Rule-based Model Research Articles

Long-Time gap crowd prediction with a Two-Stage optimized spatiotemporal Hybrid-GCGRU

Crowd prediction is a crucial aspect of modern society, facilitating numerous decision-making processes, such as hazard detection and facility maintenance. Conventional crowd prediction relies on rule-based models which involve tedious formation steps and are error-prone. Hence, this study adopts spatiotemporal deep learning models for crowd prediction in order to exploits both temporal and spatial features of the dataset. There are three major limitations in previous studies: firstly, the prediction length of the time series models is relatively short (i.e., limited to single-digit time steps); Secondly, most studies using time series prediction do not consider the features in the neighboring locations other than the target location; Finally, only the temporal pattern on regular days is exploited, whilst the data variations during holidays can greatly reduce the prediction accuracy. Therefore, a particle swarm optimized (PSO) Hybrid-Graph Convolutional Gated Recurrent Unit (HGCGRU) model, entitled PSO-HGCGRU, is proposed to increase the prediction accuracy in various situations. The PSO-HGCGRU model is composed of two parts, 1) a hybrid model structure, which allows the model adapting to both regular and holiday features, and 2) the GCGRU model, in which the graph convolutional layer to extract the spatial features and the GRU layers to extract the temporal dependencies, to undertake spatiotemporal prediction. The Two-Stage Long-Time Gap Prediction method is applied to the proposed PSO-HGCGRU model with a newly designed two-stage mechanism to optimize the length of the time gap. Results show that the proposed model generally outperforms the baseline models by around 40% in accuracy.

Incorporating Machine Learning in Computer-Aided Molecular Design for Fragrance Molecules

The demand for new novel flavour and fragrance (F&F) molecules has boosted the need for a systematic approach to designing fragrance molecules. However, the F&F-related industry still relies heavily on experimental approaches or on existing databases without considering the consequences resulting from changes in concentration, which could omit potential fragrances. Computer-aided molecular design (CAMD) has great potential to identify novel molecular structures to be used as fragrances. Using CAMD for this purpose requires models to predict the olfaction properties of molecules. A rough set-based machine learning (RSML) approach is used to develop an interpretable predictive model for odour characteristics in this work. New rule-based models are generated from RSML based on the dilution and a number of different topological indices which identify the structure-odour relationship of fragrance molecules. The most prominent rules are selected and formulated as constraints in a CAMD optimisation model. The combination of several rules was able to increase the coverage of different classes of molecules. To model the performance indicators that vary over a range of properties, a disjunctive programming model is also incorporated into the CAMD framework. A case study demonstrates the utilisation of this methodology to design fragrance additives in dishwashing liquid. The results illustrate the capability of the novel RSML and CAMD framework to identify potential fragrance molecules that can be used in consumer products.

Quantitative risk assessment of railway intrusions with text mining and fuzzy Rule-Based Bow-Tie model

With the increasing traveling speed of railway transportation, rail right-of-way intrusions can cause high-consequence accidents and pose severe challenges to railway safety. Although intrusion detection technologies have been widely studied and applied, they can only support in-event inspection and post-event control. In the current complex environment, there is a critical need to analyze the causal chain of railway intrusions and mitigate safety risks before or during the risk evolution process. This paper developed a novel methodological framework on the cause-consequence model based on the text mining techniques and fuzzy bow-tie modeling to systematically investigate the railway intrusion risks. In order to mine both critical factors and their interrelationships, a lexical co-occurrence analysis was carried out on a customized corpus of intrusion accident recordings. Then structured bow-tie diagrams were developed based on the networks generated by unstructured data. To overcome the data uncertainty issue, this paper utilized cause-consequence-based probabilistic analysis and fuzzy theory to quantify the risks involving the occurrence probability of top events and outcomes in terms of expert judgements. The application of the proposed bow-tie model was demonstrated based on the case of the Hualien Derailment accident. The findings based on the bow-tie model and historical accidents in this research have systematically summarized basic events and causal chains. Ultimately, they can be utilized by researchers and practitioners both to identify the critical risk factors and enhance railway safety via proactive and reactive measures.

A New Multilayer Belief Rule Base Model for Complex System Modeling

Belief rule base (BRB) model suffers from the rule explosion problem when it is applied in modeling complex systems. The rule explosion problem makes it difficult to establish and optimize the BRB model effectively. Aiming at this problem, a new multilayer BRB (MLBRB) model is proposed, which consists of the extracting block and the processing block. In the extracting block, the inputs are first divided into two groups, and then each group is used to construct a hierarchical BRB model. The outputs of extracting block are treated as the inputs of the processing block. To obtain the optimal MLBRB, the layerwise learning strategy and the layer adaptive growth strategy are proposed to optimize these two blocks, respectively. A case study for the safety assessment of the liquefied natural gas storage tank is conducted to verify the effectiveness of the proposed method.

Fuzzy Rule-Based Classification Method for Incremental Rule Learning

Granularrules have been extensively used for classification in fuzzy datasets to promote the advancement of artificial intelligence. However, due to the diversity of data types, how to improve the readability of the extracted granular rules while ensuring efficiency is always a challenge. Since granular reduct in granular computing (GrC) can simplify real complex problem and dataset, this article carries out granular rule learning from the perspective of granular reduct by taking formal concept analysis (FCA)-based GrC method as a framework. Specifically, for achieving classification task, we first propose a method to update the granular reduct, and then explore the updating mechanism of fuzzy granular rule in a reduced dataset. Second, a novel fuzzy rule-based classification model named FRCM is presented for fuzzy granular rule learning. In order to verify the effectiveness of the proposed model, some numerical experiments for incremental learning and fuzzy rule mining are conducted to demonstrate that FRCM can achieve the state-of-the-art classification performance.

Horizontal Federated Learning of Takagi–Sugeno Fuzzy Rule-Based Models

In this article, we elaborate on a design and realization of fuzzy rule-based model in the horizontal federated learning framework. Traditional machine learning in distributed environment often involves sharing sensitive information with other sites or transferring data to a central server on which a global model is trained. These situations increase the communication overhead and pose serious threats to the privacy of sensitive data. Federated learning opens up the possibility for collaboratively training a global model on a basis of distributed on-site data without sacrificing data privacy. While fuzzy rule-based models have been used in system modeling due to their substantial modeling abilities and good interpretability, the implementation of fuzzy rule-based models in a distributed environment without compromising data privacy still requires careful consideration. This article proposes a two-step federated learning approach to train a global model on a basis of private data located across different sites without their centralization. The first step concerns the determination of the structure of the data through federated collaborative clustering. Subsequently, a shared global model is trained jointly by all the participating clients. An advantage of the proposed method is that it achieves high accuracy without violating data privacy. A series of experimental studies are conducted to gain a detailed insight into the realization steps and demonstrate the effectiveness of the proposed method.

Coordinated Volt/Var Control of PV and EV Interfaced Active Distribution Networks Based on Dual-Stage Model Predictive Control

This article presents a dual-stage coordinated control approach for voltage regulation and congestion management of active distribution networks (ADN) in the presence of photovoltaic (PV) generators and electric vehicle stations (EVS). The proposed scheme operates on rule-based model predictive control (RBMPC) to optimally manage the settings of the regulating devices, i.e., on-load tap changer (OLTC), distribution static synchronous compensator (DSTATCOM), PV generators, and EV inverters that possess different temporal characteristics. The first (hourly time-scale) and the second (one-minute time-scale) stages of the dual-stage coordinated control mechanism are designed to correct the long-term and short-term voltage fluctuations, respectively. The objectives of the proposed approach are to minimize the number of OLTC operations (first stage) and changes in set-points of PV and EV inverters, and DSTATCOM (second stage) in addition to minimization of slack variables, energy loss, and voltage error at EV stations. In both the stages, predefined rules are set for MPC to optimize different objectives on the basis of the magnitudes of bus voltages. Simulations are performed on 33-bus and 38-bus distribution networks to test the efficacy of the control approach. It is observed that the proposed approach could mitigate the voltage variations as well as line congestion for different scenarios.

Automated Assessment of Balance Rehabilitation Exercises With a Data-Driven Scoring Model: Algorithm Development and Validation Study.

BackgroundBalance rehabilitation programs represent the most common treatments for balance disorders. Nonetheless, lack of resources and lack of highly expert physiotherapists are barriers for patients to undergo individualized rehabilitation sessions. Therefore, balance rehabilitation programs are often transferred to the home environment, with a considerable risk of the patient misperforming the exercises or failing to follow the program at all. Holobalance is a persuasive coaching system with the capacity to offer full-scale rehabilitation services at home. Holobalance involves several modules, from rehabilitation program management to augmented reality coach presentation.ObjectiveThe aim of this study was to design, implement, test, and evaluate a scoring model for the accurate assessment of balance rehabilitation exercises, based on data-driven techniques.MethodsThe data-driven scoring module is based on an extensive data set (approximately 1300 rehabilitation exercise sessions) collected during the Holobalance pilot study. It can be used as a training and testing data set for training machine learning (ML) models, which can infer the scoring components of all physical rehabilitation exercises. In that direction, for creating the data set, 2 independent experts monitored (in the clinic) 19 patients performing 1313 balance rehabilitation exercises and scored their performance based on a predefined scoring rubric. On the collected data, preprocessing, data cleansing, and normalization techniques were applied before deploying feature selection techniques. Finally, a wide set of ML algorithms, like random forests and neural networks, were used to identify the most suitable model for each scoring component.ResultsThe results of the trained model improved the performance of the scoring module in terms of more accurate assessment of a performed exercise, when compared with a rule-based scoring model deployed at an early phase of the system (k-statistic value of 15.9% for sitting exercises, 20.8% for standing exercises, and 26.8% for walking exercises). Finally, the resulting performance of the model resembled the threshold of the interobserver variability, enabling trustworthy usage of the scoring module in the closed-loop chain of the Holobalance coaching system.ConclusionsThe proposed set of ML models can effectively score the balance rehabilitation exercises of the Holobalance system. The models had similar accuracy in terms of Cohen kappa analysis, with interobserver variability, enabling the scoring module to infer the score of an exercise based on the collected signals from sensing devices. More specifically, for sitting exercises, the scoring model had high classification accuracy, ranging from 0.86 to 0.90. Similarly, for standing exercises, the classification accuracy ranged from 0.85 to 0.92, while for walking exercises, it ranged from 0.81 to 0.90.Trial RegistrationClinicalTrials.gov NCT04053829; https://clinicaltrials.gov/ct2/show/NCT04053829

Design of a Rule-based Decisive Model for Optimizing the Load Balancing in a Smart Grid Environment

Design of a Rule-based Decisive Model for Optimizing the Load Balancing in a Smart Grid Environment

Multicenter Validation of Natural Language Processing Algorithms for the Detection of Common Data Elements in Operative Notes for Total Hip Arthroplasty: Algorithm Development and Validation.

BackgroundNatural language processing (NLP) methods are powerful tools for extracting and analyzing critical information from free-text data. MedTaggerIE, an open-source NLP pipeline for information extraction based on text patterns, has been widely used in the annotation of clinical notes. A rule-based system, MedTagger-total hip arthroplasty (THA), developed based on MedTaggerIE, was previously shown to correctly identify the surgical approach, fixation, and bearing surface from the THA operative notes at Mayo Clinic.ObjectiveThis study aimed to assess the implementability, usability, and portability of MedTagger-THA at two external institutions, Michigan Medicine and the University of Iowa, and provide lessons learned for best practices.MethodsWe conducted iterative test-apply-refinement processes with three involved sites—the development site (Mayo Clinic) and two deployment sites (Michigan Medicine and the University of Iowa). Mayo Clinic was the primary NLP development site, with the THA registry as the gold standard. The activities at the two deployment sites included the extraction of the operative notes, gold standard development (Michigan: registry data; Iowa: manual chart review), the refinement of NLP algorithms on training data, and the evaluation of test data. Error analyses were conducted to understand language variations across sites. To further assess the model specificity for approach and fixation, we applied the refined MedTagger-THA to arthroscopic hip procedures and periacetabular osteotomy cases, as neither of these operative notes should contain any approach or fixation keywords.ResultsMedTagger-THA algorithms were implemented and refined independently for both sites. At Michigan, the study comprised THA-related notes for 2569 patient-date pairs. Before model refinement, MedTagger-THA algorithms demonstrated excellent accuracy for approach (96.6%, 95% CI 94.6%-97.9%) and fixation (95.7%, 95% CI 92.4%-97.6%). These results were comparable with internal accuracy at the development site (99.2% for approach and 90.7% for fixation). Model refinement improved accuracies slightly for both approach (99%, 95% CI 97.6%-99.6%) and fixation (98%, 95% CI 95.3%-99.3%). The specificity of approach identification was 88.9% for arthroscopy cases, and the specificity of fixation identification was 100% for both periacetabular osteotomy and arthroscopy cases. At the Iowa site, the study comprised an overall data set of 100 operative notes (50 training notes and 50 test notes). MedTagger-THA algorithms achieved moderate-high performance on the training data. After model refinement, the model achieved high performance for approach (100%, 95% CI 91.3%-100%), fixation (98%, 95% CI 88.3%-100%), and bearing surface (92%, 95% CI 80.5%-97.3%).ConclusionsHigh performance across centers was achieved for the MedTagger-THA algorithms, demonstrating that they were sufficiently implementable, usable, and portable to different deployment sites. This study provided important lessons learned during the model deployment and validation processes, and it can serve as a reference for transferring rule-based electronic health record models.

Deep belief rule based photovoltaic power forecasting method with interpretability

Accurate prediction of photovoltaic (PV) output power is of great significance for reasonable scheduling and development management of power grids. In PV power generation prediction system, there are two problems: the uncertainty of PV power generation and the inexplicability of the prediction result. The belief rule base (BRB) is a rule-based modeling method and can deal with uncertain information. Moreover, the modeling process of BRB has a certain degree of interpretability. However, rule explosion and the inexplicability of the optimized model limit the modeling ability of BRB in complex systems. Thus, a PV output power prediction model is proposed based on a deep belief rule base with interpretability (DBRB-I). In the DBRB-I model, the deep BRB structure is constructed to solve the rule explosion problem, and inefficient rules are simplified by a sensitivity analysis of the rules, which reduces the complexity of the model. Moreover, to ensure that the interpretability of the model is not destroyed, a new optimization method based on the projection covariance matrix adaptation evolution strategy (P-CMA-ES) algorithm is designed. Finally, a case study of the prediction of PV output power is conducted to illustrate the effectiveness of the proposed method.

Design of metaheuristic rough set-based feature selection and rule-based medical data classification model on MapReduce framework

Abstract Recently, big data analytics have gained significant attention in healthcare industry due to generation of massive quantities of data in various forms such as electronic health records, sensors, medical imaging, and pharmaceutical details. However, the data gathered from various sources are intrinsically uncertain owing to noise, incompleteness, and inconsistency. The analysis of such huge data necessitates advanced analytical techniques using machine learning and computational intelligence for effective decision making. To handle data uncertainty in healthcare sector, this article presents a novel metaheuristic rough set-based feature selection with rule-based medical data classification (MRSFS-RMDC) technique on MapReduce framework. The proposed MRSFS-RMDC technique designs a butterfly optimization algorithm for minimal rough set selection. In addition, Hadoop MapReduce is applied to process massive quantity of data. Moreover, a rule-based classification approach named Repeated Incremental Pruning for Error Reduction (RIPPER) is used with the inclusion of a set of conditional rules. The RIPPER will scale in a linear way with the number of training records utilized and is suitable to build models with data uncertainty. The proposed MRSFS-RMDC technique is validated using benchmark dataset and the results are inspected under varying aspects. The experimental results highlighted the supremacy of the MRSFS-RMDC technique over the recent state of art methods in terms of different performance measures. The proposed methodology has achieved a higher F-score of 96.49%.

A Health Evaluation Method of Complex Systems Based on Evidential Reasoning Rule Considering Perturbations

The complex system has been widely applied in aerospace, chemical industry, manufacturing, military equipment and other fields. To ensure its safe and reliable operation, it is necessary for technicians to grasp the health state of the system effectively. However, if there are perturbations in the observation data, the evaluation result may be inaccurate. Besides, the effectiveness of evaluation result can be affected by perturbation, which means that the threshold of perturbation intensity should be determined. To solve these problems, in this paper, a new health evaluation method of complex systems is proposed based on the evidential reasoning (ER) rule by considering perturbations. First, the structure of the evaluation model is constructed. Second, an ER rule-based health evaluation model is proposed, and the idea of perturbation analysis is introduced for the robustness analysis of the model. Third, a parameter optimization model is established, and the upper bound of perturbation intensity is estimated accurately. The threshold can measure the effectiveness of the evaluation result and the anti-perturbation ability of the system in engineering practice. Finally, a case study of the WD615 model diesel engine is conducted to validate the effectiveness of the proposed method.

Structural rule-based modeling with granular computing

In order to analyze the dynamic behaviors of complex systems in the era of big data, a new rule-based modeling approach is proposed in this paper. This approach considers structural information mining and granular computing (GrC) in rule-based modeling, it also aims at utilizing granular fuzzy intervals to reflect systems’ behaviors on uncertainty. Major contributions can be summarized as three points. First, using DBSCAN’s advantages in clustering arbitrarily-shaped data, DBSCAN is applied to extract structural information as the basis of rules in complex systems. Second, GrC is leveraged for rule formation and effective rule-based modeling, e.g. granules constructed in input space to refine structural DBSCAN clusters, granular intervals constructed in output space with the principle of justifiable granulating to reflect system dynamic behaviors. Third, experimental analysis based on three different design scenarios was studied on both synthetic data and publicly available datasets. Through comparative discussion, the proposed approach can outperform the conventional rule-based models, specifically having an improvement ratio of 0.18% to 1.48% on the proposed comprehensive indicator. Therefore, it can be concluded that the proposed approach has the feasibility and advantages of reflecting the structural and uncertainty characteristics of system dynamic analysis.

Sequential Pattern Mining Approach for Personalized Fraudulent Transaction Detection in Online Banking

Financial institutions face challenges of fraud due to an increased number of online transactions and sophisticated fraud techniques. Although fraud detection systems have been implemented to detect fraudulent transactions in online banking, many systems just use conventional rule-based approaches. Rule-based detection systems have a difficulty in updating and managing their rules and conditions manually. Additionally, generated from the few fraud cases, the rules are general rather than specific to each user. In this paper, we propose a personalized alarm model to detect frauds in online banking transactions using sequence pattern mining on each user’s normal transaction log. We assumed that a personalized fraud detection model is more effective in responding to the rapid increase in online banking users and diversified fraud patterns. Moreover, we focused on the fact that fraudulent transactions are very different from each user’s usual transactions. Our proposed model divides each user’s log into transactions, extracts a set of sequence patterns, and uses it to determine whether a new incoming transaction is fraudulent. The incoming transaction is divided into multiple windows, and if the normal patterns are not found in the consecutive windows, an alarm is sounded. We applied the model to a real-world dataset and showed that our model outperforms the rule-based model and the Markov chain model. Although more experiments on additional datasets are needed, our personalized alarm model can be applied to real-world systems.

Accurately Identifying Cerebroarterial Stenosis from Angiography Reports Using Natural Language Processing Approaches.

Patients with intracranial artery stenosis show high incidence of stroke. Angiography reports contain rich but underutilized information that can enable the detection of cerebrovascular diseases. This study evaluated various natural language processing (NLP) techniques to accurately identify eleven intracranial artery stenosis from angiography reports. Three NLP models, including a rule-based model, a recurrent neural network (RNN), and a contextualized language model, XLNet, were developed and evaluated by internal–external cross-validation. In this study, angiography reports from two independent medical centers (9614 for training and internal validation testing and 315 as external validation) were assessed. The internal testing results showed that XLNet had the best performance, with a receiver operating characteristic curve (AUROC) ranging from 0.97 to 0.99 using eleven targeted arteries. The rule-based model attained an AUROC from 0.92 to 0.96, and the RNN long short-term memory model attained an AUROC from 0.95 to 0.97. The study showed the potential application of NLP techniques such as the XLNet model for the routine and automatic screening of patients with high risk of intracranial artery stenosis using angiography reports. However, the NLP models were investigated based on relatively small sample sizes with very different report writing styles and a prevalence of stenosis case distributions, revealing challenges for model generalization.

A Syntactic Information-Based Classification Model for Medical Literature: Algorithm Development and Validation Study.

BackgroundThe ever-increasing volume of medical literature necessitates the classification of medical literature. Medical relation extraction is a typical method of classifying a large volume of medical literature. With the development of arithmetic power, medical relation extraction models have evolved from rule-based models to neural network models. The single neural network model discards the shallow syntactic information while discarding the traditional rules. Therefore, we propose a syntactic information–based classification model that complements and equalizes syntactic information to enhance the model.ObjectiveWe aim to complete a syntactic information–based relation extraction model for more efficient medical literature classification.MethodsWe devised 2 methods for enhancing syntactic information in the model. First, we introduced shallow syntactic information into the convolutional neural network to enhance nonlocal syntactic interactions. Second, we devise a cross-domain pruning method to equalize local and nonlocal syntactic interactions.ResultsWe experimented with 3 data sets related to the classification of medical literature. The F1 values were 65.5% and 91.5% on the BioCreative ViCPR (CPR) and Phenotype-Gene Relationship data sets, respectively, and the accuracy was 88.7% on the PubMed data set. Our model outperforms the current state-of-the-art baseline model in the experiments.ConclusionsOur model based on syntactic information effectively enhances medical relation extraction. Furthermore, the results of the experiments show that shallow syntactic information helps obtain nonlocal interaction in sentences and effectively reinforces syntactic features. It also provides new ideas for future research directions.

A Hierarchical Approach to Interpretability of TS Rule-Based Models

Interpretability of fuzzy rule-based models has always been of significant interest to the research community and the research in this area led to a number of far-reaching results. In this study, we briefly revisit the methodology and concepts of interpretability of Takagi–Sugeno (T-S) rule-based models and develop a conceptual framework involving several levels at which rules are interpreted. The layers at which interpretability is positioned are structured hierarchically by starting with the initial fuzzy set level (originating from the design of the rules), moving to information granules of finite support (where interval calculus is engaged) and finally ending up with symbols built at the higher level. As T-S rule-based models are endowed with local functions forming the conclusion parts of the rules, with the use of the principle of justifiable granularity, we develop a way of forming an interpretable conclusion in the form of information granule. To facilitate interpretability of conditions of the rules, multidimensional fuzzy sets (coming as a result of clustering) are decomposed into a Cartesian product of 1-D fuzzy sets and the quality of the resulting decomposition is evaluated. The quality of granular rules is assessed by analyzing the relationship between specificity of condition and conclusion information granules. The rules emerging at the level of symbols are further interpreted by engaging linguistic approximation, which helps approximate a collection of linguistic terms of subconditions producing a linguistic summarization in the form <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">τ</i> (inputs are <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">A</i> ) consisting of a certain linguistic quantifier <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">τ</i> . The performance of summarization is provided in the form of ranking of the relevance of the rules. Experimental studies using publicly available data are completed and analyzed.

A Hybrid Rule-Based and Data-Driven Approach to Driver Modeling Through Particle Filtering

Autonomous vehicles need to model the behavior of surrounding human driven vehicles to be safe and efficient traffic participants. Existing approaches to modeling human driving behavior have relied on both data-driven and rule-based methods. While data-driven models are more expressive, rule-based models are interpretable, which is an important requirement for safety-critical domains like driving. However, rule-based models are not sufficiently representative of data, and data-driven models are yet unable to generate realistic traffic simulation due to unrealistic driving behavior such as collisions. In this paper, we propose a methodology that combines rule-based modeling with data-driven learning. While the rules are governed by interpretable parameters of the driver model, these parameters are learned online from driving demonstration data using particle filtering. We perform driver modeling experiments on the task of highway driving and merging using data from three real-world driving demonstration datasets. Our results show that driver models based on our hybrid rule-based and data-driven approach can accurately capture real-world driving behavior. Further, we assess the realism of the driving behavior generated by our model by having humans perform a “driving Turing test,” where they are asked to distinguish between videos of real driving and those generated using our driver models.

Digital twin-driven deep reinforcement learning for adaptive task allocation in robotic construction

In order to accomplish diverse tasks successfully in a dynamic (i.e., changing over time) construction environment, robots should be able to prioritize assigned tasks to optimize their performance in a given state. Recently, a deep reinforcement learning (DRL) approach has shown potential for addressing such adaptive task allocation. It remains unanswered, however, whether or not DRL can address adaptive task allocation problems in dynamic robotic construction environments. In this paper, we developed and tested a digital twin-driven DRL learning method to explore the potential of DRL for adaptive task allocation in robotic construction environments. Specifically, the digital twin synthesizes sensory data from physical assets and is used to simulate a variety of dynamic robotic construction site conditions within which a DRL agent can interact. As a result, the agent can learn an adaptive task allocation strategy that increases project performance. We tested this method with a case project in which a virtual robotic construction project (i.e., interlocking concrete bricks are delivered and assembled by robots) was digitally twinned for DRL training and testing. Results indicated that the DRL model’s task allocation approach reduced construction time by 36% in three dynamic testing environments when compared to a rule-based imperative model. The proposed DRL learning method promises to be an effective tool for adaptive task allocation in dynamic robotic construction environments. Such an adaptive task allocation method can help construction robots cope with uncertainties and can ultimately improve construction project performance by efficiently prioritizing assigned tasks.