Search

The use of Building Information Modelling (BIM) for construction project risk management has become a growing research trend. However, it was observed that BIM-based risk management has not been widely used in practice and two important gaps leading to this problem are: 1) very few theories exist that can explain how BIM can be aligned with traditional techniques and integrated into existing processes for project risk management; and 2) current BIM solutions have very limited support on risk communication and information management during the project development process. To overcome these limitations, this research proposes a new approach that two traditional risk management techniques, Risk Breakdown Structure (RBS) and Case-based Reasoning (CBR), can be integrated into BIM-based platforms and an active linkage between the risk information and BIM can be established to support the project lifecycle. The core motivations behind the proposed solution are: 1) a tailored RBS could be used as a knowledge-based approach to classify, store and manage the information of a risk database in a proper structure and risk information in RBS could be linked to BIM for review, visualisation and communication; and 2) knowledge and experience stored in past risk reports could contribute to avoiding similar risks in new situations and the most relevant cases can be linked to BIM to support decision making during the project lifecycle. The scope of this research is limited to bridge projects; however, the basic methods and principles could be also applied to other types of projects. This research is in three phases. In the first stage, this research analysed the conceptual separation of BIM and the linkage rules between different types of risk and BIM. Specifically, an integrated bridge information model was divided into four Level of Contents (LOCs) and six technical systems based on the analysis of the Industry Foundation Classes (IFC) specification, a critical review of previous studies and the author’s project experience. Then a knowledge-based risk database was developed through an extensive collection of risk data, a process of data mining, and further assessment and translation of the data. Built on the risk database, a tailored RBS was developed to categorise and manage this risk information and a set of linkage rules between the tailored RBS and the four LOCs and six technical systems of BIM was established. Secondly, to further implement the linkage rules, a novel method to link BIM, RBS, and Work Breakdown Structure (WBS) to be a risk management system was developed. A prototype system was created based on Navisworks and the Microsoft SQL Server to support the implementation of the proposed approach. The system allows not only the storage of risk information in a central database but also to link the related risk information in the BIM model for review, visualisation and simulation. Thirdly, to facilitate the use of previous knowledge and experience for BIM-based risk management, the research proposed an approach of combining the use of two Natural Language Processing (NLP) techniques, i.e. Vector Space Model (VSM) and semantic query expansion, and outlined a new framework for the risk case retrieval system. A prototype was developed using the Python programming language to support the implementation of the proposed method. Preliminary testing results show that the proposed system is capable of retrieving relevant cases automatically and to return, for example, the top 10 similar cases. The main contribution of this research is the approach of integrating RBS and CBR into BIM through active linkages. The practical significance of this research is that the proposed approach enables the development of BIM-based risk management software to improve the risk identification, analysis, and information management during the project development process. This research provides evidence that traditional techniques can be aligned with BIM for risk management. One significant advantage of the proposed method is to combine the benefits of both traditional techniques and BIM for lifecycle project risk management and have the minimum disruption to the existing working processes.

Insects, which represent the most species-rich taxa, are extremely important ecosystem components. The diversity patterns of insects have, however, been widely ignored in biodiversity research. In my thesis, I aim to establish a basic understanding of the diversity patterns of insect assemblages in the temperate forest and forest plantation ecosystems of Northern China, and to investigate how these patterns correlate with vegetation and environmental conditions. The study aims to give further insights into the insect diversity status and measures to conserve or even enhance their diversity in the large secondary and plantation forests which have been and are currently established throughout northern China. The study focuses on two distinct insect taxa: ground beetles (Coleoptera: Carabidae) and geometrid moths (Lepidoptera: Geometridae). In the main study area located within the Changbaishan Natural Reserve (CNR) in Jilin Province, 4844 individuals (47 species) of ground beetle and 9285 individuals (155 species) of geometrid moth were sampled. In addition, 1488 ground beetles (24 species) and 2047 geometrid moths (165 species) were sampled in the secondary and plantation forest area at Dongling Mountain (DLM) in Beijing. A first important result of this work is that the α-diversity of both ground beetle and geometrid moth assemblages decreased significantly with increasing elevation at CNR. My results also show that the relationships between phyto-diversity and the diversity of insects are weak and furthermore likely to be driven by underlying environmental factors. The significant changes which have recently occurred in the plant species composition at CNR chiefly related to changes in the climatic conditions suggest that insect species are also under high pressure in this area. Finally, this study suggests that in the temperate religions of Northern China, secondary and plantation forests can potentially harbour high levels of insect α-diversity compared with mature, more pristine forests.