While the subsurface, to date, lacks clear ordering principles and design rules, there is an increasing need to more systematically plan infrastructures in the buried shallow subsurface space. This project will explore how existing (often implicit) rule sets can be integrated in a design algorithm that supports further automated and more systematic design of cables and pipes in urban undergrounds.

Before detailed construction programming takes place, engineering drawings for low voltage energy networks are being checked based on constructability criteria. This study aims to support these checks, enabling early-stage constructability analyses, by automated geometric and semantic rule checking in GIS.

Excavation damages are viewed from different perspectives. Operational workers close to the trench have a different perception of how these damages can be prevented (or sometimes justified) compared to managers and clients who work at offices and are mere distant from the construction site. What influences these perspectives? And how does their background play a role? This M.Sc. research sheds light on this dynamic to gain insights into thought and work processes.

This project aimed to support the detection of risky excavation operations based on historical datasets of damages in the Netherlands. Jiarong Li developed, using XGBoost, a data-driven model that predicted the likelihood of damage occurring in an excavation polygon. The Dutch agency Kadaster uses such polygons to exchange data on utility locations between network owners and excavator operators.

The machine learning model had a satisfactory performance with an AUC-ROC score of 0.821 and a balanced accuracy of 0.743.

This study focuses on translating the underground spatial impact of redevelopments in public space, related to heat transition, electric grid upgrading, and climate resilience.

Utility lines can be extracted from point clouds to support further 3D modelling and mapping. Doing this manually is time consuming, Jorn Kruiper investigated how to use machine learning for this task.

Therefore, Dutch law (WIBON) mandates practitioners to reduce the uncertainty in the information about the subsurface situation through the process of utility surveying. One of the most promising non-disruptive methods is the Ground Penetrating Radar (GPR). GPR is a geophysical method that uses electromagnetic waves to investigate the subsurface. This study aims to explore, explain and support the decision making between trial trenches and GPR as a utility surveying method by developing, implementing and evaluating a Decision Support Tool.

To scale up city district heating construction projects, processes need to be developed and reliable. Currently teams largely rely on experience and improvisation, making them skilled in troubleshooting, but less efficient in planning and anticipating disruptions. This thesis has explored anticipation and containment through the lens of HRO

Smart Utility Registration aims to develop a method for scanning trench data in more automated ways. It mobilizes off-the-shelf (preferably open sources) technologies to register deployed cables/pipelines. Currently, this process is laborious and involves many different workers, while new technology might enable direct registration by the jobsite crew.

This project ran from 2019-2021 and resulted in a prototype Decision Support System, which provides decision-makers with insight in the optical fiber deployment process as well as the results of operational changes they may wish to make