Integration of stochastic deterioration models with multicriteria decision theory for optimizing maintenance of bridge decks
Author(s) -
George Morcous,
Z. Lounis
Publication year - 2006
Publication title -
canadian journal of civil engineering
Language(s) - French
Resource type - Journals
SCImago Journal Rank - 0.323
H-Index - 62
eISSN - 1208-6029
pISSN - 0315-1468
DOI - 10.1139/l06-011
Subject(s) - bridge maintenance , computer science , bridge (graph theory) , maximization , optimal maintenance , markov chain , probabilistic logic , mathematical optimization , minification , deck , reliability engineering , operations research , engineering , structural engineering , mathematics , machine learning , medicine , artificial intelligence , programming language
This paper presents a new approach to optimizing the maintenance of concrete bridge decks. This approach combines a stochastic deterioration model and a multiobjective optimization model. The stochastic deterioration model is based on the first-order Markov chain, which predicts the probabilistic time variation of bridge deck conditions. The multiobjective optimization model takes into account two important and conflicting criteria: the minimization of maintenance costs and the maximization of the network condition. This approach achieves the best compromise between these competing criteria while considering the uncertainty in bridge deck deterioration. The feasibility and capability of the proposed approach are demonstrated with field data for a sample network of bridge decks obtained from the Minist\ue8re des Transports du Qu\ue9bec database. This example illustrates the effectiveness of the proposed approach in determining the optimal set of maintenance alternatives for reinforced concrete bridge decks when two or more relevant optimization criteria are taken into consideration.Cet article pr\ue9sente une nouvelle approche d'optimisation de la maintenance des tabliers de ponts en b\ue9ton qui associe un mod\ue8le stochastique de d\ue9t\ue9rioration et un mod\ue8le d'optimisation multi-objectif. Le mod\ue8le stochastique de d\ue9t\ue9rioration est bas\ue9 sur une cha\ueene de Markov de premier ordre 1 qui pr\ue9dit la variation probabiliste dans le temps de l'\ue9tat des tabliers de pont. Le mod\ue8le d'optimisation multivariable tient compte de deux crit\ue8res importants contradictoires : la minimisation des co\ufbts de maintenance et la maximisation de l'\ue9tat du r\ue9seau. La pr\ue9sente approche g\ue9n\ue8re une solution qui atteint le meilleur compromis entre ces crit\ue8res contradictoires, tout en consid\ue9rant l'incertitude de la d\ue9t\ue9rioration des tabliers de pont. La faisabilit\ue9 et la capacit\ue9 de l'approche propos\ue9e sont d\ue9montr\ue9es dans un r\ue9seau de tabliers de pont obtenu de la base de donn\ue9es du minist\ue8re des Transports du Qu\ue9bec. Cet exemple illustre l'efficacit\ue9 de l'approche propos\ue9e \ue0 d\ue9terminer l'ensemble optimal des options de maintenance des tabliers de pont en b\ue9ton arm\ue9 tout en consid\ue9rant au moins deux crit\ue8res d'optimisation pertinents.Peer reviewed: YesNRC publication: Ye
Accelerating Research
Robert Robinson Avenue,
Oxford Science Park, Oxford
OX4 4GP, United Kingdom
Address
John Eccles HouseRobert Robinson Avenue,
Oxford Science Park, Oxford
OX4 4GP, United Kingdom