A macro-evolutionary multi-objective immune algorithm with application to optimal allocation of water resources in Dongjiang River basins, South China

Macro-evolution is a new kind of high-level species evolution inspired by the dynamics of species extinction and diversification at large time scales. Immune algorithms are a set of computational systems inspired by the defense process of the biological immune system. By taking advantage of the macro-evolutionary algorithm and immune learning of artificial immune systems, this article proposes a macro-evolutionary multi-objective immune algorithm (MEMOIA) for optimizing multi-objective allocation of water resources in river basins. A benchmark test problem, namely the Viennet problem, is utilized to evaluate the performance of the proposed new algorithm. The study indicates that the proposed algorithm yields a much better spread of solutions and converges closer to the true Pareto frontier compared with The Non-dominated Sorting Genetic Algorithm and Improving the Strength Pareto Evolutionary Algorithm. MEMOIA is applied to a water allocation problem in the Dongjiang River basin in southern China, with three objectives named economic interests (OF ₁), water shortages (OF ₂) and the amount of organic pollutants in water (OF ₃). The results demonstrate the capabilities of MEMOIA as well as its suitability as a viable alternative for enhanced water allocation and management in a river basin.