Assessing the structural condition of pavements through nondestructive testing (NDT) at the network level is essential for effective maintenance management. To facilitate proactive maintenance, road agencies require consistent and reliable threshold criteria for evaluating pavement structural conditions. This study focuses on defining and benchmarking threshold values for structural indices derived from the Falling Weight Deflectometer (FWD), a widely used NDT device. In Cambodia, where distress surveys and historical data are limited, we employ a stochastic approach using the Markov hazard model (MHM) to predict the surface deterioration, expressed through the International Roughness Index (IRI). Our predictions are based on two sets of IRI measurements from pavements classified as Double Bituminous Surface Treatment (DBST) and Asphalt Concrete (AC), taken during initial and follow-up inspections conducted one year apart (2021-2022 or 2022-2023). By converting continuous IRI values into discrete condition states, the MHM addresses the probabilities of transitioning between these states, following an exponential hazard function. The model incorporates critical factors such as bearing capacity (represented by structural indices), traffic loading (in terms of Equivalent Single Axle Loads, ESALs), and environmental influence (such as average annual rainfall) to enhance prediction accuracy. Key structural indices derived from FWD testing include maximum deflection, effective structural number, subgrade resilient modulus, and deflection bowl parameters (Base Layer Index, Middle Layer Index, and Lower Layer Index), which are expected to correlate strongly with IRI deterioration. Bayesian inference and the Markov Chain Monte Carlo (MCMC) technique are used to estimate the unknown parameters, calculate deterioration rates, and generate deterioration curve (life expectancy). The structural indices values corresponding to the life expectancy of DBST and AC pavements are identified as critical thresholds of each pavement type. These thresholds are also determined for varying levels of traffic load and rainfall. As expected, the results indicate that DBST pavements deteriorate faster than AC pavements, leading to more conservative threshold criteria for AC. Finally, we establish benchmarking criteria for structural indices expressed as health ratings, ensuring uniform standards for interpreting FWD data across regions. By applying these reliable threshold criteria, road agencies can prioritize which sections need immediate rehabilitation, preventive maintenance, or no intervention.