A new method for timing traffic signals in heavily traveled intersections is introduced. The effectiveness of the new method was evaluated in a simulation study. In 50 simulations, the mean improvement in traffic flow in a particular intersection was 654.1 vehicles per hour, with a standard deviation of 311.7 vehicles per hour. Approximately what sample size is needed so that a 98% confidence interval will specify the mean to within ± 50 vehicles per hour?