The best countermeasure will not work if it is used for the wrong problem. We have to gather the
right information to understand the problem to get the best results. Information used in traffic
safety studies is usually related to road conditions or traffic characteristics.
Condition diagrams are drawings, more or less to scale, that show the locations of curves, traffic
control devices, fixed objects, etc. An example of a condition diagram is shown in Figure 8.
If you want to consult with another person, or refer to a manual, a condition diagram can help
you to remember details about the location you are investigating. It also provides a record for
later use, should another problem crop up nearby.
To prepare a condition diagram, you will need a measuring wheel (or a distance-measuring
computer), a clipboard and paper, and a pencil. A tape measure may also be useful to measure
offsets and lane widths. Start at a location that will be easy to find again, such as a cross culvert,
and set the wheel to zero.
Walk towards the other end, and measure the locations of:
When you are done, make a drawing of the road and mark the locations of items you recorded.
Remember that this diagram may be subpoenaed in a court case, so have a plan to remedy any
defects you find. On the other hand, you could use the diagram to show a missing sign was there
on the date of the condition diagram.
The best way to measure traffic volume is with mechanical counters. If counters are not
available, the peak hour traffic volume can be used to estimate the average daily traffic. This is
easier than standing out on the roadside all day.
Estimating average daily traffic (ADT)
In rural areas, peak hour traffic is usually about 15 percent of the total daily volume. In
urban areas, it is usually about 11 percent. The peak hour often occurs between 4 pm and
6 pm. Divide that period into 15-minute increments, and count the number of cars that
go by during each 15-minute increment. The peak hour is the four consecutive 15-minute
periods with the highest number of cars. It may be 4:45 to 5:45, for example. Take the
total number of cars during that hour and multiply by 7 for rural areas, or 9 for urban
areas. That is usually close to the average daily traffic.
The 85th percentile speed is used when making decisions about everything from curve design to
speed limit determination. This is the speed at or below which 85 percent of all drivers drive at a
particular location under good weather, visibility, and traffic volume conditions. In other words,
15 percent of traffic exceeds this speed. It assumes that most drivers are reasonably prudent and
will not drive too fast for road conditions. The 85th percentile speed is sometimes called the
prevailing speed or the running speed.
The 85th percentile speed can be measured with a police radar or laser speed measurement unit,
but they can be costly. Since they will not be used for enforcement, expensive police radars are
not needed. Non-enforcement radar units can cost less than $300. If radar is not available, and
traffic volumes are moderate, you can follow other cars and note the speed they are driving.
This should be done repeatedly to get a good idea of the range of travel speeds on the road.
To get a statistically valid measurement of the 85th percentile speed, the speed of at least 50
cars should be measured, and 100 measurements are desired. More measurements yield more
accuracy. On low-volume roads or when using the car-following method, 50 observations may
be impractical. You may be forced to make an estimate based on too few observations. Having
some data is better than just guessing. For many traffic applications, the 85th percentile speed is
rounded up to the nearest five mph, so high accuracy is not always needed.
Road safety studies
There are two main types of road safety studies. In a road safety audit, auditors examine a road,
looking for potential trouble spots. Accident analysis uses police accident reports to determine
what is causing accidents. They both have strengths and weaknesses, but they complement each
Road safety audits
Over the past few years, a new tool has emerged to improve roadway safety. A road safety audit
is a formal examination of an existing road, or a future road or traffic project by an independent
team of specialists. The team assesses the safety of a roadway or project, and prepares a report
that identifies potential safety problems.
Road safety audits are proactive. The team looks for problems before someone gets hurt.
Unlike accident studies, road safety audits can be done at any stage in a road’s life. They can be
performed in the planning and design stages of a new project. A work zone plan can be audited
to solve traffic problems. During construction, the road may be audited to fix problems before
the project is completed. Lastly, an audit may be performed on an existing road, to help find and
correct safety deficiencies.
The audit report usually gives some indication of the urgency of defects found. A minor defect
in a location where a crash is unlikely will be given a low priority. A defect that could result in
frequent, severe crashes will be given a high urgency. This helps you focus your resources in a
Road safety audits do have a few disadvantages. The team may not spot accident patterns that
would show up on an accident diagram. Choose team members carefully to avoid biases and
make sure the team has the experience needed. An expressway designer may not be a good
auditor on a gravel road.
Accident analysis is a study of police accident reports, looking for common factors in accidents.
It has some real advantages. If a lot of similar accidents have been occurring, the answer may be
obvious. For example, if more than 40 percent of the accidents happen on wet pavement, look for
slippery pavement or bad drainage.
On the other hand, accident analysis has some serious drawbacks. It cannot be done until several
years after a project has been completed. Crashes have already happened, and people may have
already been hurt. Also, it is only as good as the information in the police officers’ reports.
More importantly for local agencies, it does not work as well on low-volume roads. Instead
of three years of accident data, ten years or more may be needed before a pattern emerges.
Currently, the local accident surveillance system does not precisely locate crashes off the state
system that occur between intersections, but changes are planned to improve accuracy and
usability of the system.
Other traffic studies
Depending on what information is needed, other types of traffic studies can be performed.
Some of these include delay studies, intersection turning counts, or traffic signal studies. The
Manual of Transportation Engineering Studies from the Institute of Transportation Engineers has
information on many kinds of traffic investigations.