Zangeneh, P (2021) Knowledge representation and artificial intelligence for management of socio-technical risks in megaprojects. PhD thesis, University of Toronto, Canada.
Abstract
Megaprojects are probabilistically dependent systems prone to progressive failures that are undertaken in significantly incentivized economic and political domains. Current processes in definition, estimation, and financing of these large projects often exclude relevant sources of socio-technical risks with overarching effects on the project. Applications of the artificial intelligence methods in forecasting project performance and outcomes considering socio-technical sources of risks have been mainly ad-hoc solutions catered to specific needs, project types, and data availability. This research proposes a high-level framework for dealing with socio-technical risks by utilizing available sources of data, expert knowledge, and the most applicable analytics methodologies. The framework consists of representation, quantification, and inference connected through a loop of dynamic learning. The representation part defines various sources of risk in an expandable data format and with universal semantics, considering the nature of risk factors as to when and how they should be collected, in conjunction with project performance measures. An ontology is developed for such representation using the linked data and the semantic web format. The quantification aims to measure such sources of risk, which could vary from machine learning algorithms' applications over remote sensing data to pure qualitative judgments in discrete scales. The quantification was exemplified by creating a remoteness risk index, called Nighttime Remoteness Index (NIRI) for risk and resilience assessment of remote projects. The remoteness index takes nighttime satellite imagery as input and produces a remoteness index using machine learning algorithms. The index was validated based on two census bases remoteness indices of Australia and Canada. The inference part aims to incorporate the effects and dependencies of socio-technical risk factors on project outcome variables by defining an expandable object-oriented Bayesian Network (OOBN). The models can be trained based on the collected data from knowledge representation or based on expert knowledge, and different extents of their combinations. The employed methodologies allow for evergreen process of data collection and process optimization in the previous three sections. The framework creates a systematic approach to capturing and modeling project socio-technical risks, enables quality controls across the process, and can be applied to various risk sources such as the Environmental, Social, and Governance (ESG) factors.
| Item Type: | Thesis (Doctoral) |
|---|---|
| Thesis advisor: | McCabe, B Y |
| Uncontrolled Keywords: | failure; optimization; semantic web; artificial intelligence; forecasting; governance; learning; megaproject; quality control; remote sensing; Australia; Canada; project performance; quantification; machine learning |
| Index terms: | megaproject, financing, satellite, machine learning, methodology, ontology, artificial intelligence, quality control, modelling, project performance, risk factor, remote sensing, estimation, Canada, Australia, project outcome, bayesian network, semantic web, governance, forecasting, judgment, semantics, socio-technical, quantification |
| Subjects: | Geography, research methods, environmental hazards, networking, project completion, analytical methods, strategic project management, economic analysis, research products and data, project management theory and practice, education and knowledge transfer, probabilistic model, measurement and scaling, dispute resolution, business, project delivery, financial and cost management, prediction and forecasting, artificial intelligence |
| Topics: | Sustainability, Geographical Context, Project Management, Engineering Principles, Legal Issues, Quality Management, Governance, Business Strategy, Cost Management, Research Practice, Digital Applications |
| Descriptive scope: | 4 PCTA |
N.B. Descriptive scope is a count of how many of the five facets of empirical research are indicated by the words used in title, abstract and keywords. It is not intended as a judgement on the research; merely a count of the kind of word we would expect to indicate Phenomenon, Concepts, Theoretical framing, Empirical techniques, Analytical techniques. If all five are present, then a code of “5 PCTEA” will indicate this. If you feel the coding for this record is questionable, we welcome discussion around the terms we matched or the way we categorized them. The facet you would expect may not be coded, or a facet may be coded inappropriately. This can also bear on a larger question, of which facets should be treated as defining in construction management research. Please get in touch, and we will look at it. More details here