RISK-BASED MAINTENANCE OPTIMIZATION OF AGING. HIGHWAY BRIDGE DECKS

Zoubir Lounis

Institute for Research in Construction,
National Research Council, Ottawa, Ontario, Canada;
E-mail: Zoubir. Lounis@nrc. gc. ca

Abstract

This paper presents a practical approach for maintenance optimization of a network of aging highway bridge decks that integrates a stochastic deterioration model based on Bogdanoff’s cumulative damage theory with an effective multi-objective optimization approach. The multi-objective maintenance optimization takes into account all relevant objectives, such as improving bridge deck condition, minimizing maintenance costs, and minimizing traffic disruption and associated user costs. The consideration of these three objectives enables to take full advantage of the available bridge inspection data and implicitly lead towards the minimization of the risk of failure due to bridge deck deterioration and maintenance activities. A multi-objective optimality index is proposed as an optimality criterion for priority ranking of the deficient bridge decks for maintenance. The obtained optimal maintenance project prioritization strategy achieves a satisfactory trade-off or compromise between the selected relevant and competing optimization objectives. The proposed approach is illustrated on a small network of ten bridge deck projects that are optimized for maintenance.

Introduction

The deterioration of reinforced concrete bridge decks due to reinforcement corrosion is recognized as the main cause of failure of bridge decks and it is estimated that one-third to one-half of the projected bridge rehabilitation costs in North America are related to bridge deck deterioration (Weyers 1998; Lounis and Mirza 2001). The corrosion of the reinforcing steel is caused by the chlorides (from deicing salts), which penetrate the concrete cover and destroy the protective passive film on the steel reinforcement. As the corrosion products accumulate, they generate high tensile stresses, which eventually lead to irreversible damage such as concrete cracking, delamination and spalling. Furthermore, corrosion leads to reduction of concrete and reinforcement cross sectional areas, loss of bond between steel and concrete, reduction in strength, and ductility. The effects of corrosion are compounded by other deterioration factors, such as initial damage (e. g. due to shrinkage cracking), increased traffic loads, freeze-thaw cycles, poor workmanship, and inadequate maintenance.

The extensive deterioration of highway bridge decks in North America and the limited funds allocated for their maintenance present considerable technological and economic challenges for bridge owners and managers, namely: (i) which bridge decks to maintain; (ii) when to maintain them; and (iii) how to maintain them (i. e. identify the most effective maintenance strategy, which could be a patch repair, overlay, partial depth replacement and overlay, cathodic protection, total replacement, etc.). The importance of a pro-active maintenance policy is confirmed by the study carried out by Dunker and Rabbat (1990) on the performance of highway bridges included in the U. S. National Bridge Inventory (NBI), which showed that bridge deterioration varied considerably from state to state with the highest deterioration level being observed in the central and southeastern states, while the lowest deterioration was observed in the southwestern states. This considerable difference in structural deficiency between the different states was attributed to differences in design, construction, inspection, funding and most importantly maintenance policies.

Different approaches to maintenance optimization have been implemented in the different bridge management systems ranging from simplified economic models to advanced Markovian decision processes. In the literature on bridge management, the main optimization objective used for maintenance optimization is the minimization of the present value life cycle cost, which represents all the costs incurred throughout the life cycle of a bridge structure, including, the costs of design,

723

M. Pandey et al. (eds), Advances in Engineering Structures, Mechanics & Construction, 723-734.

© 2006 Springer. Printed in the Netherlands.

construction, maintenance, repair, rehabilitation, replacement, demolition, and in some instances users’ costs, and possibly costs of failure. Most of these systems are based on a single-objective optimization, namely, the minimization of the maintenance costs.

Given the importance and high consequences of failure of highway bridge structures, a risk-based maintenance management methodology can be more effective and objective as it enables the optimization of different types of structures and systems from different bridges within a network by considering not only their probability of failure but also their consequences of failure. The implementation of a quantitative risk-based bridge maintenance management, however, is very complex task due to the difficulties of assessing quantitatively the probability and the consequences of failure, especially for a large network of bridge structures.

The failure modes of bridge decks may include the loss of serviceability (e. g. excessive cracking, concrete delamination and spalling, and deformation), loss of functionality (e. g. poor traffic conditions, inadequate deck geometry/approach roadway alignment, limited clearance, etc.), and possibly the partial or total collapse. As mentioned earlier, the loss of serviceability and loss of functionality, although not catastrophic or life-threatening, are by far the most frequent failure modes for bridge decks and involve significant cumulative costs. The consequences of failure of highway bridges are multiple and may include loss of life, injury, excessive maintenance costs, user costs, traffic disruption, environmental impacts, etc. It is clear that some of these consequnces are incommensurable and cannot be evaluated in monetary terms.

To overcome the above difficulties, a multi-objective maintenance optimization approach that enables to consider all relevant criteria of a risk-based approach, namely the maximization of condition rating, minimization of maintenance costs, and minimization of user costs is presented in this paper. The proposed approach can address the requirements of many decision-makers (bridge owners/ managers/ engineers) seeking to satisfy implicitly and/or explicitly several objectives at the same time in planning the maintenance of a large network of aging bridge decks. The proposed approach enables a better evaluation of the effectiveness of maintenance strategies in terms of several criteria and determines the optimal solution that achieves the best trade-off between all criteria (including conflicting ones, such as condition and cost).