|Raised medians to separate oncoming traffic, combined with clear and concise signing, help ensure driver safety.|
Although the continuous flow intersection (CFI) is a relatively new concept in the United States, the approach has been in use in Mexico for decades, on more than 40 roadways. Two Mexican-born businessmen, Francisco Mier and Belisario Romo, were co-inventors of the CFI, obtaining a U.S. patent for the design in 1987. But, the concept is primarily based on an old idea of a fellow countryman, engineer Arturo Cedeno.
|Delays are reduced by moving the left turn bay to the left of oncoming traffic.|
|Proper traffic signal timing allows left turns to proceed without stopping oncoming traffic.|
Because part of the delay at a typical high-volume intersection comes from the left-turn cycle of the traffic signals—the through traffic must wait for the traffic turning left—the basic idea behind the CFI is to locate the left-turning lane to the left of facing traffic, eliminating the left-turn signal phase. This left-turn bay is accessed by creating a mid-block signalized intersection. Traffic is permitted to load the left-turn bay, crossing the oncoming traffic lanes during the signal phase servicing cross-street traffic, thus eliminating any traffic conflicts. The signal phase servicing through traffic also services protected left-turn movements. To reduce potential confusion regarding the location of the left-turn lane, the left-turn lane and the straight-through lane are separated by some type of barrier or traffic island.
Since the left-turning traffic no longer has to cross the on-coming traffic, accidents are reduced. Elimination of the left-turn signal increases the amount of “green” time on the main-line route, increasing traffic capacity and reducing traffic delay. Additionally, pedestrian safety is improved with the CFI design, according to research. Pedestrians cross at times when there are no conflicts with turning vehicles. However, pedestrians require two sequential signal phases to complete a street crossing.
The first CFI intersection in the United States was a prototype completed in New Jersey in 1994. This was followed by full-scale projects in New York and Maryland. In 2004, ABMB Engineers, which convinced Mier to allow his U.S. patent to expire in 2003, accepted an invitation to visit the St. Louis office of the national engineering firm for which I worked as a transportation project manager. ABMB Engineers introduced our clients—including the Missouri Department of Transportation (MoDOT) and the St. Louis Department of Highways and Traffic—to the innovative CFI design.
However, as with many innovative ideas, initial interest in the CFI was tempered with many questions and concerns regarding the concept, requiring time for many to become comfortable with the idea. Two major concerns typically surface when considering a CFI:
- increased cost versus a conventional intersection; and
- driver confusion and comfort with using the design.
Construction costs are higher than for a conventional intersection because of increased right-of-way requirements and the need for additional coordinated signal controllers. But, at locations where high traffic volumes may usually require a grade separation, the cost savings can be considerable.
Regarding the driver confusion issue, research from Mexico and subsequent surveys undertaken on the initial New York project found that 80 percent of first-time users easily adjusted to the design. Following one week of use, virtually 100 percent of daily drivers expressed positive comments about the design. Any lingering negatives expressed by users were successfully addressed with adequate advance signing to clear confusion.
Time to try it
By late 2005, I had begun work as the transportation group leader for another civil engineering design firm, Engineering Design Source, Inc. (EDSI). In early 2006, EDSI partnered with another local engineering firm, Pickett Ray & Silver (PRS) to submit a proposal for a conventional intersection layout detailed in a traffic study provided by Crawford Bunte Brammeier (CBB). PRS’ client was local land developer G. J. Grewe, who was expanding a large development just south of the intersection.
After several months with no apparent progress on the project, MoDOT requested additional analysis to the traffic study, including consideration of constructing a CFI. CBB learned in its analysis that a CFI would improve service levels for the intersection compared with the conventional intersection under consideration, and that the levels of service would further improve as the area’s development progressed.
When my contact at PRS told me that the project was back on track, he said that MoDOT now wanted the developer to build some “crazy continuous flow intersection.” I was shocked and excited, and informed PRS that I was very familiar with CFIs, having introduced the concept to MoDOT a couple years previously. MoDOT later confirmed that it had found the 2004 presentation interesting, and since then had looked for the right opportunity to try it out. We submitted a revised proposal and were contracted to begin work on roadway and traffic signal design.
While the additional traffic studies turned the idea into a reality, the extra analysis required an ambitious design schedule to meet the developer’s needs: seven weeks to complete the design, four to six weeks for MoDOT to review the plans, and two more weeks for any further revisions.
We instituted a “divide and conquer” approach with weekly design team meetings to facilitate collaboration on the project. Included in these weekly discussions were G. J. Grewe, PRS, EDSI, CBB, and general contractor R.G. Ross Construction Company. The city of Fenton was actively incorporated into discussions during the early phase of the work; MoDOT was frequently consulted as the work unfolded. We learned early the value of such a collaborative effort that enabled us to address and resolve issues quickly and cost effectively.
Consequently, we submitted our plans on schedule. MoDOT’s review required slightly less than four weeks, so we had no problem delivering our final plans on the delivery date
CFI lessons learned
We learned a number of lessons that are important for others considering the design of a CFI.
First, signing is a critical component of traffic control. The left turn lanes on Route 30 needed to be signed well in advance of the crossovers. A field review of the existing signage was conducted to ensure proper relocation of signs to be moved, and to avoid the possibility of existing signs conflicting with new signage.
Second, frequently scheduled meetings linking all parties involved were critical in seamless project integration. Since this was the first time all parties were involved in this type of endeavor, there was tremendous collaborative value from team insight into issues that arose during the design and construction.
Third, complementing industry guidelines with seasoned team judgment leads to smarter design solutions. Whenever considering innovative design ideas, it is helpful to use industry standards as the foundation, while remaining flexible in consideration of adaptations to these principles. In some cases we erred more conservatively in our approach; in others we realized our scenario allowed greater flexibility in interpretation.
The intersection was substantially complete by Nov. 19, 2007, the date of the ribbon-cutting ceremony. MoDOT said that the intersection is “working great,” and has created very little driver confusion. According to MoDOT projections, the CFI design should currently reduce delays at this intersection from an average of 25 seconds per vehicle to a little more than 17 seconds per vehicle. In 20 years, assuming a 25 percent increase in the number of vehicles, this type of intersection should reduce delays from almost two minutes to just about 30 seconds.
John Hock, P.E., P.T.O.E, is transportation manager and senior roadway engineer at Engineering Design Source, Inc., a St. Louis-based civil engineering and surveying firm. He can be reached at 636-537-5585, ext. 52, or firstname.lastname@example.org.