Data-driven Construction Research Articles

Data-assisted non-intrusive model reduction for forced nonlinear finite elements models

AbstractSpectral submanifolds (SSMs) have emerged as accurate and predictive model reduction tools for dynamical systems defined either by equations or data sets. While finite-elements (FE) models belong to the equation-based class of problems, their implementations in commercial solvers do not generally provide information on the nonlinearities required for the analytical construction of SSMs. Here, we overcome this limitation by developing a data-driven construction of SSM-reduced models from a small number of unforced FE simulations. We then use these models to predict the forced response of the FE model without performing any costly forced simulation. This approach yields accurate forced response predictions even in the presence of internal resonances or quasi-periodic forcing, as we illustrate on several FE models. Our examples range from simple structures, such as beams and shells, to more complex geometries, such as a micro-resonator model containing more than a million degrees of freedom. In the latter case, our algorithm predicts accurate forced response curves in a small fraction of the time it takes to verify just a few points on those curves by simulating the full forced-response.

Data-driven construction of maritime traffic networks for AI-based route prediction

Abstract Predicting the routes of maritime traffic can improve economic efficiency, decrease ecological impact, and improve safety at sea. Over scales that are small (few hundred meters) and large (dozens to hundreds of kilometers), vessel trajectories have successfully been predicted by deep learning and (static) network-based approaches, respectively. We present an approach for medium to large scales (few kilometers) where (a) a maritime traffic network is automatically constructed from AIS messages, and (b) vessel trajectories are predicted as most likely paths through the network. Using three regions (Stavanger, Tromsø, and Oslo), we show that the network can capture up to ∼ 90 per cent of all maritime traffic (excluding pleasure craft) with a median absolute error of ∼ 80 meters. Vessel paths are sequences of waypoints and legs (nodes and edges) and are map-matched onto the network from vessel trajectories. Once mapped, we predict future paths for two subproblems – (i) known destination, and (ii) unknown destination. We use four algorithms (Dijkstra, Markov, MOGen, GRETEL). For known destinations, we find that Dijkstra performs best. In Stavanger (Tromsø, Oslo), Dijkstra predicts 64 (42, 68) per cent of path segments correctly and keeps the median path error below 15 (33 and 55) meters. For unknown destinations, performance depends on the forecast horizon (the number of legs k to predict). For k ≤ 5, Markov is best and predicts 62 (48, 72) per cent of legs correctly. For k > 5, GRETEL performs best and predicts 54 (47, 63) per cent of legs correctly. For some types of vessels, models improve by considering vessel type. For passenger vessels, models specific to them predict ∼ 10 per cent better paths with half the distance error. For tankers, paths (and distance errors) are 6 (20) per cent worse. For auxiliary vessels, path quality is unchanged, but distance error improves ∼ 36 per cent.

Model reduction on manifolds: A differential geometric framework

Using nonlinear projections and preserving structure in model order reduction (MOR) are currently active research fields. In this paper, we provide a novel differential geometric framework for model reduction on smooth manifolds, which emphasizes the geometric nature of the objects involved. The crucial ingredient is the construction of an embedding for the low-dimensional submanifold and a compatible reduction map, for which we discuss several options. Our general framework allows capturing and generalizing several existing MOR techniques, such as structure preservation for Lagrangian- or Hamiltonian dynamics, and using nonlinear projections that are, for instance, relevant in transport-dominated problems. The joint abstraction can be used to derive shared theoretical properties for different methods, such as an exact reproduction result. To connect our framework to existing work in the field, we demonstrate that various techniques for data-driven construction of nonlinear projections can be included in our framework.

Multi-cell-line learning for the data-driven construction of mechanistic metabolic models.

Mammalian cells are commonly used as hosts in cell culture for biologics production in the pharmaceutical industry. Structured mechanistic models of metabolism have been used to capture complex cellular mechanisms that contribute to varying metabolic shifts in different cell lines. However, little research has focused on the impact of temporal changes in enzyme abundance and activity on the modeling of cell metabolism. In this work, we present a framework for constructing mechanistic models of metabolism that integrate growth-signaling control of enzyme activity and transcript dynamics. The proposed approach is applied to build models for three Chinese hamster ovary (CHO) cell lines using fed-batch culture data and time-series transcript profiles. Leveraging information from the transcriptome data, we develop a parameter estimation approach based on multi-cell-line (MCL) learning, which combines data sets from different cell lines and trains the individual cell-line models jointly to improve model accuracy. The computational results demonstrate the important role of growth signaling and transcript variability in metabolic models as well as the virtue of the MCL approach for constructing cell-line models with a limited amount of data. The resulting models exhibit a high level of accuracy in predicting distinct metabolic behaviors in the different cell lines; these models can potentially be used to accelerate the process and cell-line development for the biomanufacturing of new protein therapeutics.

Iterative data-driven construction of surrogates for an efficient Bayesian identification of oil spill source parameters from image contours

Iterative data-driven construction of surrogates for an efficient Bayesian identification of oil spill source parameters from image contours

Invited commentary: where do the causal DAGS come from?

How do we construct our causal directed acyclic graphs (DAGs)-for example, for life-course modeling and analysis? In this commentary, I review how the data-driven construction of causal DAGs (causal discovery) has evolved, what promises it holds, and what limitations or caveats must be considered. I find that expert- or theory-driven model-building might benefit from some more checking against the data and that causal discovery could bring new ideas to old theories.

A data-driven construction method of aggregated value chain in three phases for manufacturing enterprises

Under the dual influence of dual-carbon policies and the complex international situation, manufacturing enterprises urgently need to build digital value chain ecosystems to adapt to changes in market demand. However, in this process, manufacturing enterprises face many challenges such as multi-stage test data and heterogeneous information sources, large errors in product demand prediction, and uncertainty in maintenance thresholds and maintenance sequences. To address these challenges, a data-driven value chain integration approach is proposed in this paper to build a digital value chain ecosystem for enterprises. In the product design phase, a Bayesian assessment technique is introduced to integrate information from multiple sources. In the product sales phase, a two-stage sales prediction method is proposed. A planning model with variance maximization, correlation minimization, and estimation error minimization as decision objectives is introduced for weight search with ensemble learning. In the after-sales stage of the product, a cost-optimized dynamic maintenance strategy with a maintenance threshold as a constraint is constructed using the number of maintenance and the maintenance interval as decision variables. In addition, cost-based maintenance prioritization is included. Finally, the applicability of the proposed method is verified by the production data of a real manufacturing enterprise. Our findings suggest that the proposed approach helps manufacturing enterprises to aggregate, co-create, and upgrade their value chains. At the same time, manufacturing enterprises should actively build digital value chain ecosystems and dynamically adjust their maintenance strategies according to the production process and life cycle of their products to obtain the best economic benefits.

Data-Driven Construction of Operation Evaluation System for Large Medical Equipment in Children's Hospital

Through data collection and analysis, the method of evaluating the operation quality of large medical equipment in children's hospital is explored and suggestions on the use and configuration of large medical equipment is put forward. Collect the equipment operation data through the Internet of Things, and combine the hospital HIS, RIS, HRP and other information system data to establish the operation evaluation system of large medical equipment of children's hospital. CRITIC method is used to quantitatively evaluate single type of equipment and single equipment. Hospital big data platform realizes the longitudinal analysis of the operation data of a single large equipment and forms a visual chart, which is displayed on the PC and mobile terminals. Municipal platform can conduct horizontal analysis on the equipment operation data to realize the comprehensive quantitative evaluation of the operation level of large equipment of children's hospital and put forward suggestions for use and configuration. A large equipment operation evaluation system for children's hospital is established through data collection and analysis, and the fine management level of large medical equipment is improved.

Learning and optimization under epistemic uncertainty with Bayesian hybrid models

Hybrid (i.e., grey-box) models are a powerful and flexible paradigm for predictive science and engineering. Grey-box models use data-driven constructs to incorporate unknown or computationally intractable phenomena into glass-box mechanistic models. The pioneering work of statisticians Kennedy and O’Hagan introduced a new paradigm to quantify epistemic (i.e., model-form) uncertainty. While popular in several engineering disciplines, prior work using Kennedy–O’Hagan hybrid models focuses on prediction with accurate uncertainty estimates. This work demonstrates computational strategies to deploy Bayesian hybrid models for optimization under uncertainty. Specifically, the posterior distributions of Bayesian hybrid models provide a principled uncertainty set for stochastic programming, chance-constrained optimization, or robust optimization. Through two illustrative case studies, we demonstrate the efficacy of hybrid models, composed of a structurally inadequate glass-box model and Gaussian process bias correction term, for decision-making using limited training data. From these case studies, we develop recommended best practices and explore the trade-offs between different hybrid model architectures.

MF^2: Model-free reinforcement learning for modeling-free building HVAC control with data-driven environment construction in a residential building

Reinforcement Learning (RL) has advanced energy-efficient control of building Heating, Ventilation and Air Conditioning (HVAC) systems. Constructing a suitable RL environment for buildings is a crucial challenge. Compared to widely-used simulation-based environments, data-driven approaches offer higher training efficiency but face convergence difficulties due to influential factors, limiting their current application.To explore data-driven construction of RL environments for building HVAC systems, this study proposes two strategies for controlling room temperature setpoints in a residential building. XGBoost and Long Short-Term Memory Network (LSTM) are trained for energy consumption and room temperature prediction. One strategy predicts parameters for on-off states, while the other for power-on states. The XGBoost models are integrated into an OpenAI Gym environment. The first strategy achieves 0.8634 R2 and 0.2423 Root Mean Squared Error (RMSE) for energy consumption prediction. The R2 of room air temperature models are approximately 0.99 and the RMSE are lower than 0.31. The second strategy achieves 0.9181 R2 and 0.1042 RMSE for energy consumption prediction and similar performance for room temperature prediction. Deep Q-learning (DQN) and Deep Deterministic Policy Gradient (DDPG) algorithms are separately trained using these environments. Results show that the first strategy fails to induce the correct training of RL models, while the second strategy successfully induces a useable DDPG model for controlling building HVAC systems but fails to induce a useable DQN model. We analyze the reasons behind these observations. Compared to the original room temperature setpoint method, the DDPG-based HVAC control logic achieves a 10.06% energy-saving effect while ensuring comfort.

Single-cell reference mapping to construct and extend cell-type hierarchies.

Single-cell genomics is now producing an ever-increasing amount of datasets that, when integrated, could provide large-scale reference atlases of tissue in health and disease. Such large-scale atlases increase the scale and generalizability of analyses and enable combining knowledge generated by individual studies. Specifically, individual studies often differ regarding cell annotation terminology and depth, with different groups specializing in different cell type compartments, often using distinct terminology. Understanding how these distinct sets of annotations are related and complement each other would mark a major step towards a consensus-based cell-type annotation reflecting the latest knowledge in the field. Whereas recent computational techniques, referred to as 'reference mapping' methods, facilitate the usage and expansion of existing reference atlases by mapping new datasets (i.e. queries) onto an atlas; a systematic approach towards harmonizing dataset-specific cell-type terminology and annotation depth is still lacking. Here, we present 'treeArches', a framework to automatically build and extend reference atlases while enriching them with an updatable hierarchy of cell-type annotations across different datasets. We demonstrate various use cases for treeArches, from automatically resolving relations between reference and query cell types to identifying unseen cell types absent in the reference, such as disease-associated cell states. We envision treeArches enabling data-driven construction of consensus atlas-level cell-type hierarchies and facilitating efficient usage of reference atlases.

Quantifying Schedule Delay Risk in Construction Projects: A Data-Driven Approach with BIM and Probabilistic Reliability Analysis

Progress management is an important scheme of construction management. In practical affairs, many unknown factors may lead to potential project delays and make the schedule risky. To address this issue, this paper proposes a data-driven construction progress evaluation method that employs a probabilistic reliability analysis with BIM to effectively quantify the risk of schedule delay. By identifying critical tasks and probable failure conditions in uncertain environments, this method enables project managers to take proactive measures to keep their projects on schedule. Through our case study, we have demonstrated that the proposed method is highly effective in dynamically evaluating the project progress with both accuracy and efficiency. Besides, this system offers a more efficient way for managers to gather information, evaluate progress, and identify critical tasks, and provides project managers with a data-driven understanding of the risks and uncertainties related to project schedules, enabling them to take proactive measures to minimize the potential delays and ensure successful project delivery.

A General Contractor’s Perspective on Construction Digital Twin: Implementation, Impacts and Challenges

The Digital Twin (DT), as a real-time and data-connected virtual replica of a physical asset, introduces a new paradigm in the construction industry. To date, the use of DT in the construction phase has not been addressed sufficiently. Hence, this research studies the implementation of DT during the construction phase of projects to support general contractors’ decisions and operations. Starting from existing literature, a Construction Digital Twin (CDT) framework has been developed from a general contractor’s perspective and a case study was implemented as an application of the proposed CDT to validate the framework and demonstrate its benefits. In the selected project, the simulation of the construction operations in evaluating various “what-if” scenarios for optimum resource allocation and operation management proved the benefits of using a CDT in the construction phase of projects for general contractors. By implementing the proposed CDT framework, several impacts such as reduced costs, improved collaboration and information exchange and data-driven construction management can be anticipated.

Developing a National Data-Driven Construction Safety Management Framework with Interpretable Fatal Accident Prediction

Occupational accidents are frequent in the construction industry, containing significant risks in the working environment. Therefore, early designation, taking preventive actions, and developing a proactive safety risk management plan are of paramount significance in managing safety issues in the construction industry. This study aims to develop a national data-driven safety management framework based on accident outcome prediction, which helps anatomize precursors of fatalities and thereby minimizing fatal accidents on construction sites. A national data set comprising 338,173 occupational accidents recorded in the construction industry across Turkey was used to develop a data-driven model. The random forest algorithm coupled with particle swarm optimization was used for the prediction and the interpretability of the proposed model was augmented through the game theory–based Shapley additive explanations (SHAP) approach. The findings showed that the proposed algorithm achieved satisfactory model performances for detecting construction workers who might face a fatality risk. The SHAP analysis results indicated that both company (such as number of past accidents and workers in the company) and worker-related (such as age, daily wage, experience, shift, and past accident of the workers) attributes were influential in identifying fatalities by detecting which workers might face fatal accidents under which conditions. A construction safety management plan was developed based on the analysis results, which can be used on construction sites to detect workers/conditions that are most susceptible to fatalities. The findings of the present research are expected to contribute to orchestrating effective safety management practices in construction sites by characterizing the root causes of severe accidents.

Deep learning-accelerated computational framework based on Physics Informed Neural Network for the solution of linear elasticity

The paper presents an efficient and robust data-driven deep learning (DL) computational framework developed for linear continuum elasticity problems. The methodology is based on the fundamentals of the Physics Informed Neural Networks (PINNs). For an accurate representation of the field variables, a multi-objective loss function is proposed. It consists of terms corresponding to the residual of the governing partial differential equations (PDE), constitutive relations derived from the governing physics, various boundary conditions, and data-driven physical knowledge fitting terms across randomly selected collocation points in the problem domain. To this end, multiple densely connected independent artificial neural networks (ANNs), each approximating a field variable, are trained to obtain accurate solutions. Several benchmark problems including the Airy solution to elasticity and the Kirchhoff–Love plate problem are solved. Performance in terms of accuracy and robustness illustrates the superiority of the current framework showing excellent agreement with analytical solutions. The present work combines the benefits of the classical methods depending on the physical information available in analytical relations with the superior capabilities of the DL techniques in the data-driven construction of lightweight, yet accurate and robust neural networks. The models developed herein can significantly boost computational speed using minimal network parameters with easy adaptability in different computational platforms.

Unsupervised Data-Driven Classification of Topological Gapped Systems with Symmetries.

A remarkable breakthrough in topological phase classification is the establishment of the topological periodic table, which is mainly based on the classifying space analysis or K theory, but not based on concrete Hamiltonians that possess finite bands or arise in a lattice. As a result, it is still difficult to identify the topological phase of an arbitrary Hamiltonian; the common practice is, instead, to check the incomplete and still growing list of topological invariants one by one, very often by trial and error. Here, we develop unsupervised classifications of topological gapped systems with symmetries, and demonstrate the data-driven construction of the topological periodic table without apriori knowledge of topological invariants. This unsupervised data-driven strategy can take into account spatial symmetries, and further classify phases that were previously classified as trivial in the past. Our Letter introduces machine learning into topological phase classification and paves the way for intelligent explorations of new phases of topological matter.

Data-driven construction of stochastic reduced dynamics encoded with non-Markovian features.

One important problem in constructing the reduced dynamics of molecular systems is the accurate modeling of the non-Markovian behavior arising from the dynamics of unresolved variables. The main complication emerges from the lack of scale separations, where the reduced dynamics generally exhibits pronounced memory and non-white noise terms. We propose a data-driven approach to learn the reduced model of multi-dimensional resolved variables that faithfully retains the non-Markovian dynamics. Different from the common approaches based on the direct construction of the memory function, the present approach seeks a set of non-Markovian features that encode the history of the resolved variables and establishes a joint learning of the extended Markovian dynamics in terms of both the resolved variables and these features. The training is based on matching the evolution of the correlation functions of the extended variables that can be directly obtained from the ones of the resolved variables. The constructed model essentially approximates the multi-dimensional generalized Langevin equation and ensures numerical stability without empirical treatment. We demonstrate the effectiveness of the method by constructing the reduced models of molecular systems in terms of both one-dimensional and four-dimensional resolved variables.

Insurance pricing with hierarchically structured data an illustration with a workers' compensation insurance portfolio

Actuaries use predictive modeling techniques to assess the loss cost on a contract as a function of observable risk characteristics. State-of-the-art statistical and machine learning methods are not well equipped to handle hierarchically structured risk factors with a large number of levels. In this paper, we demonstrate the data-driven construction of an insurance pricing model when hierarchically structured risk factors, contract-specific as well as externally collected risk factors are available. We examine the pricing of a workers' compensation insurance product with a hierarchical credibility model [Jewell, W. S. (1975). The use of collateral data in credibility theory: A hierarchical model. Laxenburg: IIASA], Ohlsson's combination of a generalized linear and a hierarchical credibility model [Ohlsson, E. (2008). Combining generalized linear models and credibility models in practice. Scandinavian Actuarial Journal 2008(4), 301–314] and mixed models. We compare the predictive performance of these models and evaluate the effect of the distributional assumption on the target variable by comparing linear mixed models with Tweedie generalized linear mixed models. For our case-study the Tweedie distribution is well suited to model and predict the loss cost on a contract. Moreover, incorporating contract-specific risk factors in the model improves the predictive performance and the risk differentiation in our workers' compensation insurance portfolio.

Developing a regional environmental corrosion model for Q235 carbon steel using a data-driven construction method

Studying the impact of the environment on metal corrosion is of considerable significance for the safety assessment of buildings and the life prediction of equipment. We developed a new regional environmental corrosion model (RECM) to predict the atmospheric corrosion of Q235 carbon steel based on measured environmental data and corrosion rates obtained from one-year-long static coupon tests. The corrosion of metals varies depending on the environment; therefore, the ability of the model to distinguish such differences is crucial for accurately predicting corrosion. Herein, the regions in which the test sites were located were divided based on the basic principles of atmospheric corrosion. Furthermore, random forest was used to assess the importance of various environmental factors in the corrosion process within each region, which established a close relationship between corrosion and environmental conditions. Our results showed that the accuracy of the RECM is higher than that of the dose-response function of the ISO9223-2012 standard. The method of model construction can be realized automatically using a computer.

Synthesizing theories of human language with Bayesian program induction

Automated, data-driven construction and evaluation of scientific models and theories is a long-standing challenge in artificial intelligence. We present a framework for algorithmically synthesizing models of a basic part of human language: morpho-phonology, the system that builds word forms from sounds. We integrate Bayesian inference with program synthesis and representations inspired by linguistic theory and cognitive models of learning and discovery. Across 70 datasets from 58 diverse languages, our system synthesizes human-interpretable models for core aspects of each language’s morpho-phonology, sometimes approaching models posited by human linguists. Joint inference across all 70 data sets automatically synthesizes a meta-model encoding interpretable cross-language typological tendencies. Finally, the same algorithm captures few-shot learning dynamics, acquiring new morphophonological rules from just one or a few examples. These results suggest routes to more powerful machine-enabled discovery of interpretable models in linguistics and other scientific domains.