Paula Wallis, Reporter

Fig 1. Cut and cantilever design

Even with a money-saving design, a tunnel option to relieve one of the most congested freeways in Dalles proved too rich for taxpayers.

On February 26th 2009, the Texas Transportation Commission gave preliminary approval to a cut and cantilever design proposed by a team of investors lead by Cintra of Spain for the new LBJ Freeway project on IH-635 between US75 and IH-35E in Dallas (Fig 1).

The $4 billion project is the largest of its kind in the United States and involves rebuilding the existing eight lanes of the LBJ Freeway and adding six managed toll lanes.

"Cost in this case was the overriding factor in the Commission's decision," said Mark Tomlinson, Director of the Texas Turnpike Authority who recommended the cut and cantilever option to the Texas Department of Transportation (TxDOT).

Project map

Cintra's bid was selected as the 'Best Value Proposal' to design, build, finance, operate and maintain the 27.2km (17 miles) of toll roads and rebuild, operate and maintain the existing lanes.

The $2.7 billion PPP package by Cintra and its partners asked for only $445 million in public funding compared to $1.87 billion for the tunnel proposal submitted by the one other short listed competitor, the LBJ Mobility Group lead by ACS and Dragados, also of Spain.

Dragados has won lucrative tunneling contracts in the United States including the East Side Access tunnels and large station caverns in New York City for the Long Island Rail Road connection to Grand Central Terminal, which is valued at $1.15 billion.

But Dragados and its parent company ACS has been runner-up on other major concession projects in the US, most notable the Miami Port Tunnel in Florida, which is stuggling to stay afloat after private funding fell apart in the winning proposal from the French group lead by Bouygues. Dragados also lost out to Cintra on the 13-mile North Tarrant Express corridor in the Dallas/Fort Worth area of Texas that was provisionally awarded last month.

"We are obviously very disappointed by TxDot's decision," said Enrique Fernandez, Head of the Underground Technical Department for Dragados. Our people in the States worked very hard on this bid."

Bernardo Palicio, senior bid leader for Dragados said he too was disappointed by the decision. He said a review of the competing proposals was underway, but he would not second-guess his team's decision to stay with the tunnel option. The team still believes the tunnel option to be the best long-term solution for the region and the better proposal for keeping traffic moving during the estimated five-year construction phase with more lanes open continuously as work progresses.

"The decision was certainly not an indictment against a tunnel," said Tomlinson. "However, the procurement did allow the option of a cut and cantilever design and Cintra presented an excellent program at more than a third of the cost to taxpayers than the tunnel option."

In early feasibility studies a tunnel emerged as the preferred option to relieve congestion along this stretch of highway that currently handles 270,000 vehicles a day, almost double what it was designed for. With no room to expand capacity on the surface and no option to close down lanes during construction of an elevated corridor, a tunnel was the proposed solution.

Fig 2. Cross section of tunnel option

The original design called for two mined tubes of 1.5 miles and 1.9 miles long (2.4km and 3km) long, traveling in-line with the highway above and connecting either end to cut-and-cover transition zones to and from surface routes. The tunnels were to be designed to full 60mph freeway traffic standards and comprise three-lanes in each 60ft (18m) wide profile at road deck level.

The ACS-Dragados team remained committed to the tunnel option, but in a cost saving measure eliminated one of the tubes. Its design split the six managed lanes into three westbound lanes at grade and three eastbound lanes in a tunnel below. The eight main lanes (four in each direction) would be rebuilt on either side of the at-grade managed lanes. The design also included two frontage lanes on each side (Fig 2).

Of the roughly five miles of tunnel in the design, approximately three miles were to have been excavated using the NATM or SEM (sequential excavation method) with two miles of cut-and-cover.

The total cost of the project, however, including development costs, capital expenditures, operating expenses during construction, interest and financing fees during construction and other construction-related costs, came in at almost $3.9 billion with $1.8 billion or 47% coming from taxpayers.

"I don't think anyone would say that the tunnel knocked ACS out of the running," said Tomlinson. "There were no significant flaws in its proposal; the technology was great; its partners were very experienced; and the proposal was sound. In fact its proposal actually scored better in traffic management and disruption limitation during construction. However, TxDOT had a strict cap of $700 million in public funds for this project and Cintra's proposal at $445 million has freed up $255 million for other projects."

Martin Knights President of the International Tunnelling Association (ITA) said with tunnel options bound to be the more costly option, the industry must do a better job of selling the indirect costs of tunnels.

"In these austere times project owners are more focused on the bottom-line construction costs and the indirect costs, such as disruptions during construction and environmental concerns are being sacrificed," said Knights.

Fig 3. Cross section of winning design

Cintra's winning design calls for six managed or toll lanes (three in each direction) depressed in a 25ft deep trench between the rebuilt cantilevered main lanes (four on either side). In addition, the project includes two, and in some places three, continuous frontage lanes on each side for a total of 18 to 20 lanes (Fig 3). Cintra will design, build, maintain and operate the roadway for 52 years.

Patrick Rhode, Vice President of Corporate Affairs for Cintra US, said Cintra had been working on its proposal continuously and in earnest since 2005, allowing for a level of detail and due diligence which may not have been possible in other, shorter term proposals.

"According to the competitive bid process designed by the Owner, both a mined tunnel and a cut and partial, or full cover tunnel were admissible," said Rhode. "We did initial alternative studies where we looked in detail at several options, including tunneling, and based on these initial studies we adopted what we believed was the best solution for the Texas taxpayers, the traveling public, local residents and the region."

Toll collection will be completely electronic with no barriers and will vary throughout the day based on demand.

The project represents an estimated $2.7 billion investment and will include $800 million in government backed loans from the US Department of Transportation. It is the USDoT that is encouraging private sector participation in infrastructure development.

The winning consortium comprises Cintra (55%) and Meridiam Infrastructure (45%) and Negotiations are under way for Dallas Police & Fire Pension System to acquire a stake of approximately 10% from Cintra and Meridiam, although Cintra will keep a participation of at least 51%.

Cintra currently operates more than 1,700 miles of toll roads with a managed investment of more than $24 billion on three continents. It entered the US market eight years ago and has won major tollway projects in Indiana, Illinois and Texas. Last month it was provisionally awarded the Dallas-Fort Worth North Tarrant Express PPP with an estimated investment of $2 billion.

The Spanish were among the early adopters of the PPP model and have make inroads in the USA, which up until recently has been resistant to this financing structure for public works projects.

"The Spanish have been practicing it longer and more effectively than most countries," said Knights. "They are more suave about how they structure their financial models for privately financed infrastructure and they are very good at it."

Construction of the LBJ project is expected to begin between mid 2010 and mid 2011 and be complete by 2016.

The two proposals can be viewed at TxDOT