Computational & Technology Resources
an online resource for computational,
engineering & technology publications
Computational Science, Engineering & Technology Series
TRENDS IN CIVIL AND STRUCTURAL ENGINEERING COMPUTING
Edited by: B.H.V. Topping, L.F. Costa Neves, R.C. Barros
Multi-Criteria Optimization of Life-Cycle Maintenance Programs using Advanced Modeling and Computational Tools
D.M. Frangopol and N.M. Okasha
Department of Civil and Environmental Engineering, ATLSS Center, Lehigh University, Bethlehem, Pennsylvania, United States of America
D.M. Frangopol, N.M. Okasha, "Multi-Criteria Optimization of Life-Cycle Maintenance Programs using Advanced Modeling and Computational Tools", in B.H.V. Topping, L.F. Costa Neves, R.C. Barros, (Editors), "Trends in Civil and Structural Engineering Computing", Saxe-Coburg Publications, Stirlingshire, UK, Chapter 1, pp 1-26, 2009. doi:10.4203/csets.22.1
Keywords: computational tools, life-cycle, performance measures, lifetime functions, redundancy, availability, maintenance, multi-criteria optimization, genetic algorithms.
The use of advanced modeling and computational tools is essential in the proper management process of civil infrastructure. For a long time, life-cycle optimization has been recognized as an implicit and essential objective of engineering design, construction and maintenance actions. For the establishment of recommended practical criteria and methods for the achievement of that objective, it is necessary to adopt a formal decision framework, based on the identification and evaluation of an adequate set of quantitative indicators to describe the performance of a given system under uncertainty .
The objective of this keynote paper is twofold. First, the paper aims to shed light on some of the most common performance indicators obtained by advanced modeling and computational tools, and discuss their efficiencies and cost issues. Second, the paper presents two approaches for finding optimum maintenance strategies for deteriorating civil infrastructure systems through multi-criteria optimization and using genetic algorithms with applications. These approaches use different problem formulations and types of performance indicators.
A comparison conducted in this paper between the computational costs of the condition index , the instantaneous probability of failure , the lifetime survivor function  and the cumulative probability of failure  revealed that the lifetime survivor function is the least computationally expensive modeling tool whereas the cumulative probability of failure is computationally the most expensive one. From the study, a trade-off between accuracy and computational cost was observed.
Realistic examples of application of the proposed models are described. The results obtained show the importance of preventive maintenance in reducing the life-cycle cost of structures, but also the importance of essential maintenance action in keeping structures safe and serviceable. The use of multi-criteria optimization provides the decision maker with a set of optimal solutions that help in understanding the compromise and interaction between life-cycle performance and cost and from which the best can be selected for each specific case.
purchase the full-text of this chapter (price £20)