...In fact, in most instances, collisions and severe injuries increase
after signals are installed.
The Manual on Uniform Traffic Control Devices suggests that
signals should not be installed under the premise of reducing collisions,
unless five or more occur per year. This was predicated on a comprehensive
California study which determined that of all signals installed over a five
year period, the only ones that enjoyed a reduction in collisions were those
that had at least five occurring annually beforehand. Another study, the
Schoene Study, indicated that an increase in collisions should be expected
at intersections having fewer than ten right-angle collisions per year prior
to signalization.
The purpose of signals is to improve the overall flow of traffic. Signals
can be helpful in reducing right-angle collisions under certain
circumstances, but almost always increase other type of collisions such as
rear-end and left turn collisions. Unfortunately, signals can also lull
pedestrians into a false sense of security.
...against guidelines established after many years of study throughout
the country. If conditions do not meet established guidelines, signalization
will only cause more collisions, congestion and delay. When these conditions
are met, an engineer then needs to consider geometric conditions and spacing
to ascertain whether or not a signal could be helpful.
Signals installed at collector and local street intersections may be
necessary to improve access to major streets, but they can result in more
neighborhood cut-through traffic.