Lieyun Ding

Decision support analysis for safety control in complex project environments based on Bayesian Networks

This paper presents a novel and systemic decision support model based on Bayesian Networks (BN) for safety control in dynamic complex project environments, which should go through the following three sections. At first, priori expert knowledge is integrated with training data in model design, aiming to improve the adaptability and practicability of model outcome. Then two indicators, Model Bias and Model Accuracy, are proposed to assess the effectiveness of BN in model validation, ensuring the model predictions are not significantly different from the actual observations. Finally we extend the safety control process to the entire life cycle of risk-prone events in model application, rather than restricted to pre-accident control, but during-construction continuous and post-accident control are included. Adapting its reasoning features, including forward reasoning, importance analysis and background reasoning, decision makers are provided with systematic and effective support for safety control in the overall work process. A frequent safety problem, ground settlement during Wuhan Changjiang Metro Shield Tunnel Construction (WCMSTC), is taken as a case study. Results demonstrate the feasibility of BN model, as well as its application potential. The proposed model can be used by practitioners in the industry as a decision support tool to increase the likelihood of a successful project in complex environments.

Feedforward Analysis for Shield-Ground System

AbstractGround surface settlement is an important measurement in identifying potential damages for shield tunneling. Identifying the relationship between shield parameters and the resulting settlement is of vital importance to the reasonable adjustment of the shield parameters so as to control settlement development. However, many other factors, besides the shield parameters, affect settlement, which makes shield-ground interaction complicated. Therefore, a better method is necessary for extracting the shield-ground relationship for the purpose of steering shield tunneling. This paper proposes a method that incorporates smooth relevance vector machine (sRVM) and particle swarm optimization (PSO) for shield steering with concern for settlement control. First, smooth relevance vector machine with adaptive Gaussian kernel function is used to establish the relationship between the identified factors and the settlement. Particle swarm optimization is then applied to identify the appropriate kernel parameters. ...

Probabilistic risk assessment of tunneling-induced damage to existing properties

There is an intrinsic risk associated with tunnel construction, particularly in urban areas where a number of third party persons and properties are involved. Due to the limited availability of data for accidents and the complexity associated with their causation, it is therefore necessary to combine available historical data and expert judgment to consider all relevant factors to undertake a realistic risk analysis. Thus, this paper presents a hybrid approach that can be used to undertake a probabilistic risk assessment of the risks associated with tunneling and its likelihood to damage to existing properties using the techniques of Bayesian Networks (BN) and a Relevance Vector Machine (RVM). A causal framework that integrates the techniques is also proposed to facilitate the development of the proposed model. The developed risk model is applied to a real tunnel construction project in Wuhan, China. The results derived from the project demonstrated the models ability to accurately assess risks during tunneling, specifically the identification of accident scenarios and the quantification of the probability and severity of possible accidents. The potential of this risk model to be used as a decision-making support tool was also explored.

Bim-Based Risk Identification System in tunnel construction

AbstractThis paper presents an innovative approach of integrating Building Information Modeling (BIM) and expert systems to address deficiencies in traditional safety risk identification process in tunnel construction. A BIM-based Risk Identification Expert System (B-RIES) composed of three main built-in subsystems: BIM extraction, knowledge base management, and risk identification subsystems, is proposed. The engineering parameter information related to risk factors is first extracted from BIM of a specific project where the Industry Foundation Classes (IFC) standard plays a bridge role between the BIM data and tunnel construction safety risks. An integrated knowledge base, consisting of fact base, rule base and case base, is then established to systematize the fragmented explicit and tacit knowledge. Finally, a hybrid inference approach, with case-based reasoning and rule-based reasoning combined, is developed to improve the flexibility and comprehensiveness of the system reasoning capacity. B-RIES is u...

Predicting Safety Risks in Deep Foundation Pits in Subway Infrastructure Projects: Support Vector Machine Approach

AbstractAccurately predicting risks during the construction of deep foundation pits is pivotal to ensuring the safety of the workforce of public and adjacent structures. Existing methods for assess...

An improved Dempster–Shafer approach to construction safety risk perception

Abstract This paper proposes a novel hybrid approach that merges fuzzy matter element (FME), Monte Carlo (MC) simulation technique, and Dempster–Shafer (D–S) evidence theory to perceive the risk magnitude of tunnel-induced building damage at an early construction stage. The membership measurement in FME is used to construct basic probability assignments (BPAs) of influential factors within different risk states. An improved evidence fusion rule that integrates the Dempster’ rule and the weighted average rule is developed to synthesize multi-source conflicting evidence. A new defuzzification method, Centre of Distribution (COD), is proposed to achieve a crisp value that represents the final safety risk perception result. A confidence indicator, δ , is put forward to measure the reliability of the safety risk perception result. A comprehensive information fusion framework that incorporates 14 influential factors is proposed to perceive the risk magnitude of tunnel-induced building damage. Six existing buildings adjacent to the excavation of Wuhan Yangtze Metro Tunnel (WYMT), China, are utilized as a case study to verify the effectiveness and applicability of the proposed approach. Results indicate that the proposed approach is capable of (i) achieving a more accurate result for safety risk perception, and (ii) identifying global sensitivities of input factors throughout a series of MC simulation enabled iterations. A discussion on how to define a reasonable membership function for configuration of BPAs is further presented. The authors recommend that the constant coefficient λ that affects the shape of the defined correlation function in BPA (Basic Probability Assignment) constructs should have a value of three, and the risk perception result can thus reach up to the highest reliability level. This approach can enable a comprehensive preliminary safety risk perception during tunnel design phases, which can further substantially reduce the risk of building damage induced by tunneling excavation.

A Web-Based System for Interface Management of Metro Equipment Engineering

In recent years, a variety of equipment automatic control systems have been applied to Chinese new metro projects such as Integrated Supervisory Control System(ISCS), Building Automation System(BAS), Fire Alarming System(FAS), etc. As with the improvement of automation in metro equipment engineering, the number of physical interfaces between sub-systems, the complexity of interface technology, the number of responsibility interfaces between different participants have increased dramatically. Traditional approaches for interface management (IM) such as interface communication meetings or interface related documents cannot adapt to modern automation level in metro equipment engineering any more. However, it remains quite common that the owners of metro projects are inexperienced in interface management and lack of standard IM approaches and effective IM tools. In order to improve the efficiency of IM and reduce the IM risks by minimizing the cost for interface conflicts and interface reworks, this paper comes up with an integrated interface model (IIM) which extracts interface information from scattered technical documents and then identifies, classifies and expresses the interface information again through structured and standardized format according to practical interface information requirement. Besides, the model proposes approaches to calculate the exact value of matching degree for technical interface and to track responsibility interface information during the whole construction stage. Based on IIM, the development and application of a web-based integrated interface management system (WIIMS) is presented in the study. The application of this system can provide technical guidance; information sharing and decision support for IM and further facilitates process control for the construction of metro equipment engineering.

Automated detection of workers and heavy equipment on construction sites: A convolutional neural network approach

Abstract Detecting the presence of workers, plant, equipment, and materials (i.e. objects) on sites to improve safety and productivity has formed an integral part of computer vision-based research in construction. Such research has tended to focus on the use of computer vision and pattern recognition approaches that are overly reliant on the manual extraction of features and small datasets (

Topological mapping and assessment of multiple settlement time series in deep excavation: A complex network perspective

Abstract This study proposed a novel methodology that integrates complex network theory and multiple time series to enhance the systematic understanding of the daily settlement behavior in deep excavation. The original time series of ground surface, surrounding buildings, and structure settlement instrumentation data over an excavation time period were measured into a similarity matrix with correlation coefficients. A threshold was then determined and binarized into adjacent matrix to identify the optimal topology and structure of the complex network. The reconstructed settlement network has nodes corresponding to multiple settlement time series individually and edges regarded as nonlinear relationships between them. A deep excavation case study of the metro station project in the Wuhan Metro network, China, was applied to validate the feasibility and potential value of the proposed approach. Results of the topological analysis corroborate a small-world phenomenon with highly compacted interactions and provide the assessment of the significance among multiple settlement time series. This approach, which provides a new way to assess the safety monitoring data in underground construction, can be implemented as a tool for extracting macro- and micro-level decision information from multiple settlement time series in deep excavation from complex system perspectives.

A Dynamic Decision Approach for Risk Analysis in Complex Projects

Underground metro tunnels present a popular solution to relieve the pressure of surface transportation systems worldwide. However, tunnel construction inevitably generates soil displacements and deformations, which may affect the safety performance of the surface road operation. This paper presents a systemic dynamic decision approach based on dynamic Bayesian networks (DBNs), aiming to provide guidelines for the dynamic safety analysis of the tunnel-induced road surface damage over time. The potential uncertainty and randomness underlying tunnel construction is modeled by following a discrete-time Markov chain process. A detailed step-by-step procedure is proposed, including risk/hazard identification, and DBN-based predictive and diagnostic analysis. A case study concerning the dynamic safety analysis in the construction of the Wuhan Yangtze Metro Tunnel is presented. Results demonstrate the feasibility of the proposed approach, as well as its application potential. The relationships between the DNB-based and BN-based approaches are further discussed based on the results. The proposed approach can be used as a decision tool to provide support for safety assurance in tunnel construction, and thus increase the likelihood of a successful project in a dynamic environment.