Markov Chain Performance Modeling for Slurry Seal and Thin Asphalt Overlay Using Bayesian Approach
I’m glad to announce that our paper on development on pavement performance models for preventative maintenance using Bayesian approach has been presented for the 12th International Congress on Civil Engineering.
This abstract of this paper is as follows:
Preventative maintenance strategies have been commonly used by highway agencies to preserve pavements. For optimal preventative maintenance treatments, highway agencies need to properly incorporate the effect of these treatments into the pavement performance models, for which lack of sufficient historic performance data is the main problem. Markov chain models have been successfully used to circumvent this problem. The core of Markov chain prediction model is its transition probability matrix (TPM) which provides the probability of transiting the condition from one state to the next or remaining in the same state. Therefore, it is important to be able to update this matrix as more actual performance data is accumulated. This paper uses an approach to update the TPM using Bayesian theorem. This approach was used for developing the TPMs for Markovian performance models for slurry seal and thin asphalt overlay using the Long-term Pavement Performance database. The existing pavements condition prior to the treatment application was used as the main criterion to distinguish six pavement families. Then, for these pavement families, Markov chain models were developed and validated. Finally, the maintenance performance models and the net present value for slurry seal and thin overlay treatments were assessed.
Authors: Masih Beheshti, Prof. Nader Tabatabaee