A case study of utilising digitalisation techniques to achieve efficiency and carbon benefits during piling projects

This paper demonstrates how digitalisation techniques can be used to create a database of piling projects, from which valuable sustainability and cost reduction insights can be drawn. Machine learning and Artificial Intelligence (AI) are applied to analyse and make these insights that can be used in the pile design decision making process. Python and Optical Character Recognition (OCR) techniques are used to rapidly extract the relevant data from the site records of over 1,000 piles from more than 10 historic piling jobs and compile them into a single database. This process is both quicker and less error prone compared to traditional techniques. By applying carbon factors to the material quantities, we are able to compare projects based on carbon used per m2 GIA. The compilation of data enables the detailed analysis of various factors, including piling technique, in-situ ground conditions and pile geometry, which combined with engineering knowledge allowed insights to be made into the sustainability of different design decisions and their outcomes such as on the amount of overbreak . The quantity of concrete used when piling correlates well with the amount of carbon emissions produced during piling works. This paper also compares the carbon emissions from driven pile techniques and the construction of other embedded retaining wall structures. The insights from building a piling database like this can facilitate collaboration between designers and contractors in developing pile designs that are both sustainable and economical by reducing the amount of material usage on site. Increasing the size of the database over time will further benefit the decision-making process and increase the reliability of the results, enabling greater efficiencies in pile design and reducing the embodied carbon of piled foundations.