Development of Machine Learning Models for Soil Predictions based on CPT Measurements and Preliminary Research and Creation of Framework for Assessing Machine Learning Projects in AEC In a Perspective of mulidisiplinary consultancies and Change management in AEC

Abstract

Strategic digital transformation of civil engineering (CE) consulting firms in architecture, engineering, and construction (AEC) requires the implementation of business innovation and execution models more than exciting new technology (Kane, Palmer, Phillips, Kiron, & Buckley, 2015). Which models and how to employ them involves understanding the niche industry in question, the potential trajectory of the industry, current tools and methods, and how people and resources apply. Today a gap exists between civil engineers and cutting-edge technology and knowledge management. Newer technology does not allow for civil engineers to sit idly by as it once did. Instead, they must adapt and be open to educating themselves as the industry progresses. Through development in this thesis of a machine learning model for predicting soil based on data from the equipment used in ground surveys with lab reports as ground truth labels and the start of a preliminary theoretical framework to identify and rank the feasibility for potential machine learning problems. This thesis will develop and propose a substantial step forward for AEC multidisciplinary consulting firms that navigates to potential desired outcomes by providing a deeper understanding of the worth of data and what is leading in the implementation of new technology such as artificial intelligence.