Optimizing the maintenance strategies for a network of green infrastructure: An agent-based model for stormwater detention basins. (15th March 2023)

• Record Type: Journal Article
• Authors: Assaad, Rayan H.; Assaf, Ghiwa; Boufadel, Michel
• Is Part Of: Journal of environmental management. Volume 330(2023)
• Publication Date: 2023-03-15
• Subjects: Detention basins -- Agent-based modeling -- Maintenance strategies -- Optimization -- Green infrastructure -- Stormwater management; Environmental policy -- Periodicals; Environmental management -- Periodicals; Environment -- Periodicals; Ecology -- Periodicals; 363.705
• DOI: 10.1016/j.jenvman.2022.117179
• Languages: English
• ISSNs: 0301-4797

Abstract

Abstract: Various stormwater best management practices and green infrastructures (GIs) are recommended to address flooding, stormwater runoff, water quality, and sustainability. While detention basins are considered one of the main GI strategies, their benefits cannot be fully realized without properly maintaining them and making sure that they stay operational. Therefore, this paper used agent-based modeling (ABM) to devise an optimal maintenance program for detention basins to ensure that they function properly and continue to perform their water quality and flood control functions. More specifically, the following 2 agent types were incorporated in the model: 1) the detention basins were considered as static agents, and 2) the service teams responsible for the operation (maintenance, repair, and replacement) of the detention basins were considered as active agents. The developed ABM was applied for the entire network of stormwater detention basins in Newark, NJ. Sensitivity analysis was conducted to identify the most critical variables affecting the total cost of operating the network of detention basins as well as the functioning percentage of detention basins. In addition, optimization was implemented to determine the best maintenance program or policy that minimizes the total cost of operations, while also making sure that a desired functionality level or threshold is achieved for the entire network of detention basins. Finally, the ABM was statistically validated using a total of 10, 000 Monte Carlo runs and 99% confidence intervals. The optimization results showed that, in order to minimize the total cost of maintaining the entire network of detention basins and ensure that at least 80% of the basins are in a functioning state at the end of the planning horizon, the decision-maker should implement the following maintenance program or strategy: have 2 service teams for the operations of the detention basins, follow a replacement policy, and replace detention basins after 3 maintenance periods. Also, the identified optimal maintenance program or strategy would result with an average total annual cost of around $4, 085, 000, where the average annual repair cost is around $2, 572, 200, the average annual maintenance cost is around $19, 700, the average annual replacement cost is around $763, 100, and the average annual service team cost is around $730, 000. The proposed ABM for detention basins can be extended to other GIs as well as to different geographical areas. The usage of ABM has the advantage to reduce the subjectivity in developing plans for managing GIs. Highlights: Agent-based modeling could be implemented for deciding optimal maintenance strategies for a network of detention basins. The optimal maintenance strategy was based on minimizing total cost and ensuring at least 80% of the basins are functional. The identified best maintenance program leads to an average total annual cost of around $4, 085, 000. The determined optimal maintenance strategy was statistically validated using Monte Carlo runs and 99% confidence intervals. Using agent-based modeling has the advantage of reducing the subjectivity in developing plans for managing detention basins.