Abstract: Attempts are often made to find an optimal coordinated control plan for a group of traffic signals in urban areas, such that platoons of vehicles can proceed through a continuous series of green lights without stopping, thereby reducing considerable unnecessary delay. One main objective of coordinated control is to provide the maximum bandwidth progression by adjusting the offset and phase sequence pattern for each signalized intersection. Most models and methods currently being used are based on mixed-integer linear programming (MILP), which is a mathematical optimization model originally formulated by Little et al. in 1966. A basic limitation of the MILP is that it adds a large number of complex constraints for traffic control networks. In this paper, a new simple method is introduced for finding the optimization plan of arterial and network control by maximization of variable bandwidth. This method has not any "loop" constraints and a completely new approach compares to the MILP method applied in this field. In addition, the results of two examples this paper provided validity and feasibility of this new approach in real applications of traffic signal coordination control.

Key words: traffic engineering, traffic control, variable bandwidth progression, offset, NEMA phase sequence pattern

摘要： 在城市交通控制中，很多研究尝试为多个信号交叉口设计最优的协调控制方案，从而使得车队在通过多个交叉口时能够获得一系列连续的绿灯信号，无需停车、不间断地运行，这样可减少一些不必要的停车延误时间，而这种停车延误在现实中往往比较可观. 此类协调控制的主要目标之一是获取最大绿波带，通过调整各个交叉口之间的相位差以及各个交叉口本身的相序来实现. 目前在实际中得到应用的有关模型和方法大多是以混合整数规划为基础的一种线性最优化模型，最早由Little等人于1966年提出. 这种模型的基本局限性在于，对于面控系统，需要引入数量众多的复杂约束. 本文提出了一种新型简便的最优化方法，通过最大化可变绿波带实现对干线和网络各信号交叉口的最优协调控制. 从应用范围来看，该方法与混合整数规划方法的不同点在于规避了对实施协调控制的道路网络的封闭性要求，因而是一种全新的方法. 此外，本文还利用2个实例对该方法在交通信号协调控制中的有效性和实用性进行了验证.

关键词: 交通工程, 交通控制, 可变绿波带, 相位差, NEMA 相位系列