METHODOLOGY FOR USING A GENETIC ALGORITHM TO SOLVE URBAN PLANNING PROBLEMS

Abstract

With the development of computer technology, algorithms for solving problems based on artificial intelligence are gaining popularity. The genetic algorithm is one of them. The article describes using of this algorithm in the context of urban planning problems.  The genetic algorithm is primarily created to search for extremes of complex functions, where the variable can be described as a vector. A significant number of urban planning problems can be represented as follows. The article describes the methodology for solving urban planning problems using this algorithm. The main advantage of the proposed algorithm is its relative ease of implementation and a wide range of issues that can be solved with its help. The main disadvantage of this algorithm is that it is obtained the result, with a certain probability, can reflect the local minimum of the function, as well as the need to change the parameters of the genetic algorithm individually for each individual task.  The article gives an example of using the proposed methodology to solve the problem of optimizing the location of work place. It is also proposed to apply a genetic algorithm to optimize the traffic light cycle, automate the process of building planning and develop traffic management schemes.  The implementation of the proposed methodology can significantly increase the quality of urban planning projects, as well as reduce the time spent on their implementation.