LAUSR

Abstract A realistic assessment of the time and cost to complete a tunnel using tunnel boring machine (TBM) is directly related to the ability to accurately predict machine advance rate. Advance rate depends on various parameters including rock and rock mass properties, machine characteristics, and working conditions. Review of available literature shows that advance rate and utilization are much less studied than penetration rate, perhaps due to the difficulties in evaluating the interaction between input parameters, which make the advance rate hard to predict. The TBM advance rate can be estimated from either knowledge of penetration rate and utilization or directly from an evaluation of machine, ground, and operational factors by applying empirical formulas. This paper reviews different methodologies used in the prediction of advance rate and highlights potential pitfalls in the process. For this purpose, the capabilities of some commonly applied prediction models are discussed and their validity is examined using a comprehensive database of recently completed tunnels. The results of a comparing the estimated and field/observed advance rates indicate that most predictive models overestimate the performance. The results underline the current shortcomings of TBM advance rate estimation. A new model that is developed based on statistical analysis of a database of more than 300 TBM projects records is introduced for direct prediction of advance rate. The results show, the proposed new model’ estimated advance rate values are in good agreement with the actual values with a high coefficient of determination (79%). This clearly indicates a step forward in improving hard rock TBMs’ performance and project planning evaluation.