Civil infrastructure systems are the foundation of economic growth and prosperity in all nations. In recent years, infrastructure rehabilitation has been a focus of attention in North America and around the world. A large percentage of existing infrastructure assets is deteriorating due to harsh environmental conditions, insufficient capacity, and age. Ideally, an assets management system would include functions such as condition assessment, deterioration modeling, repair modeling, life-cycle cost analysis, and asset prioritization for repair along a planning horizon. While many asset management systems have been introduced in the literature, few or no studies have reported on the performance of either optimization or heuristic tools on large-scale networks of assets.
This research presents an extensive comparison between heuristic and genetic-algorithm optimization methods for handling large-scale rehabilitation programs. Heuristic and optimization fund-allocation approaches have been developed for three case studies obtained from the literature related to buildings, pavements, and bridges with different life cycle cost analysis (LCCA) formulations. Large-scale networks were constructed for comparing the efficiency of heuristic and optimization approaches on large-scale rehabilitation programs. Based on extensive experiments with various case studies on different network sizes, the heuristic technique proved its practicality for handling various network sizes while maintaining the same efficiency and performance levels. The performance of the genetic algorithm optimization approach decreased with network size and model complexity. The optimization technique can provide a high performance level, given enough processing time.
Identifer | oai:union.ndltd.org:LACETR/oai:collectionscanada.gc.ca:OWTU.10012/7043 |
Date | January 2012 |
Creators | Binhomaid, Omar |
Source Sets | Library and Archives Canada ETDs Repository / Centre d'archives des thèses électroniques de Bibliothèque et Archives Canada |
Language | English |
Detected Language | English |
Type | Thesis or Dissertation |
Page generated in 0.0026 seconds