Deploy a variable speed limit system only after the software systems required to support it are mature and reliable.
Experience from iFlorida Model Deployment
Made Public Date


United States

iFlorida Model Deployment Final Evaluation Report


The iFlorida Model Deployment, which was started in May 2003, called for the Florida Department of Transportation (FDOT) District 5 (D5) to complete the design, build, and integration of the infrastructure required to support operations in 2 years. The required infrastructure was extensive, spanned numerous stakeholders, and included many technologies that were new to FDOT D5, such as sophisticated traffic management center (TMC) operations software, a wireless network deployed along I-4, an interface to Florida Highway Patrol Computer Aided Dispatch (FHP CAD) data, statewide traffic monitoring, and many others. The iFlorida plans also called for deployment of these technologies in ways that required coordination among more than 20 stakeholders. It was an ambitious plan that would result in dramatically different traffic management operations for FDOT D5 and other transportation stakeholders in the Orlando area.

In implementing the iFlorida plan, FDOT faced many challenges ranging from higher failure rates than expected for some field hardware to difficulties with the Condition Reporting System (CRS) and Central Florida Data Warehouse (CFDW) software. "Despite these challenges, it can be readily claimed that the overall iFlorida Model Deployment was successful," noted in the final evaluation report for the iFlorida Model Deployment, published in January 2009.

The difficulties associated with the iFlorida Model Deployment provided many opportunities to identify lessons learned from the experiences they had. The most important of these are presented below in a series of lessons learned articles.

Lessons Learned

As part of the iFlorida Model Deployment, FDOT deployed the field hardware needed to support a variable speed limit (VSL) system on a portion of I-4 and maintained a network of loop detectors on I-4 that could support determination of when reduced speed limits should be implemented. Due to problems with CRS, which was meant to be software system for the FDOT District 5 (D5) regional traffic management center (RTMC), a functioning VSL system could not be started prior to the completion of the iFlorida project evaluation (January 2009). Nevertheless, FDOT’s experience with VSL system deployment did bring to light a number of lessons learned that might benefit others considering VSL:

  • Identify statutory and regulatory speed limit requirements before considering the use of variable speed limits. Statutory restrictions limit the applicability of VSL on I-4. The minimum speed limit of 40 mph, combined with the normal speed limits of 50 and 55 mph on I-4 in Orlando, meant that speed limits could be varied only over a small range. The requirement for an engineering and traffic investigation before speed limits could be changed was also problematic. FDOT determined that it could perform this type of investigation once to identify the types of traffic conditions that would warrant lower speed limits, and RTMC staff could then change speed limits after verifying that the specified conditions for lowering speed limits had been met.
  • Develop the concept of operations for VSL before designing the VSL system and validate it against historical data. Because of the long lead time for deploying the iFlorida field equipment, FDOT contracted to deploy the VSL signs before completing the VSL concept of operations, specifying that the signs be deployed in the area where congestion most often occurred. The VSL concept of operations emphasized the benefits of lowering speed limits upstream from a congested area. A comparison of the VSL sign locations to historical patterns of recurring congestion indicated that the signs were to be located to cover the area upstream of the points where eastbound congestion typically began, though not the area upstream of the full extent of the congestion at its peak. The VSL signs appeared to be west of the area congestion typically occurred in the westbound direction. FDOT noted that the agency had discussed each of these issues prior to selecting the site for deploying the VSL signs; however, because these issues were not addressed in the concept of operations document, the basis for the VSL design was not documented.
  • Ensure that the algorithms for recommending speed limit changes can detect and correct for low vehicle speed observations that are not related to congestion. A review of historical speed data identified a number of cases where low speed measurements did not appear to be related to congested conditions. In some cases, this appeared to be caused by a faulty detector (e.g., the detector consistently reported low speeds). In others, a single low value would be embedded in a series of otherwise normal values. In either case, the algorithms for recommending lower speed limits should include methods to detect and disregard low vehicle speed measurements that are not related to congestion. (In the case of iFlorida, the problems with the CRS made this very difficult or impossible to do.) Problematic detectors might also be taken offline until repaired, in which case the algorithms must be able to adapt to the fact that detectors may be taken offline. The robustness of those algorithms could be verified by applying them to historical data.
  • Require operator approval of all speed limit changes. FDOT's design of the VSL system called for an automated system to monitor traffic and weather conditions and recommend lower speed limits when conditions appeared to warrant them. RTMC operators would be required to investigate traffic conditions and approve the recommendation only when warranted by current traffic conditions. Since any algorithm for recommending speed limits is likely to fail on occasion, this approach will prevent those failures from resulting in changing speed limits at inappropriate times.
  • Deploy a VSL system only after the systems required to support it are mature and reliable. Because of its participation in the Model Deployment and the required milestones, FDOT had to deploy infrastructure and develop the operating system concurrently. This resulted in the deployment of VSL signs before the CRS software for supporting those signs had proven reliable. These signs remained set at the fixed speed limits when the initial version of the CRS software did not operate reliably and 2 years of additional work by the CRS contractor did not remedy the problems. Thus, FDOT had to bear the cost of deploying VSL signs without obtaining the benefit of using variable speed limits.

After the CRS software for RTMC was abandoned, FDOT began working with a new contractor to migrate to a different traffic management application, SunGuide, for the RTMC. By August 2007, FDOT was using SunGuide at the RTMC. By November 2007, SunGuide supported most of FDOT's basic traffic management needs and FDOT's confidence in the system was growing. At about that time, FDOT elected to restart its delayed VSL project. In December 2007, FDOT reviewed the VSL concept of operations and began to develop a new approach for triggering speed limit changes. The agency also began testing the SunGuide software capabilities to support VSL operations. As the national evaluation was ending, it appeared that FDOT's experience with VSL on I-4 was ready to begin. The VSL system is expected to improve mobility and efficiency of traffic operations along I-4 near Orlando.

iFlorida Model Deployment Final Evaluation Report

iFlorida Model Deployment Final Evaluation Report
Publication Sort Date
Robert Haas (SAC); Mark Carter (SAIC); Eric Perry (SAIC); Jeff Trombly(SAIC); Elisabeth Bedsole (SAIC): Rich Margiotta (Cambridge Systematics)
United States Department of TransportationFederal Highway Administration1200 New Jersey Avenue, SEWashington, DC 20590
Goal Areas
System Engineering Elements