|FHWA > HfL > Projects > California Demonstration Project: Pavement Replacement Using a Precast Concrete Pavement System on I-15 in Ontario > Project Details|
California Demonstration Project: Pavement Replacement Using a Precast Concrete Pavement System
The highway within the project limits is a busy commercial and light industrial corridor with a 2005 average annual daily traffic (AADT) of about 43,450, with 2.9 percent heavy commercial vehicles. Within the project limits, the existing highway had three at-grade signalized intersections and three at-grade nonsignalized intersections with significant congestion and numerous safety issues. The general project location is indicated in figure 1. Figure 2 shows the project limits.
Figure 1. Project location. (Courtesy Caltrans, Google™)
This segment carries eight lanes of traffic—four mainline lanes in each direction—with auxiliary lanes accommodating merging traffic from area crossroads. The estimated ADT is 200,000 vehicles with 6 percent trucks. Caltrans evaluated detailed traffic estimates and projections for the planning and design of this project. Historical and projected traffic data on I-15 are presented in table 1 for south of the junction with I-10 and in table 2 for north of the junction with I-10.
A large amount of traffic merges into and out of the eastbound and westbound lanes of I-15; Route 60 at the south end of the project carries six to eight lanes of traffic and I-10 toward the north end of the project carries eight lanes of traffic.
The two outer lanes were rehabilitated in both directions under this project, which amounted to about 12 lane-mi of continuous lane replacement and intermittent slab replacement. Other roadway portions that underwent rehabilitation included interchange ramps, freeway-to-freeway connectors, and asphalt shoulders. To support the major rehabilitation activities and accommodate traffic flow and detours during the construction work, the project also entailed median paving, new median barriers, widening of the inside shoulder, bridge widening, and structure crossings. Also included in the project was other bridgework, which consisted of deck rehabilitation, replacement of structure approach slabs, and upgrading of bridge approach rails.
After an extensive planning process, Caltrans began design activities in 2004. The large traffic volumes and the inability to shut down or even significantly reduce traffic flow made necessary a critical review of construction and staging options during the planning, design, and construction phases. The presence of the interchange is an added challenge to maintaining traffic during construction; as many as six lanes approach the interchange in each direction. Caltrans adopted an integrated preconstruction analysis approach to compare all feasible scenarios for the project limits and select the best approach in terms of schedule, traffic delay, and total cost. Consequently, Caltrans used traffic modeling tools and traffic network analysis tools in its preconstruction analysis, which was conducted in two phases:
CA4PRS for Comparison of Alternate Rehabilitation Scenarios
Caltrans used the CA4PRS software tool to compare alternative construction scenarios and analyze the preferred alternative. The purpose of analyzing the preferred alternative was to provide a detailed estimate of the number of working days and closures needed for each construction stage. Caltrans analyzed five alternatives to determine the most efficient in terms of road user costs (RUC) based on each phase of construction and production rates for each alternative.
The study validated the preferred alternative (as-built) based on construction schedules, traffic, and RUC of each alternative. The five alternatives and the estimated closure duration are as follows:
Table 3 is a summary of the results from the analysis.
**Total cost = 1/3 RUC + Agency Cost
The main component of the as-built alternative, or option 1, was the maintenance-of-traffic plan. This alternative required that the project include bridge widening on all bridge structures within the project limits and paving of the median. Also, four lanes of traffic were maintained by shifting the mainline traffic onto the median shoulders while the two outermost lanes were rehabilitated, as illustrated in figure 2. This process was repeated for both directions. The medians received an asphalt overlay to handle the mainline traffic volumes. Several structures were widened to accommodate traffic though the medians. However, because of conflicts with existing median bridge columns at the I-10/15 interchange crossover structures, the median traffic was shifted back onto the mainline area. At this location, traffic was reduced from four to two lanes in one direction to provide a safe work area to rehabilitate the roadway.
Caltrans estimated that the as-built rehabilitation activities would take 35 55-hour weekend closures and 410 working days, making this the least expensive alternative in RUC and bottom-line total costs. The actual project was completed in just 32 weekends and 410 weeknights.
Figure 2. As-built lane use plan
Both the contraflow and progressive continuous alternatives would have resulted in high traffic delays and high RUC. These alternatives are more suitable for short-duration projects. Figure 3 shows an illustration of contraflow lane use.
Figure 3. Contraflow lane use plan.
Traditional rehabilitation based on 1,220 nights of temporary lane closures would have caused only modest delays, but given the long project duration would have resulted in high RUC. The final alternative using CSOL was close in value to the as-built alternative, but it would have had higher delays and similar to the contraflow alternative would have closed one half of the interstate, as shown in figure 4.
Figure 4. CSOL lane use plan.
Table 3 compares the CA4PRS-estimated road user costs and time delay plus the cost to the agency for each alternative. CA4PRS calculated the total cost by combining one third of the RUC and the agency cost.
To anticipate traffic delays and impacts, a mesoscopic (between macroscopic and microscopic) traffic model of the local arterial network was created and analyzed using Dynameq software.
Normally, a network large enough to accommodate projects of this size would require a labor-intensive model and time for each model run. Consequently, analyzing several scenarios would be very time-consuming and expensive. Often, funds are not available in a project budget to perform the complex modeling that might be necessary to fully analyze the impacts of a project on traffic if traditional microsimulation models are the only available option for analysis.
Dynameq enabled planners to evaluate congested network scenarios with dynamic equilibrium benchmarks, a time-varying version of the same well-understood equilibrium assignments that have provided consistency for comparison in static analysis for years. Dynameq's equilibrium traffic assignment results represent user optimal network conditions that are immediately useful as an upper-bound on network performance. The few dozen iterations normally required to converge to a dynamic traffic assignment equilibrium took less time than a single assignment by conventional microsimulators.
Dynameq provided a more simplified yet realistic traffic model that was calibrated with fewer parameters. It performed simulations more quickly than microscopic models, allowing more time for analyzing multiple scenarios. This meant that the Dynameq model was more cost-effective to develop and run for a project of this scale. Dynameq helped designers analyze the impacts of the most significant freeway-to-freeway connector closures and adjust the project staging accordingly to minimize impact on the traveling public.
Two traffic studies were performed under this project. The first was a preliminary study and the second involved a more detailed analysis of the results of the first study. Specifically, Dynameq analyzed the detailed construction staging plan and performed an operational analysis of the primary detour routes for six key stages in the construction process when detours were considered critical. The six staging scenarios analyzed are summarized below:
Security Paving Co. Inc. was awarded $51,863,899.55 to complete this project, based on low-bid selection. PCPS was $4.6 million of the total contract. The contractor’s bid for the PCPS elements was a unit price of about $1,500 to 1,574/yd³ ($418/m²), totaling $4.6 million on the entire project. This price included slab fabrication and shipping to the site, existing pavement removal, installation, grouting, and grinding after installation to meet smoothness requirements. The bid item did not include joint sealing, which was covered under a separate bid item
The overall project limits and the PCPS section are shown in figure 5. PCPS was placed in the two outermost lanes and in some cases in only the outermost lane of northbound I-15 for about 1.2 mi between East Jurupa Street and Ontario Mills Parkway. Areas that did not receive PCPS were rehabilitated with traditional continuous lane replacement or random panel replacement. The existing pavement was 9- to 12-in thick concrete over a 5-in cement treated base (CTB). The alignment also included a fair degree of horizontal and vertical curves, as shown in figure 6. The existing outer lane was typically 12 ft wide, but varied to as much as 13 ft near gore areas.
Figure 5. Segment of project that used PCPS for slab replacement. (Source: Google Maps)
Figure 6. Alignment of roadway that received PCPS slab installation, with horizontal and vertical curves included in the alignment.
Caltrans designed the project to allow for a portion of PCPS to be constructed behind temporary concrete barrier, with no closure, which served as the contractor's installation learning curve. Prior to this project, Caltrans had conducted accelerated testing of the Super-Slab® concrete slabs to confirm their structural capacity under heavy load simulation. The results indicated service life beyond the current design requirements.
Project construction began in 2009. As is often the case, the contractor changed several elements of the construction staging plan and was aggressive in accelerating the construction process while also minimizing the number of weekend closures. Caltrans observed that the contractor had identified the demolition operation to be the critical operation in the slab replacement process. These changes included performing random slab replacements during the night work, sometimes paving two lanes wide on connectors, and combining stages.
In all, about 18 closures were used, which fetched the contractor the full incentive offered on the contract. The construction zone also experienced lower traffic delays than anticipated. It is not clear if this reduction was due in part to the slowed economy and public awareness efforts.
The mix design used for the fabrication of the PCPS panels is shown in table 4. Several other segments of the project used concrete mixes designed to achieve pavement opening strengths at various ages—14, 12, 8, and 4 hours. These mix designs were used in different areas and lanes of the project, depending on the opening time criteria for each specific area. Mix designs used for other opening time requirements are summarized in table 4. Note that all aggregate blends met the Caltrans gradation requirements.
Road Safety Audits
An RSA is a formal evaluation of the safety standards of a project and is conducted by an independent, multidisciplinary audit team. The safety performance examination may be performed at any stage of the project, as early as the preconstruction stage (i.e., a future project in its planning and feasibility and design stages), during construction (work zones, preopening stages), or in the postconstruction, in-service stage (existing roads). The goal is to promote safety by identifying issues or project features that can result in unintended and harmful incidents and making improvements that can mitigate the condition.
The analysis typically involves the integration of multimodal safety concerns and the consideration of human factors in the design. An RSA also considers the safety of all road users—passenger cars, pedestrians, pedal cyclists, motorcyclists, and large trucks. In special cases, it may also consider public safety vehicle users (police or fire), maintenance vehicles, older drivers, etc. When RSAs are performed along a specific roadway segment, they also consider the interactions at the project limits by examining connections to existing infrastructure beyond the limits and looking at the segment or intersection from the point of view of users entering and exiting it.
While promoting the awareness of safe design practices, RSAs are a step further than traditional safety reviews. An RSA is essentially a process through which the project team takes the time and makes the effort to identify all project elements as a whole and examine how the various elements interact with each other, especially the combination of minimum standards from each perspective. For example, what are the implications of providing a minimum-radius curve on an approach to an intersection where the minimum stopping sight distance is provided? Can vehicles (especially trucks) safely brake?
Finally, the goal of an RSA is not simply to identify potential problems, but also to identify potential solutions. The RSA audit process often proactively seeks mitigation measures to address these risks. For instance, it may be as simple as setting up a stop sign at a specific project location or additional signs during a construction phase. RSA recommendations might also be more involved. Some questionable elements may be unavoidable in a design, such as when constraints (geometric, fiscal, etc.) limit the project. For example, limited land availability may result in the need to incorporate a horizontal curve with a radius below the minimum design value for anticipated speeds. The RSA can identify potential measures to identify this hazard (appropriate signing) and induce lower approach speeds (narrower lanes or transverse rumble strips), which can be implemented at reasonable expense during construction.
RSA for I-15 HfL Project–Recommendations and Implementation
For the I-15 HfL project, the RSA analysis was conducted at the preconstruction stage and the scope included the entire length of the project from the I-15/Route 60 separation to the Seventh Street undercrossing in Rancho Cucamonga. The description of the RSA evaluation and the recommendations were documented in a report submitted to Caltrans.¹
First, the RSA team noted that the project design had incorporated several features that would greatly enhance safety during construction:
Next, the RSA team analysis involved identifying project elements with safety concerns and assigning them a risk rating on a scale of A through F (lowest risk level through highest risk level). The risk ratings were based on standard combinations of frequency and severity of crashes caused by each safety issue, as shown in table 6.
Using the risk ratings as a basis, the RSA team categorized each identified safety concern and made several recommendations for improving safety, especially during the construction phase of the project. These recommendations are tabulated in table 7. The table also identifies the recommendations Caltrans incorporated into the project as well as reasons for not incorporating recommendations.
The Super-Slab® precast concrete panel system manufactured by the Fort Miller Co., Inc. was used to replace cracked concrete pavement on portions I-15 near the interchange with I-10 in a complex traffic pattern area where ramps and auxiliary lanes merge with mainline lanes. A total of 662 panels were placed in continuous lane rehabilitation and 34 panels were used to replace individual failed existing panels. Figure 7 is a view of the outer lanes of northbound I-15 just before the I-10 overpass and connecting ramps.
Figure 7. View of the outer lanes of northbound I-15 just before the I-10 overpass.
The Super-Slab® system was designed as a reinforced concrete pavement with panels typically fabricated 12 ft wide by 16 ft long. These dimensions varied as required to meet the planned geometry and final grade of the pavement. Panel thickness varied from 9 to 12 in. The required thicknesses, superelevation, and warp were determined for each slab and documented in the plan documents, and each slab was fabricated for delivery to the site in accordance with the contractor’s schedule. A typical detail for a pavement cross section is shown in figure 8.
Figure 8. Typical plan details for Super-Slab®.
A typical panel was cast with load transfer dowels and dowel pockets on opposite ends. Once the panels were placed, the pockets were filled with high-strength grout to create load transfer capability from panel to panel. Panels were placed on a precisely graded subgrade surface, accomplished by using hand-operated grading equipment. Full panel support was assured by injecting bedding grout under the panels through grout distribution grooves cast into the bottom of the panels.
The Super-Slab® system was designed to match the warp of the roadway surface caused by superelevation and/or cross slope. The x, y, and z values of every corner of every panel were computed before fabrication and were used to cast each panel and to level the base surface before installation. This was important because some panels were placed in a superelevation.
Panels were precast about 25 mi from the project at a facility that provided covered casting beds and steam curing. Casting forms were adjusted to account for the warp of each panel. Conventional materials were used in the manufacturing process, such as lifting hardware, epoxy-coated steel reinforcing, and high-early concrete mix made with Type III cement. The panels were leveled with a hand-operated roller screed and then received longitudinal brooming and tining. Figure 9 shows the reinforcing steel workers have arranged in a form before casting. Workers used the roller screed to level the panels, as seen in figure 10. Figure 11 shows the surface texture of a freshly made panel.
Figure 9. Epoxy-coated steel reinforcement is shown in a typical panel layout.
Figure 10. Workers use a roller-type screed to level the panels.
Figure 11. Longitudinally textured surface of a freshly made panel.
Paramount to any rigid pavement as well as PCPS is the reliance on continuous support from the base. To promote full contact with the base, the panels included bedding grout ports and grooves designed to evenly spread the bedding grout under the installed panels. Foam gaskets were attached to the underside of the panels to help contain the bedding grout and prevent the bedding grout from infiltrating the dowel pockets. Once the panels are placed in the field, the dowel pockets are injected with high-strength grout. This grouting operation is separate from the bedding grout operation and involves a different type of grout. Figure 12 shows the underside of a panel. Note the gaskets, dowel pockets and grooves for the bedding grout. Figure 13 shows the dowel bars protruding from the panel and the tie bar pockets at 90 degrees from the dowels.
For this project, tie bars were drilled into the longitudinal side of adjacent existing panels. The dowel bars were securely covered to protect the dowels’ epoxy coating during casting and transport. The finished panels were stored in order of shipping on three-point dunnage to maintain the designed warp of the individual panels. Panels stored flat or on more than three points tend to creep out of their intended profile.
Figure 12. Underside of a panel. Note the gaskets, dowel pockets, and grooves for the bedding grout.
Figure 13. Underside of a panel. Note dowels bar and tie bar pockets.
The existing panels (often cracked at midlength) were sawcut in half longitudinally so the contractor could use the smallest excavator practical to lift the pieces off the roadbed and into dump-bed trucks for removal. This technique limited the amount of base disturbance. Pavement removal is illustrated in figure 14. Removing the existing panels was the critical task governing the amount of production the contractor could achieve during each closure period. On average, the contractor could place 32 panels a night.
Figure 14. Existing panels were removed in manageable pieces.
After the panels were removed, the contractor used a milling machine to level the cement-treated base (CTB) as necessary. A milling machine with minimal distance between the cutting head and the housing is necessary to allow the machine to get as close to the excavated area as possible. Figure 15 shows the milling machine in operation.
Figure 15. A milling machine is used to level the CTB as needed.
A thin layer of bedding sand was placed and leveled with a track-mounted, hand-operated screed. The tracks used by the screed can be seen in Figure 16 as workers push the screed forward. These tracks are set by survey equipment, to ensure precise grading that incorporates panel warp and profile corrections. Workers made three passes, each time progressively leveling the base to the final grade. Grade control of the bedding layer is critical to ensure fully supported panels. Any small pockets under the panels can reliably be filled with the bedding grout.
Figure 16. Workers use a track-mounted screed to level the base.
The Super-Slab® panels were strong enough to allow the crane to sit on the previously placed panel while setting the next panel, as shown in Figure 17. This made for effective use of critical space in the work zone.
Figure 17. A panel is set in place.
Two types of grout were used on this project (see Figure 18). A high-strength grout capable of achieving 2,500 psi in 2 hours or less was used to fill the dowel packets each night the panels were installed. The lane was open to traffic for one day with only the dowels grouted. Shims were sometimes used between the panels until bedding grout was injected under the panels the following night. Finally, the joints were sealed and the panels were diamond ground as needed. The finished PCPS lanes are shown in Figure 19.
Figure 18. Grouting operation.
Figure 19. Finished PCPS.
Installation and Production Rates for Super Slab®
The contractor on the I-15 project achieved remarkable production rates. The innovators of the Super Slab® technology observed that this was the highest production in the field in tight work windows. The production rates by day, tabulated in table 8, indicate that the contractor did not need much time to familiarize the crew with the technology. It appears that good planning and attention to details contributed to the efficiency. An average of 33 slabs was installed per day and 32 slabs per nighttime work window.
The Fort Miller Co., Inc. (owner of the Super-Slab technology) and Caltrans inspectors did note some issues during installation. The base layer grading was the primary concern, and the importance of grading and the provision of full support for the slabs was recognized in the process. Slabs without the specified support were found to eventually crack in the field, and about 25 percent of the slabs cracked after installation. Caltrans found a strong correlation between the days when grading issues were identified and slabs installed on those days cracked. It is likely that other factors contributed as well, such as opening to traffic with no shimming. The concern from a performance standpoint is that thin slabs leave little margin for error.
Cracks were observed in some Super-Slab® panels shortly after the panels were installed and opened to traffic. Cores indicated that the cracks do not penetrate the full depth and the compressive strength of the cores was satisfactory. Petrographic analysis of the core samples also indicate that the material was well consolidated without the risk of durability-related problems.
The majority of this cracking was very tight, and difficult to see with the naked eye. Due to this fact, and the presence of reinforcing steel in the panels, the panels were left in place, treated with methacrylate, and are being monitored for performance.
Figure 20. Cracked panel and core.
1Margaret Gibbs, Road Safety Audit: Rehabilitation of I-15, submitted to Caltrans with Reference H-08172, Opus International Consultants (BC) Ltd., September 2008.