Consider changeable message sign (CMS) positioning, data archive requirements, and traffic demand when considering deployment of a dynamic late merge system.
Experience of the Minnesota Department of Transportation in deploying a dynamic late merge system.
Made Public Date


United States

Evaluation of 2004 Dynamic Late Merge System


Lane closures in work zones present a challenge, often impeding traveler mobility as well as safety. The Dynamic Late Merge System (DLMS) deployed by the Minnesota Department of Transportation (MNDOT) is a traffic control strategy designed to improve throughput and driver safety during such lane closures. This DLMS, which consists of three changeable message signs (CMS) and a remote traffic microwave sensor (RTMS) detector, can dynamically change its lane use instructions based on the current traffic demands. Under light traffic conditions, an early merge strategy can be utilized, but during periods of heavy demand, this can be altered to a late merge strategy.

MNDOT is in the second year of evaluating the deployment of DLMS. In 2003, MDOT deployed and tested the system in a single location; this was followed by the 2004 evaluation of DLMS at three additional sites: I-494 Northbound in Plymouth, I-94 Southbound in Plymouth, US 52 in St. Paul and I-35 around Lakeville. Five potential benefits of the new system were identified and used as measures of effectiveness to evaluate the DLMS. The potential benefits include:

  • Shorten queue lengths before work zones
  • Increase traffic capacity through work zones
  • Reduce aggressive driving
  • Decrease number of work zone related incidents
  • Reduce travel time

This report presents MNDOT’s 2004 evaluation findings separately for each site and provides a set of lessons learned for future deployments of DLMS.

Lessons Learned

At each of the deployment sites of the Dynamic Late Merge System (DLMS), three CMSs were utilized, in addition to static control signs typically used at construction lane closures. The two CMSs furthest upstream from the taper (the point where the lanes merge) encouraged drivers to "use both lanes" was being utilized as traffic approached the work zone, and observations made at each of the deployment locations provided further insights on driver behavior.

The findings from the four evaluation sites of the DLMS indicate that the percentage of drivers utilizing the discontinuous lane during the deployment increased significantly when the CMS were activated, and in addition, queue lengths were minimized. Based on MNDOT's experience with DLMS, the following set of lessons learned is presented.

  • Consider the position of the changeable message signs (CMS). The CMS should be placed on the shoulder or the median nearest to the discontinuous lane. This position makes the signs much more visible to the drivers who may have a tendency to leave the lane early instead of utilizing the full capacity. In the I-494 deployment, the CMS closest to the work zone (placed on the right shoulder) was not easily viewed by drivers, and this may have contributed to early merges.
  • In addition, the CMS closest to the taper point should be positioned adjacent to the last static merge sign. This position is advantageous because there is no contradiction in instruction from signs at different locations. This positioning also gives an additional buffer zone of distance to complete the merging procedure, which could encourage drivers to use this lane. In the I-494 northbound deployment, the CMS instructing drivers to "merge here" was placed upstream from a static merge sign. As vehicles approached the lane closure, they chose to merge early, at the site of the static sign, rather than merge into the continuous lane further upstream at the point suggested by the CMS. For the I-494 Southbound deployment, the last CMS (instructing drivers to merge) was placed adjacent to the last static merge sign, and this significantly reduced early merges, as drivers tended to merge as they approached the last CMS.

  • Archive traffic data. In order to determine the overall benefit of the traffic control strategy, it is necessary to obtain data on traffic conditions at various points along the road. This data is also important to controlling the operations of the system. As part of the contract with equipment installations, there should be a provision requiring the system to archive data on volume and speed by lane along with an additional field that indicates whether the CMS messages were active. In the I-494 Northbound deployment, for example, MnDOT's regional Traffic Management Center's camera images revealed that the system was active overnight, despite the lack of traffic demand. One method of ensuring system reliability is to require the contractor to record current system state and traffic data at regular intervals.
  • In addition, every effort should be made to ensure that the most accurate detectors are used as part of the DLMS. Currently, the remote traffic microwave sensor (RTMS) detectors have problems accurately reading traffic conditions during low traffic volume conditions. RTMS detectors use occupancy and a vehicle length constant to calculate an average speed. The different-length vehicles are separated into separate bins and the RTMS has a difficult job performing this task when there are very few vehicles to use for the readings.
  • Consider traffic volume demand. The data from these deployments suggest that there is no need for a DLMS if there are no time periods during a construction project that would see demands in excess of 1500 vehicles per hour. For lane closures that anticipate traffic volumes above this threshold, a benefits/cost analysis should be conducted to determine if these demands would be frequent enough to warrant the deployment of the DLMS. In the US 52 deployment, low traffic volumes appeared to have an impact on merging behavior. Vehicles were not commonly observed using the discontinuous lane to the merge point very frequently, due to the fact that the queue (in the continuous lane) was typically quite short.

The DLMS allows drivers to fully utilize both lanes during periods of congestion and eliminates confusion over lane use issues by indicating where drivers should merge. These instructions should aid in eliminating aggressive driving conditions, where vehicles with widely varying speeds are driving in adjacent lanes or vehicles block the discontinuous lane to prevent vehicles from moving closer to the taper location. In addition to these safety benefits, the evaluation findings indicate that the DLMS resulted in increased use of the discontinuous lane and minimized queue lengths, suggesting that the DLMS also has mobility benefits.

Evaluation of 2004 Dynamic Late Merge System

Evaluation of 2004 Dynamic Late Merge System
Publication Sort Date
Minnesota DOT

(Our website has many links to other organizations. While we offer these electronic linkages for your convenience in accessing transportation-related information, please be aware that when you exit our website, the privacy and accessibility policies stated on our website may not be the same as that on other websites.)

Goal Areas
System Engineering Elements

Focus Areas Taxonomy: