RMIT University
Browse

Employing ant colony for the optimal reduction of project risk severity

conference contribution
posted on 2024-10-31, 18:11 authored by N Zabel, Maged Georgy, Moheeb Ibrahim
Efforts undertaken in identifying, analyzing and assessing project risks are only made good use of when proper risk treatment strategies are decided upon and pursued. Based on the criteria established by senior management, the risk management plan goes about defining how each risk is to be handled. There are options to that end, including acceptance, avoidance, transfer and mitigation. Whilst these strategies are known to all in the industry, the decision-making process is far from easy. A research was undertaken to optimize risk treatment in construction projects, where both costs and benefits are balanced out at the project level. The paper particularly introduces Ant Colony Optimization (ACO) as a capable algorithm for the balanced selection of risk treatment strategies; that is to reduce the overall risk severity in a project at the minimum cost possible. ACO resembles the real life behavior of ants in their intelligent and guided search for food. The research is being applied in the pipeline construction sector and made use of professional knowledge and project records from a big construction company in the Middle East. The paper further presents an example project to demonstrate how ACO explores the risk treatment alternatives in a project and chooses the optimal set of strategies in such context.

History

Related Materials

Start page

398

End page

403

Total pages

6

Outlet

Proceedings of the 31st International Symposium on Automation and Robotics in Construction and Mining 2014 (ISARC 2014)

Editors

Quang Ha, Xuesong Shen, Ali Akbarnezhad

Name of conference

ISARC 2014

Publisher

University of Technology

Place published

Australia

Start date

2014-07-09

End date

2014-07-11

Language

English

Copyright

© 2014, The International Association for Automation and Robotics in Construction (IAARC)

Former Identifier

2006048536

Esploro creation date

2020-06-22

Fedora creation date

2015-01-15

Usage metrics

    Scholarly Works

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC