Skip to main content
ARS Home » Plains Area » Fort Collins, Colorado » Center for Agricultural Resources Research » Water Management and Systems Research » Research » Publications at this Location » Publication #320922

Title: Improved ant colony optimization for optimal crop and irrigation water allocation by incorporating domain knowledge

Author
item NGUYEN, DUC CONG - University Of Adelaide
item DANDY, GRAEME - University Of Adelaide
item MAIER, HOLGER - University Of Adelaide
item Ascough Ii, James

Submitted to: Journal of Water Resources Planning and Management
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 1/25/2016
Publication Date: 6/2/2016
Citation: Nguyen, D.H., Dandy, G.C., Maier, H.R., Ascough II, J.C. 2016. Improved ant colony optimization for optimal crop and irrigation water allocation by incorporating domain knowledge. Journal of Water Resources Planning and Management. doi.org/10.1061/(ASCE)WR.1943-5452.0000662.

Interpretive Summary: An improved ant colony optimization (ACO) formulation for the allocation of crops and water to different irrigation areas is developed. The formulation enables dynamic adjustment of decision variable options and makes use of visibility factors to bias the search towards selecting crops that maximize net returns and selecting water allocations that result in the largest net return for the selected crop, given a fixed amount of water. The performance of this formulation is compared with that of other ACO algorithm variants (without and with domain knowledge) for two case studies, including one from the literature and one based on an irrigation district in the South Australian reaches of the River Murray introduced in this paper for different water availability scenarios. The results demonstrate that the improved formulation is able to identify better solutions, especially for highly constrained problems, and is able to do so at significantly reduced computational effort.

Technical Abstract: An improved ant colony optimization (ACO) formulation for the allocation of crops and water to different irrigation areas is developed. The formulation enables dynamic adjustment of decision variable options and makes use of visibility factors (VFs, the domain knowledge that can be used to identify locally optimal solutions) to bias the search towards selecting crops that maximize net returns and selecting water allocations that result in the largest net return for the selected crop, given a fixed volume of water. The performance of this formulation is compared with that of other ACO algorithm variants (without and with domain knowledge) for two case studies, including one from the literature and one based on an irrigation district in the South Australian reaches of the River Murray introduced in this paper for different water availability scenarios. The results for both case studies indicate that the use of VFs increased the ability to identify best-found solutions, especially under lower water allocation regimes which correspond to more constrained search spaces. The best results were obtained when dynamic decision variable option adjustment is combined with the use of VFs as this combination was able to identify the best-found solution for all scenarios in both case studies. In addition, this algorithm variant also resulted in a marked increase in computational efficiency, reducing computational effort by around 90% for the less constrained water availability scenarios for both case studies. For the more constrained scenarios, the best-found solutions were identified within approximately 25% to 71% of the available computational budget.