The important aspects of the original network optimization system (NOS) used in the Arizona Department of Transportation (ADOT) are reviewed. The NOS has been an important instrument for the Highway Preservation Program since the early 1980s. However, no major updates have been conducted to the original NOS since its initial implementation. It was determined that there is a need to reevaluate the system since there are more than 10 years of pavement performance data available now and technology advancements in microcomputers. Several improvements should be made to the NOS model structure and the original transition probability matrices (TPMs). The factor of crack change was found to be insignificant in predicting the acceleration of pavement deterioration. Therefore, it was removed from the system. The effective rehabilitation actions were determined to be 6 instead of the original 17. In addition, new prediction models were established for all the road categories on the basis of the 13-year pavement performance data base in Arizona. The TPMs were modified with accessibility rules to improve the prediction of pavement performance. Pavement probabilistic behavior curves have been established and analyzed on the basis of Chapman-Kolmogorov equations. The new NOS structure improves the effectiveness and efficiency of the optimization. The enhanced NOS is implemented on a high-end microcomputer in the 32-bit operating environment. ADOT uses the new NOS to conduct financial analysis for more than 7,000 mi of highways with annual rehabilitation funding approaching $100 million. (A)
Abstract