Developing Penetration Rate Prediction Models to Enhance the Productivity Prediction of Microtunneling Construction Projects

• Author / Creator

  Moharrami, Saeid

• Predicting the productivity of microtunnelling construction projects is challenging due to the inherent complexities of this trenchless excavation method. One of these complexities has to do with estimating the micro-tunnel boring machine penetration rate, given that it is subject to a number of factors, including variations in ground conditions during microtunneling, uncertainty regarding underground conditions along the tunnel path, and the complexity of the mechanism underlying micro-tunnel boring machine excavation. A review of literature on predicting the penetration rate reveals a number of gaps with respect to the prediction of micro-tunnel boring machine penetration rate, including (1) the lack of a robust method for the use of machine-generated data for dynamically updating the project progress, (2) limited research on small-diameter microtunneling excavation through soft ground conditions, (3) the lack of mechanistic investigation of the mechanism governing micro-tunnel boring machine penetration into the soil, as well as the lack of a theoretical mechanistic relationship by which to quantify the influence of primary factors such as soil type, operational loads, and cutterhead characteristics, (4) the lack of integrated models by which to reduce uncertainty of micro-tunnel boring machine penetration rate in simulation-based microtunneling productivity studies, and (5) the lack of penetration rate prediction models suited for dynamic utilization during construction to update the penetration rate predictions and modify the project plan accordingly.The research presented in this thesis to enhance the prediction of micro-tunnel boring machine penetration rate and productivity in microtunneling construction, proceeding in three phases. In Phase 1, a dynamic penetration rate prediction model that uses a machine-learning approach is developed. Phase 2 involves the development of a mechanistic approach for modelling micro-tunnel boring machine penetration into soft ground. In this regard, a novel mechanistic approach (based on the contact mechanics theory) by which to model the interaction between the micro-tunnel boring machine and the ground is introduced. The developed mechanistic model is further improved by modelling in greater detail the micro-tunnel boring machine’s engagement with the ground, taking into consideration in particular the engagement between cutting blade and soil, and quantifying the influence of this engagement on the micro-tunnel boring machine’s penetration rate. In Phase 3, to enhance the production rate estimation in microtunneling construction projects, the micro-tunnel boring machine penetration rate prediction models developed in Phases 1 and 2 are integrated with simulation models. Two approaches are followed for integrating the mechanistic model for prediction of micro-tunnel boring machine penetration rate (developed in Phase 2) with operation simulation. The first approach is to use the exact mechanistic formula and incorporate it into the simulation model, while the second approach is to enhance the prediction made by the mechanistic model by leveraging observations (excavation times) made during construction and updating the initial predicted distribution of penetration rate accordingly. To integrate the dynamic machine-learning model for prediction of micro-tunnel boring machine penetration rate (developed in Phase 1) with an operation simulation model, a database of results is integrated with a simulation model. Whenever the micro-tunnel boring machine reaches specific locations along the tunnel in the simulation, it calls up the predicted penetration rate to be used for modelling excavation, and in this manner the entire microtunneling operation is simulated.The feasibility and functionality of the developed models for predicting micro-tunnel boring machine penetration rate (as well as the prediction models integrated with simulation) are validated using both actual case studies and a synthetic dataset of fifty microtunneling projects generated using the Monte Carlo approach. Ultimately this research provides practitioners and researchers with a systematic procedure for using machine-generated data and available geotechnical information during tunnelling to achieve more accurate prediction of micro-tunnel boring machine performance in dynamic geological conditions, and to update the project progress dynamically based on what is actually occurring on site. Furthermore, this research proposes a novel approach for mechanistic analysis of the interaction between the micro-tunnel boring machine and the ground, and develops a mechanistic model for micro-tunnel boring machine penetration rate that characterizes in a quantitative manner the relationship between penetration rate and the combined influence of three primary factors—soil properties, operational loads, and cutterhead characteristics.

• Subjects / Keywords

  • Penetration Rate
  • Microtunneling
  • Productivity
  • Mechanistic Approach
  • Dynamic
  • Data-Driven approach
  • Simulation
  • Trenchless

• Graduation date

  Spring 2023

• Type of Item

  Thesis

• Degree

  Doctor of Philosophy

• DOI

  https://doi.org/10.7939/r3-asxg-6s68

• License

  This thesis is made available by the University of Alberta Libraries with permission of the copyright owner solely for non-commercial purposes. This thesis, or any portion thereof, may not otherwise be copied or reproduced without the written consent of the copyright owner, except to the extent permitted by Canadian copyright law.