Various nondestructive evaluation techniques are currently used to automatically detect and monitor cracks in concrete infrastructure. However, these methods often lack the scalability and cost-effectiveness over large geometries. A solution is the use of self-sensing carbon-doped cementitious materials. These self-sensing materials are capable of providing a measurable change in electrical output that can be related to their damage state. Previous work by the authors showed that a resistor mesh model could be used to track damage in structural components fabricated from electrically conductive concrete, where damage was located through the identification of high resistance value resistors in a resistor mesh model. In this work, an automated damage detection strategy that works through placing high value resistors into the previously developed resistor mesh model using a sequential Monte Carlo method is introduced. Here, high value resistors are used to mimic the internal condition of damaged cementitious specimens. The proposed automated damage detection method is experimentally validated using a $500 x 500 x 50 $ mm reinforced cement paste plate doped with multi-walled carbon nanotubes exposed to 100 identical impact tests. Results demonstrate that the proposed Monte Carlo method is capable of detecting and localizing the most prominent damage in a structure, demonstrating that automated damage detection in smart-concrete structures is a promising strategy for real-time structural health monitoring of civil infrastructure.