This thesis presents the application of data science techniques, especially machine learning, for the development of seismic damage and loss prediction models for residential buildings. Current post-earthquake building damage evaluation forms are developed for a particular country in mind. The lack of consistency hinders the comparison of building damage between different regions. A new paper form has been developed to address the need for a global universal methodology for post-earthquake building damage assessment. The form was successfully trialled in the street ‘La Morena’ in Mexico City following the 2017 Puebla earthquake. Aside from developing a framework for better input data for performance based earthquake engineering, this project also extended current techniques to derive insights from post-earthquake observations. Machine learning (ML) was applied to seismic damage data of residential buildings in Mexico City following the 2017 Puebla earthquake and in Christchurch following the 2010-2011 Canterbury earthquake sequence (CES). The experience showcased that it is readily possible to develop empirical data only driven models that can successfully identify key damage drivers and hidden underlying correlations without prior engineering knowledge. With adequate maintenance, such models have the potential to be rapidly and easily updated to allow improved damage and loss prediction accuracy and greater ability for models to be generalised. For ML models developed for the key events of the CES, the model trained using data from the 22 February 2011 event generalised the best for loss prediction. This is thought to be because of the large number of instances available for this event and the relatively limited class imbalance between the categories of the target attribute. For the CES, ML highlighted the importance of peak ground acceleration (PGA), building age, building size, liquefaction occurrence, and soil conditions as main factors which affected the losses in residential buildings in Christchurch. ML also highlighted the influence of liquefaction on the buildings losses related to the 22 February 2011 event. Further to the ML model development, the application of post-hoc methodologies was shown to be an effective way to derive insights for ML algorithms that are not intrinsically interpretable. Overall, these provide a basis for the development of ‘greybox’ ML models.
This thesis describes the management process of innovation through construction infrastructure projects. This research focuses on the innovation management process at the project level from four views. These are categorised into the separate yet related areas of: “innovation definition”, “Project time”, “project team motivation” and “Project temporary organisation”. A practical knowledge is developed for each of these research areas that enables project practitioners to make the best decision for the right type of innovation at the right phase of projects, through a capable project organisation. The research developed a holistic view on both innovation and the construction infrastructure project as two complex phenomena. An infrastructure project is a long-term capital investment, highly risky and an uncertain. Infrastructure projects can play a key role in innovation and performance improvement throughout the construction industry. The delivery of an infrastructure project is affected in most cases by critical issues of budget constraint, programme delays and safety Where the business climate is characterized by uncertainty, risk and a high level of technological change, construction infrastructure projects are unable to cope with the requirement to develop innovation. Innovation in infrastructure projects, as one of the key performance indicators (KPI) has been identified as a critical capability for performance improvement through the industry. However, in spite of the importance of infrastructure projects in improving innovation, there are a few research efforts that have developed a comprehensive view on the project context and its drivers and inhibitors for innovation in the construction industry. Two main reasons are given as the inhibitors through the process of comprehensive research on innovation management in construction. The first reason is the absence of an understanding of innovation itself. The second is a bias towards research at a firm and individual level, so a comprehensive assessment of project-related factors and their effects on innovation in infrastructure projects has not been undertaken. This study overcomes these issues by adopting as a case study approach of a successful infrastructure project. This research examines more than 500 construction innovations generated by a unique infrastructure alliance. SCIRT (Stronger Christchurch Infrastructure Rebuild Team) is a temporary alliancing organisation that was created to rebuild and recover the damaged infrastructure after the Christchurch 2011 earthquake. Researchers were given full access to the innovation project information and innovation systems under a contract with SCIRT Learning Legacy, provided the research with material which is critical for understanding innovations in large, complex alliancing infrastructure organisation. In this research, an innovation classification model was first constructed. Clear definitions have been developed for six types of construction innovation with a variety of level of novelties and benefits. The innovation classification model was applied on the SCIRT innovation database and the resultant trends and behaviours of different types of innovation are presented. The trends and behaviours through different types of SCIRT innovations developed a unique opportunity to research the projectrelated factors and their effect on the behaviour of different classified types of innovation throughout the project’s lifecycle. The result was the identification of specific characteristics of an infrastructure project that affect the innovation management process at the project level. These were categorised in four separate chapters. The first study presents the relationship between six classified types of innovation, the level of novelty and the benefit they come up with, by applying the innovation classification model on SCIRT innovation database. The second study focused on the innovation potential and limitations in different project lifecycle phases by using a logic relationship between the six classified types of innovation and the three classified phases of the SCIRT project. The third study result develops a holistic view of different elements of the SCIRT motivation system and results in a relationship between the maturity level of definition developed for innovation as one of the KPIs and a desire though the SCIRT innovation incentive system to motivate more important innovations throughout the project. The fourth study is about the role of the project’s temporary organisation that finally results in a multiple-view innovation model being developed for project organisation capability assessment in the construction industry. The result of this thesis provides practical and instrumental knowledge to be used by a project practitioner. Benefits of the current thesis could be categorized in four groups. The first group is the innovation classification model that provides a clear definition for six classified types of innovation with four levels of novelty and specifically defined outcomes and the relationship between the innovation types, novelty and benefit. The second is the ability that is provided for the project practitioner to make the best decision for the right type of innovation at the right phases of a project’s lifecycle. The third is an optimisation that is applied on the SCIRT innovation motivation system that enables the project practitioner to incentivize the right type of innovation with the right level of financial gain. This drives the project teams to develop a more important innovation instead of a simple problemsolving one. Finally, the last and probably more important benefit is the recommended multiple-view innovation model. This is a tool that could be used by a project practitioner in order to empower the project team to support innovation throughout the project.
