EAPSI: Robust maintenance of urban underground infrastructure

Brownlow Andrew W, Seneca SC

Investigators

Abstract

The need to implement cost-effective, preventative measures to slow or reverse the deterioration of civil infrastructure is becoming increasingly important. However, while this need is recognized, it is challenging to efficiently implement infrastructure maintenance plans due to uncertainties inherent to the process. Traditionally, variations in engineering properties are accounted for by overdesigning, which is an inefficient use of money and resources. Use of reliability analysis has allowed for better understanding of how much overdesign is needed, but still results in inefficient designs. The next step in designing for uncertainty is to develop a method for finding a balance between cost and safety. This project will be conducted in collaboration with Professor Hongwei Huang at Tongji University in Shanghai, China, where a multi-million dollar research project on subway tunnel maintenance is being performed. Information gathered from this research project will be used to validate a tunnel maintenance decision-making model. The issue of cost-effectively preventing infrastructure deterioration is a problem of rapidly growing importance with a global impact. The uncertainties associated with the infrastructure maintenance process pose considerable challenges to the planning of maintenance programs for underground urban transportation systems. To characterize these uncertainties, reliability models have been widely studied for maintenance optimization, but accurate information is necessary for these reliability-based methods to be effective. Noise in the model inputs quickly cause suggested maintenance schemes to become inefficient and costly. Through implementation of a robust optimal maintenance decision method that considers design parameters (easy to control) and noise factors (hard to control) separately, the effect of noise factors can be minimized. This research will allow the validation of such a model, leading to enhanced decision-making capacity during maintenance planning. This NSF EAPSI award is funded in collaboration with the Chinese Ministry of Science and Technology.