Prediction of back-calculated layer moduli using cuckoo search algorithm for pavement asset management at a network level

The back-calculation of layer moduli using falling weight deflectometer (FWD) data is an essential part of the pavement maintenance measures in pavement asset management systems (PAMS). It is necessary to provide the back-calculation predicted layer moduli and other pavement distresses as a part of PAMS to estimate the optimal maintenance strategy. The predicted layer moduli can be introduced as a part of PAMS if the back-calculation model can be easily integrated. In this study, the cuckoo search algorithm (CSA) was used as an optimization tool to develop a back-calculation model, BACKCSA, for predicting the pavement layer moduli using FWD data. The developed model was validated by comparing it with the laboratory-measured resilient moduli (M_R) values of the field core samples obtained from five different highway sections. The variation between the laboratory M_R values and the BACKCSA model predicted layer moduli was marginal. The pavement layer moduli values obtained from the BACKCSA model were also compared with the back-calculated layer moduli obtained using the BAKFAA model. The statistical hypothesis testing revealed that the predicted layer moduli from BACKCSA and BAKFAA models were similar to the laboratory-measured M_R values. The mean absolute percentage error (MAPE) between the BACKCSA model predicted layer moduli and the laboratory-measured M_R values was 2.49% on an average, indicating a marginal error between the predicted and the measured values. One of the most significant benefits of using the BACKCSA model over the other back-calculation models is its ability to handle deflection data from any FWD equipment, in general.