Roadway design usually involves choices regarding grade selection and earthwork (transportation) that can be solved using linear programming. Previous work considered the road profile as series of interconnected linear segments. In these models, constraints are included in the linear programming formulation to insure continuity of the road, which cause sharp connectivity points at the intersection of the linear segments. This sharp connectivity needs to be smoothed out after l;he linear programming solution is found and the earth in the smoothed portion of the roadway has to be moved to the landfill. In previous research, the smoothing issue is dealt with after an optimal solution is found. This increases the work required by the design engineer and consequently increases the construction cost; furthermore, the optimal solution is violated by this smoothing operation. In this paper, the issue of sharp connectivity points is resolved by representing the road profile by a quadratic function. The continuit Roadway design usually involves choices regarding grade selection and earthwork (transportation) that can be solved using linear programming. Previous work considered the road profile as series interconnected linear segments. In these models, constraints are included in the linear programming formulation to insure continuity of the road, which causes sharp connectivity points at the intersection of the linear segments. This sharp connectivity needs to be smoothed out after l; he linear programming solution is found and the earth in the smoothed portion of the roadway has to be moved to the landfill In previous research, the smoothing issue is dealt with after an optimal solution is found. This increases the work required by the design engineer and accounts increases that the construction cost; furthermore, the optimal solution is violated by this smoothing operation. In this paper, the issue of sharp connectivity points is resolved by representing the road profile by a quadratic funct ion. The continuit