To deal with the frequent occurrence of queue spillover along urban arterial, a signal optimization method is proposed in this study to balance the queue spillover risk by resorting to an arterial partitioning approach. Based on the estimated queue length, the spillover risk of each intersection is divided into three levels, i.e., low, medium and high. The intersection with higher spillover risk will be included into one spillover cluster using the arterial partitioning approach. Then, specific signal control strategies are developed for different types of intersections (i.e., input, connecting and output) in one spillback cluster, in order to realize balanced queue spillover risk for intersections along the arterial. Last, one typical arterial was modeled using microscopic simulation VISSIM to validate the signal optimization method proposed in this study. The experimental results help shed light on both theoretical and technical aspects in active risk prevention and alleviation of queue spillover on urban arterial.