Development of a linear programming model for optimum budget allocation to the accident-prone points of rural roads

Geometry and traffic corrections of high-ways and rural roads, especially in the accident-prone points, are some of the most effective measures of accidents reduction. Because of the high cost of such measures and limitations in construction budgets, the study of how these corrections can be made for the purpose of maximizing accidents reduction is necessary. In this paper, objective function coefficients of a mixed linear programming model, of the back pack type, with two variables of continuous and discontinuous decision-making, were developed for the necessary corrections. The results obtained by solving this problem showed that from the measures proposed, two measures of enhancing the road width and embedding lighting equipment were not economical and some others such as path smoothing, safety road lining and cat-eyed performance had a higher priority. The results of sensitivity analysis, in the form of a diagram representing objective function variations relative to budget changes, also showed that increasing the budget, up to a certain limit and with a slow slope, could lead to the reduction of accidents, but following that, the slope was to a great extent slowed. This finding can help policy-makers and managers in allocating budgets for accident-prone points and saving costs.