Attachment D.pdf

1 | P a g e

South Carolina Department of Transportation Traffic Data QA/QC procedures

Table of Contents Short Term Traffic Data Collection QA/QC

• Place the equipment at the correct location

• Place equipment properly

• Equipment pickup

• Site inspection

Office Short Term Traffic Data QA/QC

• Loading data

• Review of field collection techniques

• Completeness checks

Document History

Appendix A – Examples of Counter Image Photos

Appendix B – SCDOT Count Rules

Classification Counts (including details for creating Coverage count Metrocount site list)

Volume counts (including details for creating Coverage count Metrocount site list)

Appendix C – SCDOT Class 14 Rules

Out of Range Rule

Adjusting volumes

Appendix D – SCDOT Rule Based Axle Classification Tree

2 | P a g e

Short Term Traffic Data Collection QA/QC

Field Quality Control and Quality Assurance are to ensure both SCDOT and contractor technicians:

• Place the equipment at the correct location o Ensure that all traffic counters utilized are clearly marked with the name and telephone number of the contractor (or SCDOT if performed by SCDOT staff).

o Utilizing all resources provided by SCDOT (paper maps and KMZ files), technicians will place equipment as near to the marked locations as possible.

o Identify location of schools, colleges or other high volume traffic generators near a count station.

o SCDOT Office staff will utilize submitted images to determine correct location randomly, and specifically when count data is out of range. If it is determined the incorrect location was counted, the contractor shall not charge for data collection and a recount should be collected.

• Place equipment properly o Ensure your safety. Watch for traffic at all times. Wear approved safety vest, and never turn your back on approaching traffic.

o Choose a location, preferably, on a straight, flat roadway with free flowing traffic.

Avoid rutted, pothole sections, driveways, entrances to businesses, etc… o Install road tube perpendicular to traffic flow with tight tube placement (no slapping or bouncing). Tape road tube to roadway surface in heavy volume or high speed locations. No nails should be in the wheel path or sticking up in bicycle/pedestrian paths.

o Use the site list to determine the number of counters for each location. If the site list indicates 1 counter, use 1 counter. If it is determined that the site list is incorrect, please notify office staff. Reference Appendix B for classification and volume count rules.

o Tubes must be the same total length. Spacing must be the same at both ends of the road and the middle to maintain accuracy. If using Metrocounters, spacing from the middle of each tube must be 3 feet.

o Make sure to select the correct setup for the direction the file will be collected.

Example A/B or B/A.

o It is recommended that counters be chained to poles, signs or guardrails. Counters should be off the ground to avoid water saturation.

o If construction, detours, bridge out or other highway signs which may indicate a count may be out of range, notify SCDOT with the County, Station number and reason why the count was not made so that SCDOT can estimate the count.

3 | P a g e o Monitor the traffic counters for accurate operation before leaving the traffic data collection site after equipment set-up. This usually requires watching the counter properly increment until a few vehicles have passed over it.

o Take a GPS enabled timestamped picture of the road tubes (including counter if possible) after installation is complete. Reference Appendix A for examples.

• Equipment pickup o When you arrive at the location, ensure that the tubes are still properly in place and connected to the counter. If they aren’t, check the data to determine if the proper time period (e.g., 48 hours) was collected prior to the tube becoming loose. If not, a recount should be attempted. After one recount attempt, SCDOT will accept 24-hours if 48-hours can’t be collected.

o Start and end times should be as close to the same times as possible for multiple counter locations. Recounts for multiple counter locations should have start/end times as close to the same times and on the same day of week.

o Prior to picking up the equipment, take a GPS enabled timestamped picture of the road tubes (including counter if possible). Reference Appendix A for examples.

o Pickup equipment as soon as reasonable to minimize the chance that tubes will become loose. However, equipment must remain down for the time period of the count (e.g., 48 hours). When using Metrocounters, start with the next hour. For example, if a counter is placed at 9:10AM on a Monday; then ensure that it is not picked up until after 10AM on Wednesday.

o If using Metrocounters:

1. Review Triggers when downloading the file. Sensor balance should be above

92%. In the below example, there are excess B triggers causing a balance of 81%.

4 | P a g e

2. Run the SCDOT Data Scan report to see if data passes the audit. If it fails, try to determine why; and if needed do a recount while at the location. The field technician may also send the data to the SCDOT office staff and request help if he/she is unsure why it failed the audit report.

o Do not leave nails sticking above the ground where pedestrians or bicycles can run over them.

• Site inspection o Site Inspection is a valuable part of our QC process.

o The Field technician manager will schedule a review of each SCDOT traffic count field staff at least once per year. This review will involve meeting the technicians in the field to review the technician’s techniques when placing and picking up equipment.

The manager will send the Traffic Data Manager a short synopsis of the review by email.

o The Contractor is primarily responsible for ensuring their staff is following proper procedures when placing equipment on SCDOT maintained roads. SCDOT will use the submitted images to review for proper location and placement of equipment; and will submit a report to the contractor when discrepancies are found. SCDOT reserves the right to request a field visit with Contractor staff that has more than one report submitted about him/her. If problems persist, SCDOT will instruct the Contractor that the particular Contractor’s staff member is no longer allowed to work on SCDOT maintained roads.

5 | P a g e

Office Short Term Traffic Data QA/QC

SCDOT Office Quality Assurance and Quality of traffic data is to ensure both SCDOT and contractor technicians data is as accurate as possible.

• Loading data o If using Metrocounters, batch load data to convert it to the SCDOT report format using the How to Batch Process MetroCount Files instructions. Note any files which do not convert, and review them individually to determine the cause.

o The Metrocount batch load will automatically use the SCDOT Axle Classification Tree.

However, if you manually process a file, ensure that this classification tree is being used. Reference Appendix D for the SCDOT Rule Based Axle Classification Tree.

o Load traffic data into Traffic Data Software as scheduled.

1. Verify data loaded into software using the software exception reports.

2. Review TDMA.NET software edits checks for data accuracy. There are approximately 30 data edit checks in the TDMA.NET software. Examples of data edit checks include:

• Daily check for zero values

• Invalid number of lanes

• Daily volume directional checks

• Unexpected zero volumes

• Lane distribution

• Daily peak hour volumes

• Midnight versus noon data checks

• Cross vehicle classification data checks (ensure class 2/cars are highest

% of vehicles)

• If edit checks fail, review reason to determine the need for a recount.

If the SCDOT or contractor technician has indicated construction, detours, bridge out or other highway signs which indicate a count may be out of range, estimate the count. No recount is needed.

o If a count is flagged as having excess class 14 vehicles, review the location, pictures, volume, and count history. A decision can then be made to either allow TDMA to adjust the volumes or to get a recount. Reference Appendix C for SCDOT’s rules concerning Class 14s (a.k.a, unknown classification) o If a count is identified by the Count Validation Report as being out of range, verify the location counted is correct. This is done by utilizing the traffic station maps and the photos taken by SCDOT and contractor technicians. The traffic software will locate all photos taken within a half mile from the station selected. Office staff will use historical reports to see if there were any traffic anomalies in previous years. Out of range

6 | P a g e counts are automatically determined by the traffic count software by comparing the previous year count to the current. The following table is used:

o If needed, utilize Metrocount Audit function in Metrocount software to determine data discrepancies in order to determine if a recount is needed.

• Review of field collection techniques o Each regional traffic data office staff should randomly review photos at 25 locations each month. The review should include a random sample of both SCDOT staff as well as contractors.

o Review should focus on proper location and proper setup (straight and tight tubes, counter out of the travel lane, time between put down and pick up is correct, etc…).

o If any discrepancies are noted, a report should be generated in PDF format and sent to the Traffic Data Manager who will review and then forward for proper action.

• Completeness checks o Utilize the Count Verification report to determine which stations do not have data for the current year. Check with the SCDOT technician or contractor to determine if a count can be made. If so, schedule it. If not, estimate the count.

o Add new count stations on HPMS sites and newly functionally classified routes as needed.

o Compile and review the list of paved secondary roads with old traffic count dates.

Have traffic count data collected and loaded into the traffic count software. The goal is to perform 5,000 counts per year if the budget allows.

7 | P a g e

Document History

• Issued on November 28, 2017

• Add review of sensor balance on January 29, 2018

• Added Appendix A on August 23, 2018

• Added additional information in the Place equipment properly and Equipment Pickup sections on January 14, 2019.

• Added Appendix B, C and D on March 25, 2019

• Modified Class Tree (Appendix D) on October 15, 2019. Added rules 15 to capture dump trucks with two optional axles down (5 total axles).

Approved by: _______________________________________ Director of Road Data Services

8 | P a g e

Appendix A – Examples of Counter Image Photos

These photos provide valuable information for Quality Assurance (QA) and Quality Control (QC);

notice this photograph contains:

• An overall view of the road/location

• Placement of the road tubes

• Placement of the Counter (when possible)

This second photo provides a closer view of the counter box and the placement of the excess road tube.

Counter Location

Secured Counter

Make sure Satellites are detected for GPS

Optimal satellite signal

Good satellite signal

Wait for satellite to link

No satellite signal

Make sure Camera Clock is Set

Current Date and Time

Reference: Manual Page 162

9 | P a g e

To check Counter Image Latitude and Longitude (GPS) information:

Cursor over photo and click the “right mouse button” for Properties

Select Properties for the Photo details;

Scroll down to GPS for the Latitude and Longitude values

10 | P a g e

Appendix B – SCDOT Count Rules

Classification Counts (including details for creating Coverage count Metrocount site list)

SCDOT Classification count rules

# of lanes in each direction

AADT

(both directions combined)

# of counters/site needed Comments 1 < 5,900 1

1 >= 5,900 2

One per direction. The Attribute and DirA field needs to be changed for each direction.

2 (or more) < 10,000 2 One per direction

2 (or more) >= 10,000 4 (or more)

One per lane. The Attribute, DirA, and LnA fields needs to be modified for each site.

For each field listed below, “Global” indicates that the value is always the same regardless of the site.

• FIELDS2– The value will always be “Site”. Global

• Site – Format should be CountyName-County#-Station#

• Attribute – The route type, number and auxiliary. Also include the direction if more than one record is created.

• Init – The Field technicians initials (e.g., RTA)

• Description – The from and To for the coverage count station. This will always be generated by TDMA, so there is no need for us to change it.

• LnA - If counts are per lane, this value should be incremented starting at 1 up to the number of lanes in that direction. For example, if the NB has 3 lane, then there should be 3 sites created with LnA values from 1-3.

• LnB, LnAx, and LnBx - The values will always be 0. Global

11 | P a g e

• DirA- Below are the possible numerical values for the direction, a description of these codes, and when to use each of these codes as the default value. The technician may change this value in the field.

Code Description When to use as default

1 North bound, A trigger first

Odd numbered One-Way routes, and the 1st record for Odd numbered two way routes with more than 2 travel lanes

2 East bound, A trigger first

Even numbered One-Way routes, and the 1st record for Even numbered two way routes with more than 2 travel lanes

3 South bound, A trigger first The 2nd record for Odd numbered two way routes with more than 2 travel lanes

4 West bound, A trigger first The 2nd record for Even numbered two way routes with more than 2 travel lanes

South bound A>B, North bound B>A

6 West bound A>B, East bound B>A

North bound A>B, South bound B>A

Odd numbered two way routes with 2 travel lanes

8 East bound A>B, West bound B>A Even numbered two way routes with 2 travel lanes

• DirB, DirAx, DirBx, Mode - The values will always be 0. Global

• Mode – This value will always be 0. Global. (FYI, 1 is a volume only count; 0 is a class count)

• StateBits – This value will always be 2031617. Global

• Spacing – The default values should always be 914. Global (changed from 1000 to 914, October 2018)

• Length- This value will always be 0. Global

• StartTime - The value will always be 0. Global

• Lockout – The value will always be 10. Global

• Latitude, Longitude - These value will always be 10. Global

• RemoteURL- Create this field, but leave it blank. Global

• Offset - The default value should always be 1000. Global

• ZoneName – The default value should always be “None”. Global

• ChannelMode - The value will always be 0. Global

• RsuName – The default value should always be “MC5900”. Global. (changed from MC5600 to

MC5900, October 2018)

12 | P a g e

Volume counts (including details for creating Coverage count Metrocount site list)

The creation of coverage count volume only counts will follow the same rules as #1 above, except as noted below:

# of lanes in each direction

# of counters/sites needed Comments 1 1

2 (or more) 2

One per direction. The Attribute and DirA field needs to be changed for each direction.

• LnA - The values will always be 0. Global

• DirA-o If one counter is being used:

7 if a North/South route, 8 if an East/West route.

o If one counter per direction is being used:

1 for the NB direction 3 for the SB direction 2 for the EB direction 4 for the WB direction

• Mode – This value will always be 1. Global. (FYI, 1 is a volume only count; 0 is a class count)

13 | P a g e

Appendix C – SCDOT Class 14 Rules

Out of Range Rule

1. Should be less than 10% of the total volume. If 10% or higher, count should be reviewed by office staff to determine if it can be allowed to continue with adjusting as detailed below

Adjusting volumes Class 14s most likely occur because more than one vehicle strikes the tubes in a pattern which the counter cannot distinguish. Therefore, use the following rules to account for these unknown vehicles:

1. Multiple the total Class 14 volume by 1.5. Call this value X

2. For each of the other class of vehicles, the new total volume should be:

a. New Class Volume = Original Class Volume + (X * Original Class Percentage excluding class 14s)

Example:

In the example below, class 14s total 55 or 1.73% which is below the 10% threshold. Therefore, continue adjustment.

X = 55 * 1.5 = 83. Now adjust other classifications by their respective percent of the total (excluding class 14s) times X.

For example, New Class 2 volume = 1920 + (61.52% * 83) = 1971

Original Data

Revised Data

Class Total Percent Percent excluding class 14s Class Total Percent 1 15 0.47% 0.48%

1 15 0.47%

2 1920 60.45% 61.52%

2 1971 61.54% 3 972 30.60% 31.14%

3 998 31.16%

4 1 0.03% 0.03%

4 1 0.03% 5 79 2.49% 2.53%

5 81 2.53%

6 16 0.50% 0.51%

6 16 0.50% 7 9 0.28% 0.29%

7 9 0.28%

8 1 0.03% 0.03%

8 1 0.03% 9 105 3.31% 3.36%

9 108 3.37%

10 3 0.09% 0.10%

10 3 0.09% 11 0 0.00% 0.00%

11 0 0.00%

12 0 0.00% 0.00%

12 0 0.00% 13 0 0.00% 0.00%

13 0 0.00%

14 55 1.73% 0.00%

14 0.00% Total 3176 100.00% 100.00%

Total 3203 100.00%

14 | P a g e

Appendix D – SCDOT Rule Based Axle Classification Tree October 15, 2019

• Axles are numbered in order from the front of the vehicle to the back

• Minimum and maximum axle spacing is specified in feet

• A measured spacing meets the minimum specification if it is greater than or equal to the value specified

• A measured spacing meets the maximum specification if it is less than the value specified

• Vehicles not meeting any of these rules shall receive a Class 14 designation.

Rule Class Number of axles

Axle Spacing

1 to 2 2 to 3 3 to 4 4 to 5 5 to 6 6 to 7 7 to 8 8 to 9

Min Max Min Max Min Max Min Max Min Max Min Max Min Max Min Max

0 1 2 3 5.7

1 2 2 5.7 10.5

2 3 2 10.5 13.2

3 5 2 13.2 22.5

4 4 2 22.5 40

5 2 3 5.7 10.5 6 18

6 2 4 5.7 10.5 6 18 1 3.5

7 3 3 10.5 13.2 6 25

8 3 4 10.5 13.2 6 22 1 3.5

9 3 5 10.5 13.2 6 22 1 3.5 1 3.5

10 4 3 21 40 2.7 6

11 5 3 8 22 6 22

12 5 4 8 20 6 22 1 3.5

13 6 3 6 21 2.7 6

14 7 4 6 22 2.7 6 2.7 13

15 7 5 6 22 2.7 6 2.7 13 2.7 13

16 8 3 6 26 11 40

17 8 4 6 26 2.7 6 6 44

18 8 4 6 26 11 44 2.7 12

19 9 5 6 26 2.7 6 6 46 2.7 12

20 9 5 6 26 2.7 6 6 46 2.7 14

21 9 5 6 26 6 46 2.7 11 2.7 11

22 10 6 6 26 2.7 6 6 46 2.7 11 2.7 11

23 10 7 6 26 2.7 6 2.7 46 2.7 11 2.7 11 2.7 11

24 11 5 6 26 11 26 6 20 11 26

25 12 6 6 26 2.7 6 11 26 6 24 11 26

26 12 6 6 26 11 26 2.7 6 6 24 11 26

27 13 7 6 45 2.7 45 2.7 45 2.7 45 2.7 45 2.7 45

28 13 8 6 45 2.7 45 2.7 45 2.7 45 2.7 45 2.7 45 2.7 45

29 13 9 6 45 2.7 45 2.7 45 2.7 45 2.7 45 2.7 45 2.7 45 2.7 45

15 | P a g e

Appendix D – SCDOT Rule Based Axle Classification Tree