Deterioration prediction of reinforced concrete expressway structures attacked by corrosion based on Markov chain

The performance of reinforced concrete structures experience deterioration during their lifetime due to corrosion of steel reinforcement. Three significantdeteriorations as chloride, carbonation and moisture movement are considered in the present work for service life estimation of the expressway in Bangkok, Thailand. The first objective of this study was to predict structural service life based on inspectionresults by Deterministic model and Markov Chain model. A number of locations along expressway in Bangkok were inspected by various non-destructive testing such as covering depth, concrete properties, progress of deterioration, etc. Corrosion initiation and corrosion cracking time due to carbonation, chloride and moisture were predicted. The results shown that it has very low chance that structure will becorroded or cracked due to carbonation in 100 years after being service because carbonation depth is very small compared to covering depth. However, there is a chance that reinforcing steel will be corroded and cracked due to chloride in some structural components due to severe environment or small covering depth. Moisture is also a major cause of corrosion initiation and corrosion cracking. The results show that continuous exposure to moisture for less than a day, corrosion can be initiated. However, cracking takes a longer time due to low corrosion rate. Therefore, protection of moisture exposure is important for durability of structure. The secondobjective was to compare predicted carbonation coefficient and chloride diffusion coefficient by DPT1332 Thailand standard with actual measurement. The results show that value predicted by DPT1332 standard were higher than measured value. This is due to uncertainties of actual structure such as concrete properties, environmental conditions, moisture content, etc. The modify equation prediction of carbonation coefficient was proposed based on influential factors including carbondioxide concentration, rebound hammer value, water-binder and moisture content. From this research, deterioration and risk of infrastructure safety can be predicted based on Markov Chain. The transition matrix by Markov chain will be proposedevery five years of prediction. The prediction will be classified that followed by condition states such as no damage, corrosion initiation and corrosion cracking. The results show that around 10% (263 columns) of column 10cm in 2020 will be changed from no damage to be corroded and around 9% start to be cracked due to chloride in 2025. Around 0.5% of main crossbeam in 2020 will be changed from the no damage to be corroded and around 0.4% start to be cracked due to carbonation in 2025. Around 11% of Cantilever deck and main crossbeam (92 and 62) in 2020 will be cracked due to moisture in 2025. Finally, infrastructure maintenance can be managed effectively in Thailand.