最近因为课程设计的需要提前读了Daganzo教授的经典教材Fundamentals of Transportation and Traffic Operations. 也许是因为我看多了数理化的交通文献, 所以看到Daganzo教授从实际出发, 用简单的数学方法解决复杂的工程问题, 就感觉特别神清气爽. 工程技术特别像交通这样的, 我们或许更应该以一种设计的姿态去解决问题, 不要妄想着数学立模型能解释所有问题. 我觉得这样做才能做出点art的味道, 才能出神来之笔. 有时太过science堆砌了很多math, 事情就搞僵了, 大有存天理灭人欲之势.

摘录教材中一段我喜欢的文字:

Although we have just seen that theories exist to evaluate the performance of any given strategy, numerical methods to sort through all the possibilities efficiently have not been developed. Nonetheless, it should be possible to come up with good signal plans from a physical understanding of what is happening and what one would like to see happen. For example, to eliminate long queues in a section of an arterial with closely spaced intersections, which reduce flow and may cause gridlock, we could restrict flow through an upstream signal that can store queues so as to keep things going smoothly through the section with closely spaced intersections. With the increased flow through this critical region the system delay will go down. This type of ‘common sense’ decision making (that should be obvious to anyone with some field experience) should not be discarded just because the optimization problem is difficult mathematically. Good engineering is in fact our best hope to generate sensible plans in complicated cases.