‹ Back

 

Resource & Energy Inputs and Economic & Environmental Impact Objectives Optimization

 

Genetic Algorithm based on NSGA2 method is used to find a possible solution and to solve a Multi-Objective Optimization problem involving two inputs (Resource & Energy) and two objective functions (Economic & Environmental Impact). The aim is to minimize both of the objectives simultaneously to find the optimal solutions.

The fictional company already has defined parts of its operational objectives. The three parameters population, affluence, and technology can be changed to meet the stated objective functions of economic and environmental impacts. There will not be only one solution to this problem. Increase in affluence, for example, would lead to a demand for better technologies, which adds up extra cost to maintain the same level of environmental impact. Therefore, the financial requirements will also increase.

 

         

 

Inputs (Resource & Energy) and Objectives (Economic & Environmental Impact)

 

To find the best solutions for the problem: The plot of Economic Objective vs Environmental Impact Objective gives possible output solutions for different production outputs. At every point in the plot, there is one product with an individual parametrization of the mentioned parameters (population, affluence, and technology). Every point in the plot represents different products and has different design characteristics.

The goal is to optimize the systems for the defined target parameters (Economic & Environmental Impact). We want to minimize the Environmental Impact and maximize the Economic outcome of the product or process. Therefore, there are solutions which are better than others so called Pareto Optimal solutions at the boundary of the points (Pareto Frontier). For the Pareto Optimal solutions, there are no better solutions, as for an improvement in one criterion leads to a reduction in another one. For solutions within the area under the Pareto Optimal solution, the target parameters can be changed without any tradeoffs in either of the parameters.

The goal of NSGA2 is to find the Pareto Optimal solutions out of a big possible design space for the defined target indicators. The Genetic Algorithm NSGA2 is based on the principle of Darwin’s Survival of the Fittest and is used to find the fittest solution for the defined problem. The fitness value is equal to the performance of the outputs of the systems (in the example, the product or process is the system and for the output, we get fitness from the Economic & Environmental Impact produced). The genes which cause the survival of the fittest are equal to the input parameters (population, affluence, and technology).

The NSGA2 simulation is done using Pymoo multi-objective optimization package in Python.

Two Objectives (Economic & Environmental Impact) Optimization