Minnesota Department of Transportation

511 Travel Info

Traffic Forecasting & Analysis

Coordinating MnDOT's traffic monitoring and vehicle classification programs

traffic on a highway

Traffic Volume Program

traffic data collection hierarchy graphic
This graphic represents the hierarchy of traffic data collection methods in the Traffic Forecasting & Analysis section. At the bottom of the pyramid, Weigh-In-Motion (WIM) sites collect four types of data at the fewest number of sites. At the top of the pyramid, thousands of sites statewide are used to collect one type of data, traffic volume. View larger version

Quick Look

Traffic data on all traffic volume maps is represented as Annual Average Daily Traffic (AADT), a theoretical estimate of the total number of vehicles using a specific segment of roadway (in both directions) on any given day of the year.  This estimate represents the total number of cars per year divided by 365 and is developed using factors that adjust for season, day of the week, and vehicle type.

AADT estimates are subject to many sources of variability.  Therefore, it is suggested that historical AADT’s be referenced in addition to the most currently available information. Construction effects are unavoidable when collecting traffic data.  If possible, traffic counts are scheduled before a project starts or after it is completed.  It is important to remember that construction affects traffic patterns on the entire road network, another reason why it is valuable to reference historical traffic volumes.

Traffic Volume Program Overview

cars on divided highway
Vehicles traveling along a divided highway.

Traffic volume information is used for many purposes:

  • Statewide estimation of annual Vehicle Miles Traveled (VMT)
  • Annual reporting of VMT and AADT estimates to the Federal Highway Administration (FHWA) for use in federal level travel analysis and determination of funds
  • Part of the formula for annual allocation of state funds for roadway maintenance and construction on the County and Municipal State Aid road system
  • Providing information to help facilitate decision making for planners, engineers, forecasters, businesses, and the general public

The traffic volume program is funded in two parts. Data collection activities are funded by the MnDOT district or responsible agency, while the processing, review and publication is funded via federal SP&R funds. If time and/or resources in any given count season become limited the priorities of the program will be Trunk Highway and HPMS counts, followed by CSAH and MSAS, and finally County Roads.

The traffic volume program produces AADT estimates for approximately 33,000 count locations:

  • 4,500 on Trunk Highways (TH)
  • 13,500 on County State Aid Highways (CSAH)
  • 6,500 on County Roads (CR)
  • 8,500 on Municipal State Aid Streets (MSAS)

There are several data collection methods used for AADT estimation:

Workers installing continuous count sensors on a highway
Workers installing continuous count sensors on a roadway

  • 32,500+ total Short Duration Counts (48 hours) - sites that collect traffic volume; make up majority of all count locations
  • 1,200+ Vehicle Classification sites - collect traffic volume and categorize counts by vehicle type; 20+ of these are in-pavement sensors
  • 70+ Automatic Traffic Recorder, ATR (Continuous) - devices with loops in the pavement that collect traffic volume and sometimes vehicle classification and/or speed data
  • 15+ Weigh in Motion System, WIM (Continuous) - sites that collect vehicle weight, type, speed and volume
  • 240 + Regional Traffic Management Center, RTMC (Continuous) - sites that collect volume data that is stored, processed and converted to an AADT; RTMC's primary purpose is traffic management

The majority of traffic data is collected by MnDOT District staff, but some counties and cities, especially in the Metro, submit their own count data. Approximately ½ of the trunk highway locations and ¼ of the local system locations are scheduled each year (see the Traffic Counting Schedule for details), and most traffic counting occurs between April and October of each year (more details regarding field activities are available in our Program Guidelines for Site Scheduling, Count Collection, and Count Submission).

Raw counts may be submitted anytime outside the set Traffic Counting Schedule. Submission instructions, including the necessary data format, are available in the public Counting Material folder. AADT's are estimated in the fall/winter and sent to the local county/city engineer for comment. The Official AADT's, or Traffic Volumes, are then posted to the web the following spring (Data Products).

Criteria for Determining Traffic Segment Breaks

A traffic segment is a piece of roadway between two traffic segment breaks for which a single AADT estimate is applied. The traffic volumes along segments may vary within an allowable range determined by a 1990 customer/user survey. Based on this survey, the traffic volume program developed the criteria outlined in the Acceptable Percent Change table.

Traffic Counting Schedule

Greater Minnesota

Approximately ½ of the trunk highway locations and ¼ of the local system locations are scheduled each year.  This results in all counties in greater Minnesota having their trunk highways scheduled every two years and local system roads every four years. Trunk highways are major roadways such as interstates, US trunk highways, and state trunk highways. Local system roads are any other roadway designated as a CSAH (County State Aid Highway), CR (County Road), or MSAS (Municipal State Aid Street). Starting in 2010 most County Roads will be scheduled on a 12 year cycle. Requests for data collection outside of the predetermined count schedule should be taken to the State Aid Office.

County road signage in northeastern Minnesota
County road signage in northeastern Minnesota

Sites on a 12 year “super cycle" include all County Roads that are not scheduled more often for the purpose of HPMS reporting (new policy as of 2010) and sites that have had an AADT less than 100 for two consecutive cycles. Sites are assigned one of three super cycle designations (A, B, C). One of these is assigned each cycle (4 year cycle X 3 letter designation = 12 year super cycle).

  • Super Cycle A – 2010 to 2013
  • Super Cycle B – 2014 to 2017
  • Super Cycle C – 2018 to 2021

Metro Area (Anoka, Carver, Dakota, Hennepin, Ramsey, Scott, Washington counties)

Trunk highways are scheduled over a two year period and information is published in the even years. CSAH and CR counts are provided by the counties and the information is published in the odd years. MSAS counts are provided by the municipalities and the information is published every two or four years according to the Municipal Traffic Counting Schedule.

Data Screening and Recounts

Factoring of Incoming Short Count Data

Forty-eight hour short duration traffic counts are factored using the following formula:

48 Hour Raw Traffic Count


Seasonal Adjustment Factor (SAF)


Axle Correction Factor (ACF) (Trunks Only)


48 Hour Adjusted Count

Forty-eight hour adjusted counts are then divided by two to give the final 24 hour adjusted count that is compared to history (see below: Screening of Incoming Short Count Data).

Seasonal Adjustment Factors (SAF) correct incoming data to account for differences between weekday and weekend traffic as well as fluctuations in traffic volumes from month-to-month. The complete set of factors is listed in the AADT Adjustment Factors for Short Duration Traffic Volume Counts document.  SAF's are developed using information from the 70+ Automatic Traffic Recorders (ATR's) dispersed throughout the state that continuously count traffic all year long.

SAF's are created using WARD's hierarchical clustering procedure. This procedure and subsequent analysis results in five “cluster groups” based on characteristic travel patterns. Each “cluster group” has a set of factors for each month of the year and “middle day” of the week.  All short counts are taken for 48 hours with a “middle day” of Tuesday, Wednesday, or Thursday. Every count location is assigned to one of the five “cluster groups.” Traffic count locations in close proximity to an ATR are adjusted using temporal factors calculated directly from the associated ATR. (Continuous count locations are not adjusted using factors; instead, the AADT is calculated as the annual traffic volume divided by 365). The “raw” 48 hour count totals are then adjusted by the specific month and day of the week SAF from its location's “cluster group."

Axle Correction Factors (ACF) are used on the trunk highway system only.  These factors are calculated from vehicle classification counts and correct for the heavy commercial traffic, or vehicles with three or more axles, that contribute “extra” axle hits to the raw 48 hour traffic counts.  For more information on vehicle class counts and how ACF's are determined please contact the Vehicle Classification Program.

Screening of Incoming Short Count Data

Traffic observation for manual short counts
Traffic observation for manual short counts

Incoming traffic counts are factored and the resulting 24 hour adjusted count is then compared to the most recent official AADT for the same location. If the adjusted count falls within the Acceptable Percent Change criteria, then it is determined to be valid and the official AADT will be estimated using this new data.

If the adjusted count falls outside of the Acceptable Percent Change criteria, then the count is examined further. The complete count history is reviewed to determine whether this count is similar to historic volumes. Count comments regarding construction, new development, closures, and any other explanatory information are examined to possibly validate the count information. If a recount is necessary, it is assigned based on the Recount Decision Tree.

At the end of the season when all counts and recounts are completed, official AADT's are estimated using all available data. A new AADT will not be published if the site does not have data collected or confirmed in that season.

AADT Estimation Procedures

AADT estimates are determined at the end of each count season once all available count information has been collected.  The current standardized methods for estimating AADT on all trunk highways, CSAH's, CR's, and MSAS's are illustrated below.

AADT estimation methods are based on the following assumptions:

  • Adjusted counts at any location are highly variable.
  • Adjusted counts at any location vary around a mean that is relatively constant through time.  Therefore, historic adjusted counts and AADT information can be used to validate current adjusted count information.
  • AADT at any location can change from its historic pattern when the following factors are present: new development, closure of a traffic “generator,” long-term construction, changes in the roadway network, etc.


    48 Hour Adjusted Count

    AADT Estimation Procedure

    AADT Decision Tree

    AADT Estimation Procedure

    Federal Rounding Conventions



All traffic counts are analyzed using the AADT Decision Tree and the Acceptable Percent Change table. If count information indicates a change in traffic pattern on a roadway network, GIS spatial analysis is used to identify all locations affected by this change. Historical information, including the complete count history, is analyzed in order to identify past AADT estimation errors. For a small number of locations, an accurate traffic count is not able to be obtained due to current conditions or other special circumstances.

I-35W Saint Anthony Falls Bridge in Minneapolis, completed in the fall of 2008
I-35W Saint Anthony Falls Bridge in Minneapolis, completed in the fall of 2008

For all CSAH, CR, and MSAS locations a Draft AADT is sent to the local authority for their review prior to the official AADT product release.  These local authorities have the opportunity to comment and suggest changes to the Draft AADT based on additional information they provide.  Please refer to the Draft Review Process and Timeline for a more detailed explanation of this procedure.

Additional AADT information of note:

  • Data collected from continuous count locations is validated and the AADT is estimated using alternate methods.  
  • Final AADT estimates are provided to MnDOT from a limited number of partner agencies.

The Official AADT's are then posted to the web the following spring (Data Products).

Draft Review Process and Timeline

In the fall/winter of each collection year a Draft AADT report is sent out to each county or municipality that had traffic data collected that season (see Traffic Counting Schedule).  This process allows the local authority to have a chance to comment on their traffic volumes before they are made official.  Draft AADT can be viewed spatially via the Traffic Mapping Application (static maps of the Draft AADT are available upon request). 

The deadline for comments is 10 business days from the receipt of the report. All reviews must be submitted by the date specified in the notification in order to ensure the timely completion of the official product and to allow sufficient time for preparation of the subsequent year’s traffic counting materials. 

A Draft AADT Comment Form is included with the Draft AADT report and can be returned via email or printed and returned via US mail to the following address:

Attn: Traffic Volume Program
395 John Ireland Blvd, MS 450
St. Paul, MN 55155-1899

All Draft AADT comments should include the following information:

  • The Route
  • Sequence Number (SEQNUM) or location description (nearest intersections)
  • The Draft AADT
  • Detailed explanation of why the traffic volume should be changed.  Additional documentation such as traffic counts and construction/detour route information can also be submitted for consideration.

Draft AADT review and its relationship to State Aid Needs and Design Standards

Comments will be reviewed in a timely manner and a response will be sent as soon as possible.  This response will include a description of all changes made to the Draft AADT as well as any additional information needed to explain the decisions made. After the response to the Draft AADT comments is sent, any additional comments, information, or appeals to the decisions made must be submitted within two weeks. The Official AADT product is then posted to our website the following spring (Data Products).

Construction Policy

MnDOT construction crew working on a highway project
MnDOT construction crew working on a highway project

In the AADT estimation process our overall goal is to represent “typical” traffic conditions for the current collection cycle (the time between the current year and the next scheduled collection year; see Traffic Counting Schedule).  Since these estimates are used for a number of years, it is necessary to attempt to represent “normal” conditions.  In order to minimize construction effects on AADT estimates, we have adopted the following policies:

  • Traffic counts scheduled in areas affected by construction should always be taken prior to or after a construction project is completed. This includes count locations in the surrounding area, especially detour routes.
  • If a project lasts longer than the available counting season (generally April through October) but is not considered a long term project, the following is considered:
    • When no data is available or the data from the current year is suspect, the AADT will not be updated for that count cycle, unless special circumstances are identified.
    • When there is a new count location and no data is available, an AADT estimate may be generated using the surrounding count locations as well as the ITE Trip Generation manual. Analyst discretion will be used to make the best estimate possible in these situations where a new AADT is considered critical.
    • Any traffic counts taken in the year following the scheduled cycle would be used to estimate updated AADT for the year in which they were collected.
  • If there is a long term construction project lasting more than one year, the resulting effect on traffic patterns in the area are considered “typical” for that count cycle. In this situation traffic data will be collected and AADT estimates will reflect changes in traffic patterns due to the construction project. The AADT for that cycle will be noted as impacted.
The Crosstown Reconstruction Project, completed in 2010, heavily influenced AADT on I-35W and nearby roads
The Crosstown Reconstruction Project, completed in 2010, heavily influenced AADT on I-35W and nearby roads

Current MnDOT Construction Projects

Traffic Volume Program Contacts

Overall Program Administration: Procedures, Count Cycles, Data Processing, AADT Estimation, Draft Commenting, Official AADT

MnDOT Count Managers: Field collection activities, coordination of local information (construction, detours, mowing and street cleaning)

District Count Manager Email Phone
D1 - Duluth John Hoivik john.hoivik@state.mn.us

D2 - Bemidji

Todd Larson todd.larson@state.mn.us

D3A - Baxter

Jerilyn Swenson jerilyn.swenson@state.mn.us

D3B - St. Cloud

Laurie Pomrenke laurie.pomrenke@state.mn.us

D4 - Detroit Lakes

Carl Holt carl.holt@state.mn.us

D5 - Metro

Barton Kimble barton.kimble@state.mn.us

D6 - Rochester

Jeff Rieder jeffrey.rieder@state.mn.us
D7 - Mankato Brad Bruegger bradley.bruegger@state.mn.us
D8 - Willmar Randy Lindstrom randy.lindstrom@state.mn.us

Twin Cities Metro Contacts

Counties and municipalities in the metro area take their own traffic counts on County State Aid Highways (CSAH), County Roads (CR) and Municipal State Aid Streets (MSAS).  These counts are then submitted to the Traffic Volume Program for the official MnDOT traffic volume maps.  Some metro area agencies have their own policies, procedures, and guidelines that direct their traffic monitoring activities; therefore, these agencies may not use the same methods outlined above. It is important to note that more information may be available from the following agencies regarding program standards, additional traffic volumes on roads not reported by MnDOT, traffic flow maps, or more recent count data.


(click for website)


(click for email)

City of Minneapolis Jim Steffel

City of St. Paul

Chris Gulden


Anoka County

Sean Thiel

Carver County

Steve Greengard

Dakota County

Phong Vu

Hennepin County

Peter Lemke
Ramsey County Erin Laberee
Scott County Lisa Freese
Washington County Nik Costello

For a list of contact information for all Metro City Engineers visit:

[Back to the top]

Vehicle Classification

Minnesota’s vehicle classification program is designed to gain an understanding of the volume and type of heavy commercial vehicles that utilize Minnesota’s portion of the National Highway System as well as other trunk highways, CSAH's, County Roads, and MSAS's throughout the state. This is being accomplished through an integrated system of data collection devices that includes both continuous and short-duration counting methods. Vehicle classification data is collected from weigh-in-motion sites (WIM), continuous classifiers (ATR), pneumatic tube counters, and manual counts; and the data is archived in a database for analysis and reporting purposes. Heavy commercial annual average daily traffic (HCAADT), which is the theoretical estimate of the total number of heavy commercial vehicles using a specific segment of roadway (in both directions) on any given day of the year, is derived from these counts. This estimate represents the total number of heavy commercial vehicles per year divided by 365 days and is developed using factors to adjust for season. Heavy commercial vehicles are defined as all trucks with at least two axles and six tires.

Pneumatic tube technology is used to collect vehicle classification data
Pneumatic tube technology is used to collect vehicle classification data

Vehicle classification data is collected from thousands of count sites throughout Minnesota:

  • 15+ WIM sites
  • 30+ ATR continuous classification sites that utilize Piezo electric sensors (ATR Monthly Vehicle Classification Reports)
  • 2000+ total vehicle classification count locations (Vehicle Classification Site Maps), with 1200+ active locations (approx. 95% tube, 5% manual) that are each scheduled every 6 years
  • Special counts (project related)
  • Parent/Child heavy commercial relationships established for all trunk highway traffic segments where no classification has taken place; all major segments have at least one vehicle class count for HCAADT determination.

All vehicle classifiers collect data on axle spacing for thirteen classes of vehicle types (Vehicle Classification Scheme) that are stored in the database by hour. Body type data is available for all data collected manually. Tube counts are set for 48 hours and manual counts are taken for four hours (generally from 9 AM to 1 PM), both during weekdays and between the months of April and October.  The short duration classification counts are adjusted to AADT volumes by using factors developed from the continuous counters.  These factors take into account the variations of truck volumes by month and day of week.  The parent/child relationships developed enable us to automate the process for the production of heavy commercial vehicle type volumes and estimated loadings at all HPMS segments.

Count Numbers and Cycles for Vehicle Classification Sites:

  • 1000-1999 = Trend sites
  • 2000-2999 = Special request
  • 3000-3999 = Special CSAH
  • 4000-6000 = CSAH/County Road
  • 6000-6999 = N/A
  • 7000-9000 = Update sites (6 year cycle, summer)

[Back to the top]

Automatic Traffic Recorder (ATR)

Automatic traffic recorders are permanent devices in the pavement surface that continuously and automatically collect traffic data. Currently, there are 70+ active devices maintained by MnDOT throughout the state-- 30+ in the seven-county metro area and 35+ in greater Minnesota. All stations produce vehicle volume data; however, some ATR stations collect additional types of data depending upon their equipment and sensors. ATR's (30+) equipped with multiple axle sensors per lane collect vehicle classification data to distinguish between motorcycles, cars, and trucks, and some ATR’s (35+) use permanent loop or Piezo sensors to record vehicle speed. Consequently, after the data polling process is complete, this information can be aggregated to report vehicle volume, speed, and vehicle classification characteristics.

Located along US 169 near Onamia, ATR 204 collects volume, speed, and classification data
Located along US 169 near Onamia, ATR 204 collects volume, speed, and classification data

ATR data is used by the Traffic Forecasting & Analysis Unit to estimate statewide VMT trends, calculate design hour volumes (K factors), directional splits (D factors), and seasonal adjustment factors. K and D factors affect decisions made when designing roads, and Seasonal Adjustment Factors modify 48-hour weekday traffic counts that are used to produce AADT estimates for roads throughout the state. The data also contributes to our understanding of design hour volumes which, in turn, affect decisions made in the design of roads. Since ATR stations report continuously but sometimes fail to record data, the data record for each station is augmented with contextually determined and historically valid estimates. These estimates help to eliminate bias that could be introduced by non-reporting. When estimation is not possible, an ATR is determined to be unusable or "inactive" and will not be used in reporting. ATR’s that have been affected by weather (snowfall, heavy rain, fog, etc.), construction, detours, or other forms of traffic disruption that represent "atypical" data are noted not included in factoring efforts.

There is a wealth of ATR data available on the Data Products page, a majority of which can also be found in monthly and annual reports. The ATR Monthly Volume Station Reports examine variations in monthly average daily travel (MADT), monthly average weekday/weekend travel (MAWDT/MAWET), and total volume at each ATR with the use of data tables, graphs, and qualification notes. The annual Continuous Traffic Recorder Report provides data and analysis for all ATR and WIM sites for an entire year. The data provided includes AADT and MADT figures, Station Location maps, 10-Year and Year-to-Year Percent Change of AADT maps, Highest Hourly Volume reports, and Truck Volume graphs.

smartsensor hd device
Using radar technology, SmartSensor HD devices have the capability of detecting up to ten lanes of traffic at a time

SmartSensor HD is MnDOT’s newest non-intrusive automatic traffic counting device. The SmartSensor HD from Wavetronix LLC uses radar technology to collect volume, speed, vehicle length, and length based classification. This unit has a detection range of 250 feet with the ability to detect up to ten lanes of traffic concurrently.  After installation SmartSensor HD requires little or no on-site maintenance, and are very economical when compared to other Automatic Traffic Recorders.

The Office of Traffic, Safety, and Technology (OTST) and the Office of Maintenance along with the Office of Transportation System Management (TSM) are working together to deploy SmartSensor HD to collect traffic data throughout the state. As of April 1, 2012, OTST and TSM installed ten SmartSensor HDs. Five more units will be deployed in the summer of 2012 and, in the near future, we expect to see more of these units on the roads. Consult our SmartSensor HD PDF for more information.

[Back to the top]

Weigh-In-Motion (WIM)

In addition to collecting the type of data that ATR's capture (volume, classification, speed), weigh-in-motion (WIM) devices collect axle loadings, vehicle and axle configuration, and truck volume characteristics.

MnDOT currently uses the WIM system data in Daily Weight Enforcement Reports for a number of sites.

When a vehicle passes over a WIM site, a sensor emits an analog signal whose strength is directly proportional to the axle weight of the vehicle, and the approximate weight is recorded.

A truck with an "Oversize Load" sign approaches a weigh-in-motion site
A truck with an "Oversize Load" sign approaches a weigh-in-motion site

After the volume, class, speed, and weight data from all WIM sites is processed and analyzed, the WIM Monthly Station Reports are posted online and data is submitted to the FHWA. Analysis of annual trends in the data can lead to a better understanding of changing truck fleet characteristics and truck weights.

Gross vehicle and axle weight monitoring is useful for a wide range of reasons, including:

  • Pavement design, monitoring, and research
  • Bridge design, monitoring, and research
  • Size and weight enforcement
  • Legislation and regulation
  • Administration and planning

[Back to the top]

Traffic Forecasting

Traffic forecasting is the production of future traffic volumes and loads on a specific roadway segment. The projections are derived by trending historic vehicle classification data and considering the effects that future changes in the socioeconomic factors and traffic trends will have on the particular segment.

bumper-to-bumper traffic
Traffic forecasting plays a major role in determining infrastructure capacity. It can help predict how much traffic to expect on highways twenty years from now.

The most common requests for traffic forecasts in the TFA Unit are:

  • Base and design year annual average daily traffic (AADT)
  • Base and design year heavy commercial annual average daily traffic (HCAADT)
  • Design hour volumes (DHV) with associated directional distribution (DD)
  • 20- and 30-year cumulative Equivalent Single Axle Loads (ESALs)

A basic element of roadway design is the estimation of ESALs, which are numeric factors that represent the effect axles in different configurations and weights have on pavement performance as compared to the effect of a single 18,000-pound axle.  The estimates are calculated by forecasting the traffic the road will be subjected to over its design life, then converting the traffic to a specific number of ESALs based on the composition of that traffic. This helps to develop structural and pavement design options for future roadways.

The MnESAL Traffic Forecasting Tool is an Excel program that uses AADT/HCAADT and vehicle class data to forecast 20-year BESALs and 35-year CESALs. 

The following are examples of how traffic forecasting is used in transportation policy, planning, and engineering:

  • Calculations of infrastructure capacity (e.g., how many additional lanes will be added to an interstate highway)
  • Estimations of financial and social viability of projects (e.g., how a new trunk highway route will affect a suburban area)
  • Calculations of environmental impacts (e.g., how the air and noise pollution from a future airport will affect surrounding neighborhoods)

Traffic Forecasting Contacts

District Forecaster Email Phone
D1 - Duluth James Miles james.miles@state.mn.us

D2 - Bemidji

Adam Wick adam.wick@state.mn.us

D3 - Baxter/St. Cloud

Gerry Buscher



D4 - Detroit Lakes

Bruce Bryngelson

Sara Maninga





D5 - Metro

Jim Henricksen

Michael Corbett





D6 - Rochester

Tracy Schnell tracy.schnell@state.mn.us
D7 - Mankato

Debra Yates

Scott Thompson





D8 - Willmar Yuzhe Guan rachel.guan@state.mn.us
CO - St. Paul (Office of Transportation System Management)

Shannon Foss

Gene Hicks

Quinten Geddes







[Back to the top]