| As critical components of infrastructure management systems, after enforcement of the Government Accounting Standards Board's statement 34 (GASB 34), development of a valuation approach and optimal rehabilitation and replacement polices (R&R policies) for infrastructure assets become overriding concerns to governmental agencies. To comply with GASB 34 requirements, governments must report the value of their infrastructures' assets by using a depreciation method or a modified approach. However, most existing valuation approaches, including these two, do not properly account for the change of assets' conditions and time-value of money. This may lead to inappropriate asset values, especially for infrastructures that have significantly longer service life than other capital assets.;This study proposes an asset valuation model, combined with a Markov chainbased, condition prediction and a regression-based, construction cost index prediction, to capture the change of asset condition and time value of money. The model not only will assist in predicting the future value of the asset but also will play an important role in constructing the optimal R&R policies and the timing of their application.;This study also describes the development of a two-step optimization process for R&R policies that include project and network levels optimizations. In project-level optimization, probabilistic dynamic programming analyzes the life cycle cost of a system. In network-level optimization, fuzzy logic theory optimizes the allocation of limited funds and prioritization polices for the infrastructure system.;Finally, by describing the potential applicability of the prediction model to real option theory, this study assists government agencies in implementing a comprehensive infrastructure management system to further assist in making satisfactory financial decisions. |
