96 Transportation Research Record: Journal of the Transportation Research Board, No. 2248, Transportation Research Board of the National Academies, Washington, D.C., 2011, pp. 96–103. DOI: 10.3141/2248-13 areas were more frequent and severe. Although automotive technol- ogy and vehicle design have evolved significantly, such safety prob- lems on acceleration lanes have not been solved, as demonstrated by more recent crash analyses developed on such areas (9, 10). Although scientific literature proposes several crash prediction mod- els (11–14) that relate crash frequency at interchanges to different explanatory variables such as traffic volumes and geometric design standards, there are no guidelines or research outcomes to provide designers with clear and updated criteria for appropriate geometries of the acceleration lanes based on the actual driver behavior. The main objective of this study is to evaluate driving performance on freeway acceleration lanes using a driving simulator. In this study the effects of different design variables on driver behavior during the merging maneuver were investigated, such as the influence of the length of the acceleration lane and the traffic volume on merging drivers’ speed, trajectory, and acceleration. The indicators have been analyzed and compared among different configurations to evaluate their significant variability and possible effects on the safety of the acceleration lane. Findings are consistent with previous studies, demonstrate the value of simulators to investigate how traffic and geometric features of the road affect driver behavior, and give insights into several future research ideas in order to provide practi- cal applications in traffic engineering, such as the guidelines neces- sary for the design of safer and more efficient acceleration lanes based on effective driving behavior. LITERATURE REVIEW Acceleration Lane Features and Road Safety During the past 60 years, considerable research has been conducted on the impact of various freeway interchange elements on freeway safety. McCartt et al. studied crashes that occurred around several ramps in Northern Virginia (10). Results showed that 48% of crashes occurred while exiting from the freeway, 36% occurred while enter- ing the freeway, and 16% were recorded in the central area of the interchange. Many researchers have proposed statistical models to validate a relationship between the crash rate and the design vari- ables of interchanges. Cirillo et al. proposed multiple regression models to predict crash frequencies for entire interchange areas (11). Traffic volume was found to be a key variable in predicting the interchange crash rates. In another study, it has been found that shorter acceleration lanes exhibited higher crash rates, regardless of the characteristics of merging or diverging traffic flow (12). Bauer and Harwood proposed some statistical models and found that most of the variability can be explained by the flows on the ramp and mainline freeway, the area type, and the ramp type and configura- tion (13). Bared et al. developed a statistical model to estimate the Driver Behavior on Acceleration Lanes Driving Simulator Study Alessandro Calvi and Maria Rosaria De Blasiis Acceleration lanes provide access to freeways with the aim of improv- ing traffic flow conditions and safety. Therefore, provision of an appro- priate entrance ramp and acceleration lane geometry that allows the entering vehicle to accelerate to a speed closer to the main flow speed is crucial for comfortable and safe merging maneuvers. This paper pre- sents an analysis of driver behavior on different acceleration lanes in various traffic conditions performed on an interactive driving simula- tor. Three different traffic volumes combined with two lengths of accel- eration lane were simulated and their effects on driving performances have been studied. It has been observed that driver behavior during merging maneuvers is significantly influenced by traffic volume on the main lane and that the acceleration lane length does not show any sig- nificant effect on drivers’ speed, trajectories, and accelerations. In par- ticular, as the traffic volume increases, so does the merging length of the driver; the acceleration oscillations and the number of gaps rejected also increase. The general results show that the advanced techniques of driving simulation can disclose the relationships between road design parame- ters and behavioral aspects important to create safer road infrastruc- ture. Further simulation studies are planned to confirm the findings and to strengthen and generalize the results. Acceleration lanes provide a transition from low-speed ramps or turning roadways to high-speed lanes. Acceleration lanes are built to improve traffic flow conditions, reduce interruptions, increase capacity, and improve safety. Specifically, comfortable and safe merging maneuvers can be performed with an appropriate entrance ramp and an acceleration lane that allow the entering vehicle to accelerate to a speed closer to the main lane flow speed. The merging process from an acceleration lane to the through lane of the freeway is an important aspect of safety and operating condi- tions at interchanges. Many studies have analyzed traffic flow oper- ating characteristics to establish the safest and most efficient design criteria needed to improve the efficiency of the freeway acceleration lane and its associated entrance ramp. Nevertheless, many of the design criteria for interchanges are based on 50- to 60-year-old stud- ies and are considered still valid for the recent edition of technical regulations in many countries (1–5). The safety effects of various freeway interchange elements have been studied for at least 40 years (6, 7 ). A number of research stud- ies have identified interchanges as the most likely area for freeway collisions to occur. An old study (8) showed that collisions in these Department of Sciences of Civil Engineering, University Roma Tre, Via Vito Volterra, 62-00146 Rome, Italy. Corresponding author: A. Calvi, calvi@uniroma3.it.