Cost-Effectiveness Evaluation of Different Safety Interventions Strategies on Two-Lane Rural Highways

Nowadays, an essential part of any Safety Management System is the network screening, that is the identification of sites where the greatest cost-effectiveness of safety measures is expected. In rehabilitation and reconstruction projects, geometric improvements of existing roads represent higher costs with respect to other safety improvements related to pavement, markings, sign, safety barrier or other low cost interventions. For the safety evaluation of a road section, in a previous study carried out by the authors, a Safety Index (SI) was formulated by combining three components of risk: the exposure of road users to road hazards, the probability of a vehicle being involved in an accident and the resulting consequences should an accident occur. This systematic and replicable procedure showed the advantage of applicability on two lane local rural roads where crash data are either not available or statistically not significant. Due to the significant correlation found between SI values and EB safety estimates, in this paper a methodological approach using the SI to evaluate the effectiveness of alternative investment in delivering specified levels of safety under constant budget is presented. Specifically, the cost-effectiveness ratio of alternative investments related to different intervention strategies was evaluated in terms of SI variation (ΔSI) and cost during a defined cycle life. For an allocated budget, the ΔSIs obtained using different intervention strategies can be related to an expected accident reduction and therefore they can be effectively used as ranking criteria into a Safety Management System. In the case of unlimited available budget, the research shows that Geometry strategy (G) is still the more effective one in a level terrain scenario, while in different terrain conditions the Roadside strategy (R) is to be preferred, emphasizing the highly cost-effectiveness ratio at any budget and/or terrain characteristics.