4.6 Article

Multiagent Reinforcement Learning for Project-Level Intervention Planning under Multiple Uncertainties

Related references

Note: Only part of the references are listed.
Article Management

Using Bayesian networks to estimate bridge characteristics in early road designs

Vassilis Panopoulos et al.

Summary: This study presents a methodology for estimating the characteristics of new bridges using a bridge database and Bayesian networks. It demonstrates that this methodology can provide quick and reliable estimates of bridge characteristics based on limited situational information. The results show that the proposed methodology has a higher prediction accuracy compared to naive-search databases queries, and Bayesian networks based on expert experience can achieve even better results.

INFRASTRUCTURE ASSET MANAGEMENT (2022)

Review Construction & Building Technology

Project Level Management Decisions in Construction and Rehabilitation of Flexible Pavements

Ahmed S. Mohamed et al.

Summary: This study presents and discusses the diverse perceptions and methodologies for the management system elements in the construction and rehabilitation of flexible pavements. It highlights the need for further investigations to develop more efficient systems that optimize environmental, economic, and social objectives.

AUTOMATION IN CONSTRUCTION (2022)

Article Construction & Building Technology

Upscaling Complex Project-Level Infrastructure Intervention Planning to Network Assets

Vahid Asghari et al.

Summary: This paper presents a new machine learning-based methodology to estimate optimal intervention timings and demonstrates its feasibility and efficiency through experiments. The proposed method can assist practitioners in making better decisions in asset management systems and furthering sustainability objectives.

JOURNAL OF CONSTRUCTION ENGINEERING AND MANAGEMENT (2022)

Article Management

Modelling the uncertainty of cost and t me in highway projects

Alireza Moghayedi et al.

Summary: This study developed a hybrid intelligent tool to model uncertainties in linear infrastructure projects and measure their impact on cost and time. The study found that the accumulated impact of uncertainties significantly increased the construction cost and time of infrastructure projects. Therefore, improving the accuracy of cost and time estimation for highway projects requires a combination of classical and intelligent prediction techniques.

INFRASTRUCTURE ASSET MANAGEMENT (2022)

Article Engineering, Industrial

A digital twin to train deep reinforcement learning agent for smart manufacturing plants: Environment, interfaces and intelligence

Kaishu Xia et al.

Summary: The paper proposes a data-driven approach to automate smart manufacturing systems by utilizing digital twin and Deep Reinforcement Learning. It demonstrates the integration of smart agents into industrial platforms to expand the usage of system-level digital twin and improve the digital transformation of the manufacturing industry. The study highlights the potential of combining data science and manufacturing industries through the DRL approach to automated manufacturing control problems.

JOURNAL OF MANUFACTURING SYSTEMS (2021)

Article Computer Science, Artificial Intelligence

Deep reinforcement learning for drone navigation using sensor data

Victoria J. Hodge et al.

Summary: Mobile robots like drones play a crucial role in surveillance, monitoring, and data collection in buildings, infrastructure, and environments. The research aims to achieve accurate and rapid problem localization through flexible, autonomous, and powerful decision-making mobile robots. The implementation of a generic adaptive navigation algorithm using deep reinforcement learning algorithms aims to improve accuracy and efficiency.

NEURAL COMPUTING & APPLICATIONS (2021)

Article Engineering, Civil

Optimal Timing of the Seismic Vulnerability Reduction Measures Using Real Options Analysis

Vahid Asghari et al.

Summary: This study proposed an investment valuation framework based on real options analysis to determine the optimal timing for implementing seismic vulnerability reduction measures, aiming to help communities reduce overall costs and enhance resilience.

ASCE-ASME JOURNAL OF RISK AND UNCERTAINTY IN ENGINEERING SYSTEMS PART A-CIVIL ENGINEERING (2021)

Article Computer Science, Interdisciplinary Applications

Flood mitigation in coastal urban catchments using real-time stormwater infrastructure control and reinforcement learning

Benjamin D. Bowes et al.

Summary: This research explores the use of reinforcement learning (RL) to create control policies in a virtual urban catchment model, successfully reducing flood risk and comparing different control strategies. RL can proactively manage pond levels based on current and forecast conditions, reducing flood events effectively.

JOURNAL OF HYDROINFORMATICS (2021)

Article Management

Quantifying net benefits of intervention programmes to enable their digitalised generation

Marcel Burkhalter et al.

Summary: In order to ensure infrastructure networks provide expected service, interventions are necessary. By considering service provided over time and intervention costs, a systematic method utilizing advanced algorithms helps generate optimal intervention programs for infrastructure management systems.

INFRASTRUCTURE ASSET MANAGEMENT (2021)

Article Energy & Fuels

Real-Time Autonomous Residential Demand Response Management Based on Twin Delayed Deep Deterministic Policy Gradient Learning

Yujian Ye et al.

Summary: This paper introduces a novel real-time demand response management strategy for residential households based on the twin delayed deep deterministic policy gradient (TD3) learning approach, which enables learning of neural-network-based and fine-grained demand response management policies in a multi-dimensional action space by exploiting high-dimensional sensory data.

ENERGIES (2021)

Article Engineering, Industrial

Prioritizing Postdisaster Recovery of Transportation Infrastructure Systems Using Multiagent Reinforcement Learning

Pedram Ghannad et al.

Summary: Postdisaster reconstruction of transportation infrastructures involves complex planning and implementation options under uncertainty. This paper introduces a new prioritization approach for rapid and optimized postdisaster recovery, using a multiagent system with reinforcement learning technique to evaluate recovery priorities of damaged transportation infrastructure systems. The proposed model aims to minimize social impact and help public agencies make informed decisions for distributing resources and structurally arranging disaster recovery processes of transportation systems.

JOURNAL OF MANAGEMENT IN ENGINEERING (2021)

Article Computer Science, Artificial Intelligence

Deep reinforcement learning based home energy management system with devices operational dependencies

Caomingzhe Si et al.

Summary: This paper proposes an energy management strategy for controllable loads based on reinforcement learning, utilizing mathematical models and two RL algorithms to achieve proper scheduling of home energy resources based on residents' living habits and dependency modes. Through case studies, it is verified that the proposed method can effectively schedule HERs to meet established dependency modes with a relatively small difference from the optimal solution.

INTERNATIONAL JOURNAL OF MACHINE LEARNING AND CYBERNETICS (2021)

Article Engineering, Industrial

Expediting Life Cycle Cost Analysis of Infrastructure Assets under Multiple Uncertainties by Deep Neural Networks

Vahid Asghari et al.

Summary: This study proposes a method to replace simulation modules of optimization algorithms with trained machine learning models, which can accelerate the computation speed of complex AMSs and make them applicable to all assets in a network.

JOURNAL OF MANAGEMENT IN ENGINEERING (2021)

Article Engineering, Industrial

Bayesian Monte Carlo Simulation-Driven Approach for Construction Schedule Risk Inference

Long Chen et al.

Summary: This research aims to develop a novel Bayesian Monte Carlo simulation-driven approach for construction schedule risk inference of infrastructures. The study constructs a key risk network based on network theory analysis, develops the DAG structure of a Bayesian network using DFS and A-MWST algorithms, and further develops CPTs using the leaky-MAX model. The results show that the developed approach is more appropriate to deal with risk inference of infrastructures considering its reliability, convenience, and flexibility.

JOURNAL OF MANAGEMENT IN ENGINEERING (2021)

Article Construction & Building Technology

Anomaly detection using state-space models and reinforcement learning

Shervin Khazaeli et al.

Summary: Early detection of anomalies associated with changes in structure behavior is important for ensuring their serviceability and safety. Existing anomaly detection strategies usually rely on univariate threshold values and tend to overlook the planning horizon in decision making.

STRUCTURAL CONTROL & HEALTH MONITORING (2021)

Article Engineering, Environmental

Minimizing the global warming impact of pavement infrastructure through reinforcement learning

Sophie Renard et al.

Summary: Lifecycle assessment (LCA) studies are commonly used to evaluate environmental impacts of pavement facilities. This study introduces a novel approach to LCA modeling using a type of reinforcement learning algorithm known as Q-learning, which has shown to reduce global warming impacts of pavement infrastructure by 13% to 18% in three representative case studies. This approach can help decision-makers optimize construction and maintenance plans while considering uncertainties and improve management strategies to mitigate environmental impacts.

RESOURCES CONSERVATION AND RECYCLING (2021)

Article Energy & Fuels

Optimal planning of a rooftop PV system using GIS-based reinforcement learning

Seunghoon Jung et al.

Summary: The study developed a GIS-based RL model for optimal planning of rooftop PV systems, showing its economic profitability in residential buildings in Nonhyeon district, South Korea. The model outperformed existing methods, demonstrating its potential for optimizing economic feasibility and promoting the adoption of rooftop PV systems.

APPLIED ENERGY (2021)

Article Construction & Building Technology

Autonomous construction hoist system based on deep reinforcement learning in high-rise building construction

Dongmin Lee et al.

Summary: The researchers developed an autonomous hoist system supported by deep reinforcement learning to address the issue of unnecessary trips caused by manual control, improving lifting performance and shortening project schedules.

AUTOMATION IN CONSTRUCTION (2021)

Article Construction & Building Technology

An open-source and extensible platform for general infrastructure asset management system

Vahid Asghari et al.

Summary: This paper introduces GIAMS, an open-source platform developed in Python, to facilitate collaboration and contribution to future research in asset management. The modular and open-source nature of GIAMS allows for easy extension in various asset management topics.

AUTOMATION IN CONSTRUCTION (2021)

Article Construction & Building Technology

Reinforcement learning based process optimization and strategy development in conventional tunneling

Georg H. Erharter et al.

Summary: This paper introduces a novel RL based framework for strategy development for conventional tunneling, where an agent learns to achieve a tunnel breakthrough goal by interacting with the environment. Result analysis shows that optimal policies have similarities to current practices, and the framework has the potential to discover new tunneling strategies.

AUTOMATION IN CONSTRUCTION (2021)

Article Management

Exploiting digitalisation to plan interventions on large water distribution networks

Sean Kerwin et al.

Summary: Cities heavily rely on water distribution networks for services. This paper discusses applying a methodology to larger networks, demonstrating advantages through three applications: estimating budget requirements for interventions, communicating project-level trade-offs, and investigating the impacts of intervention planning activities from other networks.

INFRASTRUCTURE ASSET MANAGEMENT (2021)

Article Management

Decision making methods to prioritise asset-management plans for municipal infrastructure

Luis Amador-Jimenez et al.

Summary: This study proposes a methodology for measuring and comparing the benefits of changing the decision-making method for infrastructure asset management, using a case study to illustrate the impact of different decision-making approaches and economic comparisons to support the selection of the preferred method. Results show the expected inferiority of the worst-first method and the trade-off method resulting in lower resource expenditure than the silo method, with higher condition levels achieved.

INFRASTRUCTURE ASSET MANAGEMENT (2021)

Article Engineering, Civil

Optimal policy for structure maintenance: A deep reinforcement learning framework

Shiyin Wei et al.

STRUCTURAL SAFETY (2020)

Article Computer Science, Artificial Intelligence

Deep neural network based framework for complex correlations in engineering metrics

Vahid Asghari et al.

ADVANCED ENGINEERING INFORMATICS (2020)

Article Construction & Building Technology

Complete coverage path planning using reinforcement learning for Tetromino based cleaning and maintenance robot

Anirudh Krishna Lakshmanan et al.

AUTOMATION IN CONSTRUCTION (2020)

Article Computer Science, Interdisciplinary Applications

Deep reinforcement learning for long-term pavement maintenance planning

Linyi Yao et al.

COMPUTER-AIDED CIVIL AND INFRASTRUCTURE ENGINEERING (2020)

Article Engineering, Civil

Stochastic optimal control methodologies in risk-informed community resilience planning

Saeed Nozhati et al.

STRUCTURAL SAFETY (2020)

Article Construction & Building Technology

Deep reinforcement learning to optimise indoor temperature control and heating energy consumption in buildings

Silvio Brandi et al.

ENERGY AND BUILDINGS (2020)

Review Automation & Control Systems

Deep Reinforcement Learning for Multiagent Systems: A Review of Challenges, Solutions, and Applications

Thanh Thi Nguyen et al.

IEEE TRANSACTIONS ON CYBERNETICS (2020)

Article Construction & Building Technology

Teaching robots to perform quasi-repetitive construction tasks through human demonstration

Ci-Jyun Liang et al.

AUTOMATION IN CONSTRUCTION (2020)

Article Energy & Fuels

Reinforcement Learning-Based School Energy Management System

Yassine Chemingui et al.

ENERGIES (2020)

Review Computer Science, Information Systems

Deep Reinforcement Learning for Traffic Signal Control: A Review

Faizan Rasheed et al.

IEEE ACCESS (2020)

Review Construction & Building Technology

Pavement asset management systems and technologies: A review

Naga Siva Pavani Peraka et al.

AUTOMATION IN CONSTRUCTION (2020)

Article Management

Defining and quantifying railway service to plan infrastructure interventions

Natalia Papathanasiou et al.

INFRASTRUCTURE ASSET MANAGEMENT (2020)

Article Management

Usefulness of quantifying effects on rail service when comparing intervention strategies

Natalia Papathanasiou et al.

INFRASTRUCTURE ASSET MANAGEMENT (2020)

Article Engineering, Civil

A Scalable Reinforcement Learning Algorithm for Scheduling Railway Lines

Harshad Khadilkar

IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS (2019)

Article Construction & Building Technology

Energy optimization associated with thermal comfort and indoor air control via a deep reinforcement learning algorithm

William Valladares et al.

BUILDING AND ENVIRONMENT (2019)

Article Engineering, Industrial

Managing engineering systems with large state and action spaces through deep reinforcement learning

C. P. Andriotis et al.

RELIABILITY ENGINEERING & SYSTEM SAFETY (2019)

Review Chemistry, Multidisciplinary

Optimization in Decision Making in Infrastructure Asset Management: A Review

Lin Chen et al.

APPLIED SCIENCES-BASEL (2019)

Article Engineering, Electrical & Electronic

Deep Reinforcement Learning Based Resource Allocation for V2V Communications

Hao Ye et al.

IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY (2019)

Article Construction & Building Technology

Impact assessment of reinforced learning methods on construction workers' fall risk behavior using virtual reality

Yangming Shi et al.

AUTOMATION IN CONSTRUCTION (2019)

Article Construction & Building Technology

LightLearn: An adaptive and occupant centered controller for lighting based on reinforcement learning

June Young Park et al.

BUILDING AND ENVIRONMENT (2019)

Article Management

Multivariate simulation modelling for adaptive long-term infrastructure planning

Mark Havelaar et al.

INFRASTRUCTURE ASSET MANAGEMENT (2019)

Article Construction & Building Technology

Deep Reinforcement Learning Algorithms in Intelligent Infrastructure

Will Serrano

INFRASTRUCTURES (2019)

Article Construction & Building Technology

Optimal control of HVAC and window systems for natural ventilation through reinforcement learning

Yujiao Chen et al.

ENERGY AND BUILDINGS (2018)

Article Engineering, Civil

Multi-criteria robust optimization framework for bridge adaptation under climate change

Alysson Mondoro et al.

STRUCTURAL SAFETY (2018)

Review Engineering, Civil

Bridge Adaptation and Management under Climate Change Uncertainties: A Review

Alysson Mondoro et al.

NATURAL HAZARDS REVIEW (2018)

Article Engineering, Civil

Bridge inspection practices and bridge management programs in China, Japan, Korea, and U.S.

Yoseok Jeong et al.

JOURNAL OF STRUCTURAL INTEGRITY AND MAINTENANCE (2018)

Article Multidisciplinary Sciences

Mastering the game of Go with deep neural networks and tree search

David Silver et al.

NATURE (2016)

Article Energy & Fuels

Reinforcement learning for optimal control of low exergy buildings

Lei Yang et al.

APPLIED ENERGY (2015)

Article Engineering, Civil

Investigating railway track asset management using a Markov analysis

Darren Prescott et al.

PROCEEDINGS OF THE INSTITUTION OF MECHANICAL ENGINEERS PART F-JOURNAL OF RAIL AND RAPID TRANSIT (2015)

Article Engineering, Industrial

Planning structural inspection and maintenance policies via dynamic programming and Markov processes. Part II: POMDP implementation

K. G. Papakonstantinou et al.

RELIABILITY ENGINEERING & SYSTEM SAFETY (2014)

Article Construction & Building Technology

Parallel Computing Platform for Multiobjective Simulation Optimization of Bridge Maintenance Planning

I-Tung Yang et al.

JOURNAL OF CONSTRUCTION ENGINEERING AND MANAGEMENT (2012)

Article Automation & Control Systems

On actor-critic algorithms

VR Konda et al.

SIAM JOURNAL ON CONTROL AND OPTIMIZATION (2003)