventory levels have dropped, service levels decreased, lead time raised and there was an increase in backlogged products and late orders numbers. The highest impact was observed for the shutdown of paths linking suppliers to warehouses, whereas the increase in demand had a comparatively minor effect. The risk analysis approach helps to identify critical products for which the time-to-recover is longer and requires more commitment to enhance their resilience. Practical implications The model serves to deduce quantitative resilience assessment from simulation, streamline the selection of recovery strategies and enable the best-informed reactive decision-making to minimize the impact. Originality/value The research brings organizing solutions to catastrophe-related emergencies in Morocco. It would contribute significantly by visualizing, examining and unveiling the effects of disruptions on a BISC and offering actionable recommendations for remedial measures.

Integration of smart charging of large-scale electric vehicles into generation and storage expansion planning: a case study in south china

This paper studies how to integrate the smart charging of large-scale electric vehicles (EVs) into the generation and storage expansion planning (GSEP), while analyzing the impact of smart charging on the GSEP of a real power system in south China. For this purpose, a random simulation-based method is first developed to provide the tractable formulations of the adjustable charging load and reserve provision from EVs. This method avoids the unrealistic assumption that EVs drive and charge every day, which often exists in prior relevant approaches. Based on the random simulation, this paper proposes a novel GSEP optimization model which incorporates the weekly adjustable charging load of EVs. In the proposed model, the total charging load of EVs can be co-optimized with the investment and operational decisions of various generation and storage units. This GSEP model is applied to a provincial power system in south China. The numerical results show that the implementation of smart charging can significantly alter the decisions of GSEP. As the participation rate of smart charging improves from 0% to 90%, there is an additional 1,800 MW installation in wind and solar power, while the need to build new batteries is noticeably reduced; also, depending on the level of EV uptake, the annualized total system cost decreases by 5.11%–7.57%, and the curtailment of wind and solar power is reduced by 10.34%–19.64%. Besides, numerical tests reveal that the traditional assumption that EVs drive and charge every day can mislead the evaluation of adjustable charging load and overestimate the daily charging power peak by averagely 24.72%.

A numerical algorithm for damage progression of reinforced concrete bridge piers during air temperature and solar radiation cycles

Reinforced concrete (RC) solid piers are unavoidably affected by air temperature changes and solar radiation during service life. Nevertheless, the deterioration of pier concrete with increased exposure age remains unclear. A novel thermal-mechanical numerical algorithm (finite element model) for characterizing the deterioration characteristics of RC structures under cyclic environmental loads has been suggested. The deterioration of RC pier concrete after 100 years of air temperature and solar radiation cycles is simulated. The numerical results demonstrate that tensile damage of pier concrete during air temperature and solar radiation cycles increases continuously with exposure age. After one year of exposure, the tensile damage value of pier surface concrete increased to 0.201 and further increased to 0.286 and 0.357 as the exposure age grew to 10 and 100 years. Pier concrete tensile damage can be mitigated by using surface short-wave absorption-reducing methods, controlling the thermal properties of concrete, and raising the concrete peak tensile strain. The proposed finite element framework and numerical solutions enable a viable simulation-based method for the analysis of bridge structural long-term deterioration behavior and design of the bridge engineering.

Capturing impacts of travel preference on connected autonomous vehicle adoption of risk-averse travellers in multi-modal transportation networks

ABSTRACT This paper proposes a reliability-based combined modal split and traffic assignment model for the multi-modal transportation network with uncertainties in which the travellers' travel preferences, vehicle ownership, and the characteristics of connected autonomous vehicles (CAVs) are considered. Travellers are classified into captive travellers and choice travellers according to their preference for travel modes and vehicle ownership. The choice travellers need to make a tradeoff among the out-of-pocket cost, comfort, and travel time reliability and make decisions between continuing to commute by subway or purchasing a human-driven vehicle (HV) or CAV. The dogit-nested-logit model is adopted to characterise the effect of travel preference on the adoption of CAVs of the choice travellers, and the multinomial logit model is adopted to capture the effect of the differences between HV and CAV on information quality and value of travel time reliability on the path choice behaviour of travellers. The problem is formulated as an equivalent variational inequality model based on the properties of the equilibrium state, and its equivalency and solution existence are discussed. A path-based algorithm is employed for solving the problem, which combines the Monte Carlo simulation-based method and the column generation method. Numerical experiments are presented to illustrate the properties of the proposed models and validate the algorithm.

PACS: Projection-driven with Adaptive CADs X-ray Scatter compensation for additive manufacturing inspection

Additive Manufacturing (AM) has revolutionised the production of custom-shaped samples through direct manufacturing from digital design models. As the internal structural integrity of these printed samples is of critical importance in diverse applications, X-ray radiography and X-ray Computer Tomography (X-CT) have emerged as widely used non-destructive imaging methods for quality control of AM samples. Unfortunately, beam hardening and X-ray scatter often degrade the quality of X-CT images, posing a significant challenge for X-ray based inspection. In addressing X-ray scatter, most of the methodologies assume fixed scanning geometries or stationary/known object characteristics, limiting their practicality in dynamic industrial scenarios where these may change over time. Simulation-based methods have been proposed that estimate and suppress scatter by accurately simulating the forward projection process. Yet, these methods assume the availability of X-CT reconstruction for simulation, thereby requiring a large number of projections (and hence scan time) for faithful X-CT reconstruction. In this work, we propose a simulation-based scatter compensation method (PACS) that eliminates the need for a prior X-CT scan. By employing few projections and nominal surface meshes of the scanned objects, the actual pose of the objects and their superficial deviation (e.g., due to printing) are estimated and used during X-ray simulations. Furthermore, as the pipeline is inherently coupled with a mesh projector, analysis of projective residuals facilitates the inspection for deformities or defects within the scanned object. To demonstrate the versatility of PACS for mitigating scatter, experiments across various inspection scenarios are conducted, and the outcomes are compared with those of a well-established scatter compensation technique. The results consistently show a higher Signal-to-Noise Ratio and Contrast-to-Noise Ratio of pore-defects, as well as lower residual errors in all examined cases.

Enhancing the building resilience in a changing climate through a passive cooling roof: A case study in Camas (Seville, Spain)

Renovating buildings is a crucial mission for the coming decades to combat energy inefficiencies and produce a resilient building stock. The inefficient thermal envelope has significant implications for occupants, especially for families lacking the financial resources to maintain a healthy and comfortable indoor environment. In scenarios where households cannot afford or utilize HVAC systems and high temperatures are more frequent, building retrofits should prioritize reducing heat loss in winter and passively lowering indoor temperatures in the cooling season. This work introduces a new solution to enhance the liveability of a social housing building in southern Spain by integrating a double-skin roof with a conventional retrofit strategy, which incorporates passive cooling techniques – night ventilation and an evaporative cooling system – for effective heat dissipation based on available commercial solutions. An extensive monitoring campaign was conducted over three years to evaluate the thermal comfort of occupants and the risk of overheating in naturally conditioned buildings before and after renovation following two impact assessment methods: real-time data-based and simulation-based. The real-time data-based method compared the indoor conditions of a dwelling with a conventionally retrofitted roof to one with a double-skin roof, indicating that the dynamic roof maintains a ceiling temperature lower than the air temperature throughout the summer while the conventional roof acts as a heating surface. On the other side, the simulation-based method compared the observed indoor conditions with a double-skin roof enabled during 2023 to the initial stage and a double-skin roof disabled scenario, using calibrated building energy models. The natural cooling roof solution almost eliminates the overheating events, leading to a reduction of 94.1% and 76.9% in discomfort degree-hours compared to the initial stage according to the ASHRAE adaptative and Fanger comfort models, respectively. Additionally, the results indicate that conventional retrofits can increase the risk of overheating when a cooling system is not considered. Integrating a cool roof solution requires only an extra implementation cost of 31€/m2 compared to a conventional solution, having an operational cost that represents less than 3% of the minimum monthly income in the social housing district.

Real-Time Estimation of Origin–Destination Matrices Using a Deep Neural Network for Digital Twins

The digital twin, a real-time replica of physical systems, has garnered attention as a promising tool to strategize and evaluate solutions for complex real-world issues. However, developing digital twins in the field of transportation faces significant challenges related to the real-time estimation of dynamic origin–destination (OD) matrices constrained by computation time. To bridge this gap, microscopic traffic simulations with real-time synchronization are being researched. Nonetheless, the computational issue persists, emphasizing the need for more efficient OD estimation methods. In this regard, our objective is to reduce computation time in simulation-based methods by developing a data-driven metamodel using a deep neural network. The proposed model serves to map the correlation between the OD matrix and detector data. This model simplifies the computational process using hidden layers, rather than calculating complex interactions between vehicles in the traffic simulation. Compared to conventional methods, we evaluate the capability to estimate a reasonable OD matrix within a restricted time and validate our approach using detector data from Daejeon City, South Korea. The findings indicate that by combining our data-driven metamodel with the simultaneous perturbation stochastic approximation, it becomes feasible to estimate a reasonable OD matrix within a stipulated time frame, compared to the conventional method. Given a 1-min time frame, the proposed method outperforms the conventional simulation-based method by improving the calibration performance of traffic flow by 44.5 percentage points. This paper proposes a practical and versatile approach for real-time OD estimation, laying the foundation for extending microscopic traffic simulation into the digital twin.