Search Google Appliance


Inventory

Review your existing bicycle and pedestrian count program and create an inventory.  In addition to transportation staff, reach out to parks departments, business districts and health departments in your area.  All are potential data collectors.   

As data sources are found, categorize them into permanent count stations and short duration counts: 

  • Short duration counts can be from one hour to multiple months and can be collected by hand or machine. 
  • Permanent count stations are automated counters that count continuously 24-hours per day for at least one year at a given location.

Below are the essential questions to ask:

  • Where are they counting?  
  • What are they counting? 
  • What technology do they use?
  • How long have they been counting there?  
  • When, how often, and for how long do they count?
  • Are these intersection counts or counts on a road or path segment?
  • Has count accuracy been evaluated? 

Once you have inventoried the count data, the next step is to assemble it into one useable format. Multiple regions are working on web-based bicycle and pedestrian clearinghouses which can share and accept data.  Examples of these are shown under resources, below. 

Resources

The Minnesota Department of Transportation published a report in 2013, which describes its inventory of the state’s bicycle and pedestrian count programs and offers recommendations.

The Los Angeles metropolitan area has a bicycle count data clearinghouse which recommends manual count formats, allows partner agencies to upload data and makes the data publicly available.

In the Philadelphia metropolitan area, the Delaware Valley Regional Planning Commission (DVRPC) provides online access to its bicycle and pedestrian counts.

Similarly, Arlington, Virginia offers access to their count data through an online site which allows data to be sorted by weather and time.

Many other jurisdictions are offering their bicycle and pedestrian count data in various formats online.  Below is a partial list:

•    Boulder, Colorado
•    Minneapolis, Minnesota
•    New York City, New York
•    Portland, Oregon
•    Seattle, Washington

Unfortunately, many of the sites above to do not show all of the count data available in a given city or region.  For example, the Boulder site only shows short-duration bicycle and pedestrian counts collected as part of the regularly motor vehicle counting program, and does not provide access to data from their many permanent bicycle count stations. 

For this reason, more efforts on creating centralized data clearinghouses for bicycle and pedestrian count data are needed.  To that end, the National Institute of Transportation and Communities (NITC) has assembled a pooled-fund of local, regional, state and federal agencies, which will create an online national data archive for non-motorized traffic count data.  Work on the archive is scheduled to start in March 2014.  For more information, contact Hau Hagedorn.

 


GUIDANCE FROM THE TRAFFIC MONITORING GUIDE 2013, SECTION 4.4.1

The review of existing continuous counts should review and assess the following: 

Overall Program Design

  •  Existing monitoring locations and why they were chosen.
  •  Existing equipment and any noted performance/accuracy limitations.
  •  Who is using existing data, and for what decisions?
  • Is the existing data sufficient? If not, what are the additional needs and their priorities?
  •  If there is no existing data, who would like data and for what decisions?

Traffic Patterns

If existing continuous count data is available, it should be analyzed to determine typical traffic patterns and profiles:

  •  How do counts vary throughout the day?
  • How do counts very by day of week?
  • How do counts very by month or season?
  • How do counts vary for inclement weather and other special events?
  • How does traffic vary by street functional class and the presence of bike or pedestrian facilities?
  • How do traffic patterns and profiles compare at different locations in areas with different land use and  demographic characteristcs?

Note that the count magnitude may not be similar, but the time-of-day, DOW, or month-of-year patterns may be similar in shape or overall profile. These patterns of variation will ultimately be used to create groups of similar locations (called factor groups) that can be used to factor (i.e., annualize) short-duration counts to an annual volume estimate. If continuous non-motorized count data is not available, short-duration counts can be used to estimate the traffic patterns that may be typical. However, because of the higher variability of pedestrian and bicyclist count data, short-duration counts should be used with great caution. Short-duration counts cannot be used to determine monthly variability and, depending on the duration of the counts, may not be indicative of typical DOW variability. In addition, inclement weather or other special events may skew the time-of-day patterns in short-duration counts. In most cases, some data is better than no data in establishing typical traffic patterns.

Data Processing

In reviewing the current program and existing non-motorized data, one should also understand the basics of how data is processed by the field equipment and loaded into its final repository, whether that be a stand-alone spreadsheet, a mode-specific database, or a traffic monitoring data warehouse. The following elements should be considered: 

  • What formats (e.g., data structure, time intervals, metadata) are available and/or being reported from the field equipment?
  • What quality assurance and quality control processes are applied to the field data?
  • Are suspect or erroneous data flagged and/or removed?
  • What summarization or adjustment procedures are applied to the field data?
  • How does the current process/system address missing data (e.g., due to equipment hardware, software, or communications errors)?
  • Are estimated or imputed values flagged or documented with metadata?
  • Are the non-motorized data stored/integrated with motorized data? Alternatively, is there an entirely seperate process? 
  • Are data summarization processes automated to the fullest extent possible? At what points are manual review and/or intervention required?

Subjective data manipulation or editing should be avoided. Instead, appropriate business rules and objective procedures can be used in combination with supporting metadata to address missing or invalid data. 

 

 

Summary Statistics

The final step in reviewing the existing program is to consider summary statistics, both those that are currently computed as well as those that may be needed. Permanent count locations should be providing count data 24 hours per day, 365 days per year; however, this continuous data stream is often summarized into a few basic summary statistics, like annual average daily traffic. Because of the greater monthly variability of non-motorized traffic, other summary statistics may be more relevant:

  • Seasonal average daily traffic (includes those months that contain at least 80 percent of the annual  traffic)(seasonal average daily traffic SADT) is a traffic statistic used by the National Park Service in  recreational areas that have very high seasonal peaking (e.g., very high use in summer with low use in  winter));
  • Average daily traffic by month and day of week; and
  • Peak hour volumes for peak seasons (i.e., different user types in summer and winter for shared use  paths).

The review of existing and needed summary statistics should be based on those users and uses that have been identified earlier in the process. In this way, one can ensure that that variety of users has the required information to make decisions.