TY - JOUR TI - Probabilistic life cycle cost optimization of bridges DO - https://doi.org/doi:10.7282/T38S4MXK PY - 2013 AB - The main goal is to develop time-dependent bridge deterioration curves and to perform a Probabilistic-Bridge-Life-Cycle-Cost-Optimization (BLCCO-p) using them. Currently, a commonly accepted methodology considering the bridge as a system with all of its components doesn’t exist according to a report by the National Cooperative Highway Research Program published in 2003. Therefore, research is needed to develop a methodology for the BLCCO taking a systems approach and ensure that the most appropriate course of action is taken for bridge improvements. Bridge deterioration curves are developed here based on nonlinear optimization analysis of the National Bridge Inventory (NBI) database using Markov-Chain process. Most of the existing models consider only the age of the bridge and the Average Daily Traffic (ADT) and disregard other important factors such as climatic regions, bridge length and material type in formulating the objective functions. The deterioration curves developed in this dissertation categorize each bridge according to their climatic region, length, ADT and material type. Bridge deterioration curves are formed for superstructure, deck and substructure of the bridges. This deterioration is simulated by a Markov-Chain process. Unexpected event (seismic, scour, etc.) occurrences are also considered. In the BLCCO model, which is essentially a set of economic principles and computational procedures to obtain the most economical strategy for ensuring that a bridge will provide the services for which it was intended, bridge deterioration curves are used as the decision tool for the repair or the replacement of the bridge. In addition, detour traffic analysis, cost and effectiveness of rehabilitation or replacement activities, user, accident, agency costs and discounting models are included. The BLCCO is formulated as a mixed-integer nonlinear optimization model. The BLCCO model uses genetic algorithm to reach the optimal total cost and Monte-Carlo simulation as a risk analysis technique to do the optimization probabilistically (BLCCO-p). Additional major outcomes of this dissertation are (1) updating of deterioration curves using Bayesian methodology, (2) a combined methodology for reliability index and bridge condition index to be used in deterioration models, (3) performing a parametric study for BLCCO using traffic volume growth, discount rate, user cost weight, probability of unexpected events. KW - Civil and Environmental Engineering KW - Bridges--Cost KW - Life cycle costing LA - eng ER -