Abstract: It is an important task for urban designers to calculate the urban road's elevation.In this paper,the authors first analyze the factors which influence urban road's elevation designing:(a) road longitudinal slope conformed to the demands of traffic and road surface drainage,(b) the least construction cost,and(c) other design requirements.In general,the designers calculate the elevation of next road surface control point based on one known elevation of one control point,the elevation of the rest unknown road control points can be deduced one by one through such method.Once the elevation of one road surface control point changes,the elevation of others is accordingly adjusted,and so designers must spend much time finishing this work, and the resulting design scheme isn't the best one.For example,the cubage of excavation and fill of earth and stone is not the least,and transportation expenses are not the least,either.The authors establish the genetic mathematical model on designing urban road's elevation based on the corresponding constraint conditions and the objective of the least cost,and state that this mathematical model is a complex non-linear one.It is very difficult,even unfeasible to get the solution of it using those traditional searching methods or Operational Research.According to the authors' research,recently there is really no feasible method to handle it.Genetic algorithms(GA) is a robust searching technique,and it is useful and effective for dealing with this kind of mathematical model.This paper is an attempt to make use of simple genetic algorithms(SGA) for designing urban road's elevation,and it shows that GA is perhaps a promising approach in comparison with the traditional manual method.