Economic Analysis

A key aspect of HfL demonstration projects is quantifying, as much as possible, the value of the innovations deployed. This entails comparing the benefits and costs associated with the innovative project delivery approach adopted on an HfL project with those from a more traditional delivery approach on a project of similar size and scope. The latter type of project is referred to as a baseline case and is an important component of the economic analysis.

For this economic analysis, Mn/DOT supplied most of the cost figures for the as-built project. The assumptions for the baseline case were determined from discussions with Mn/DOT and pertinent national literature.

Construction Time

Full closure was in effect for 4 months (122 days). All four lanes of TH 36 were open to traffic in 6.5 months (200 days). Traditional staged construction methods would have impacted traffic for an estimated 19 months (580 days). Therefore, the resulting impact to those who rely on the highway was shortened by 380 days, affecting them less than half the time required by traditional construction methods. The full closure allowed the contractor unfettered access to the roadway to haul materials, begin the overhead bridgework, and conduct paving operations without the interruptions caused by contending with live traffic. The locked incentive date encouraged the contractor to open the roadway no later than scheduled, while lane rental provided the flexibility to temporarily close a lane when needed to address finishing and cleanup activities.

Construction Costs

Table 3 presents the differences in construction costs between the baseline and the as-built alternatives. All of the cost estimates and construction assumptions were provided by Mn/DOT and stem from a technical meeting of the Mn/DOT construction and estimating staff held to examine different approaches to construct the project. In the baseline scenario, the project would have been constructed with traditional methods in which traffic on TH 36 would have been maintained on a four-lane bypass built next to the existing highway. The as-built case used shorter bypass lanes to accommodate traffic crossing the median at the ends of the project and short access lanes to McKnight Street. Bypass costs for the baseline scenario are based on the as-built bid items and increased by 15 percent to account for inefficiencies inherent in working around live traffic. Traffic control also would have been a major contributor to the baseline costs, resulting from additional items such as barriers, delineators, and temporary lighting.

Mn/DOT staff concluded that special provisions in the contract for contract time ($2,925,000) and lane rental ($220,000) would not apply to the baseline case and should be eliminated from the winning bid for a more accurate cost comparison. The comparison in table 3 assumes that both alternatives would have the same end design and the major differences between the two options are from the bypasses and traffic control. Building the project under traffic would have resulted in $3,884,493 ($32,097,055 - $28,212,562) in additional construction costs, or an extra 14 percent, when compared with the as-built scenario.

Traffic Impacts During TH 36 Full Roadway Closure

Overview of Anticipated Detour Routes

The full closure of TH 36 during May and June 2007 between White Bear Avenue and Century Avenue (TH 12) required diversion of between 39,400 and 47,500 vehicles per day. Project planners identified through and local detours in both the eastbound and westbound directions of travel. The through detour westbound began at the interchange between I-694 and TH 36, followed I-694 southbound to I-94, continued on I-94 westbound until reaching I-35E, and then traveled north on I-35E until it reconnected with TH 36. For the eastbound through detour, planners assumed that drivers would exit TH 36 at Maplewood Drive (TH-61), travel north to I-694, and follow I-694 eastbound until reconnecting with TH 36. This detour was identified because project planners assumed that motorists would want to avoid ongoing construction on I-35E and I-694 north of TH 36.

For drivers with local origins and destinations in the vicinity of the full roadway closure, project planners assumed that County Highway C to the north of TH 36 and County Road B to the south would serve as the primary detour routes. Drivers were expected to access those routes using White Bear Avenue or Century Avenue, depending on whether they were traveling from west to east or east to west. 

Traffic Impacts Examined

Baseline Scenario

Mn/DOT’s project staff projected that the traffic impacts of a more conventional or baseline approach could be divided into three distinct stages. The first stage includes the construction of two temporary bypass lanes in each direction along the southern limits of TH 36. The duration of this stage (stage 1) was estimated to extend from April 15 2007 to July 15 2007 and have the following impacts:

• TH 36 within the project limits had signalized intersections at McKnight Road, First Street, and Margaret Street. On the first day of building the bypass, Margaret Street and First Street would have been closed while McKnight Road would have stayed open.
• Periodic off peak lane closures would occur on McKnight Road.
• Eastbound TH 36 would have required off peak lane closures to facilitate bypass construction.  These off peak closures would begin at 7 a.m. and end at 2:30 p.m.

The second stage (stage 2) would consist of constructing the permanent westbound roadway over the course of a full year (July 15 2007 to July 15 2008). The traffic impacts would have been the following:

• Margaret Street and First Street would have remained closed and traffic would have been detoured onto McKnight Road.
• Periodic off peak lane closures would have continued on McKnight Road.
• Traffic on TH 36 would have been routed onto the four temporary bypass lanes.
• Off peak lane closures would have likely occurred on the westbound bypass lanes to accommodate the construction of the new westbound TH 36 permanent roadway.

Construction of the permanent eastbound TH 36 roadway would have taken place in stage 3 from July 15 2008 to June 15 2009 with the following traffic impacts:

• Margaret Street and First Street would have remained closed and traffic would have been detoured to McKnight Road.
• Periodic offpeak lane closures would have continued on McKnight Road.
• TH 36 traffic would have been switched from the bypass to a single lane in each direction on the new westbound roadway.
• No temporary or off peak closures would have been expected because there would have been only one lane of traffic in each direction.

Thus the projected three stages of construction would have extended between April 2007 and July 2009 for a total of 26 months.  Such construction would have necessitated construction with live traffic in close proximity which would have offered tremendous logistical challenges to construction operations considering the volumes of earthwork that was moved on the project particularly at certain locations where deep cuts were made and 20+ ft retaining walls were built to allow for a facility with no at-grade intersections.  The safety considerations for such construction would also have been substantial. 

As-Built Scenario

Analysis of the impacts of the full closure on traffic was accomplished using traffic volume data collected on the surrounding freeway segments by the Minneapolis-St. Paul Transportation Management Center. On the arterial streets assumed to be the primary detour routes for local drivers, a number of portable traffic counters were installed at key locations to track diversion during the closure. Analyses were limited to evaluating changes in traffic volumes during the morning and evening travel periods; no speed or travel time data were gathered during the project. Volume-to-capacity computations on the detour routes identified few locations where congestion may have occasionally developed, so it was assumed that diverted traffic had a minimal effect on detour route speeds and travel times. Consequently, only the travel impacts on diverted traffic (the additional travel distances and differences in travel speeds on the detour routes compared with TH 36) had to be considered in the analysis. That is, traffic already using the various alternative routes was not significantly affected by the addition of the diverted traffic on those routes. 

Summary of Results

As Built Scenario—Changes in Traffic Volumes

Overall, for the as-built scenario, each of the identified detour routes experienced some increase in traffic volumes during the roadway closure, indicating that they were being used by diverted drivers. Table 4 summarizes the change in traffic counts on the identified detour routes during the 6 a.m. to 10 a.m. and 2 p.m. to 9 p.m. analysis periods. Also shown is the estimate of traffic normally on TH 36 during those same time periods between McKnight Road and White Bear Avenue, assuming that 20.5 percent of the AADT occurs during the 6 a.m. to 10 a.m. period and 42.2 percent occurs during the 2 p.m. to 9 p.m. period (these assumptions were made from a sample of automatic traffic recorder count data in four States—California, North Carolina, Ohio, and Washington—obtained through the FHWA Highway Safety Information System as part of a National Cooperative Highway Research Program study).

In the morning period, nearly all of the TH 36 traffic that was forced to divert by the full roadway closure can be accounted for on the four major facilities serving as alternative routes.   The difference (40 vehicles) is only 0.4 percent of the traffic normally using TH 36. In the afternoon period, however, a more significant proportion (28.3 percent) of normal TH 36 traffic could not be accounted for on the four major alternative routes. As has been found in other evaluations of travel pattern changes during major highway projects, most of this difference may be the result of drivers searching out their own alternative route to TH 36 in the afternoon rather than using those designated as part of the traffic management plan for the project.

Additional Road User Costs

As stated above, the results of the traffic volume analysis implied that the addition of traffic on the detour routes did not have a significant effect on speeds or travel times during the total roadway closures. Therefore, road user costs caused by the closure were the result of the traffic normally using TH 36 having to travel an additional distance and time each day when the closure was in place. Project planners performed a simple travel time analysis to assess the additional travel distance and time of the four alternative routes (eastbound through, eastbound local, westbound through, and westbound local). In reality, the traffic volume data indicated that drivers did not follow the prescribed detours by direction. Instead, both directions of travel on each facility designated as a detour route saw increases occur in both the morning and afternoon-evening periods. Consequently, the changes in travel times for each alternative route could be applied to the total amount of additional traffic on each facility listed in table 5, which summarizes the additional travel times and distances associated with each route. The portion of the afternoon traffic that could not be accounted for on the four detour facilities was distributed in proportion to the traffic increases measured on those facilities, as it was assumed that drivers finding their own alternative route experienced similar additions to their trip times and distances as measured for the defined alternatives.

Project planners indicated that 2.9 percent of the TH 36 traffic consisted of commercial vehicles and used the following values of time and vehicle operating costs for commercial vehicles and automobiles:

• Values of Time
  • Automobile  = $12.63/h
  • Commercial Vehicle = $20.41/h
• Vehicle Operating Costs
  • Automobile = $0.27/mi
  • Commercial Vehicle = $0.75/mi

Applying these unit costs to the values in table 5 results in the following estimated additional road user costs per day of closure:

1,608.4*(.029*$20.41 + 0.971*$12.63) + 100,710*(.029*$0.75 + 0.971*$0.27)    
= $20,677+$28,594
= $49,271/day

The road user cost from employing the full lane closure was $6,011,062 (122 days*$49,271/day).

The road user costs associated with traditional staged construction were not computed but considering that the construction would have been spread over a time span that was at least 6 time greater than the as built scenario (780 days for the traditional construction versus 122 days for the as built), the road user costs in terms of delay cost and safety costs would have been very high.  Mn/DOT’s assessment was that the neglible vehicle operating costs due to temporary by passes for the traditional construction would have been more than easily offset by the increased delay and safety costs.  However, these higher costs were not quantified due to the unavailability of reliable user delay information.

Cost Summary

Traditional construction methods would have cost $3,884,493 more to complete this project compared to the as-built case. The impact to motorists in terms of vehicle operating costs added $6,011,062 in road user costs to the as-built case.  However, according to Mn/DOT, the additional road user cost would have in terms of increases in delay and safety costs for the traditional staged construction would have made this method of construction cost-prohibitive.  Moreover, this option presented many logistical challenge for construction and motorist safety.  As a result, the as built scenario was the only robust and viable option available for the construction of TH 36.