A report warned the state about the Washington Bridge in 2015. Why was nothing done?

Nearly a decade before the westbound Washington Bridge was hastily closed to avoid collapse, the engineering firm the state hired to study it found structural problems and called for repairs.

But many, and possibly most, of the key repairs recommended by engineering firm AECOM in a 2014-2015 evaluation of the bridge were unaddressed by the time the bridge was closed Dec. 11, 2023.

Traffic mostly flowed across the bridge as usual over the next several years after the AECOM report.

Construction workers and bridge inspectors paused for a few hours one hot day in July 2022 when President Joe Biden's motorcade crossed the span on its way from Brayton Point to T.F. Green International Airport, emails show. There was no evidence they were concerned he was traveling on anything dangerous.

But the problems with the bridge hadn't gone away. They were hidden within the concrete girders and cantilevers undetected until, during painting of the bridge's steel span late last year, someone spotted a snapped tie-down rod and noticed that the whole structure was "bouncing."

Why it matters

A forensic analysis by engineers is due out by the end of the month and was delayed to coordinate with the state’s legal team.

Many questions about what went wrong with the bridge may be answered, but many will likely remain unanswered.

• One is whether the bridge could have been saved if the repairs recommended by AECOM in 2014 had been done immediately or if there had been a weight restriction placed on the bridge.

• Even if the bridge could not have been saved by repairs, would earlier attempts to fix the steel beams have identified the problem sooner and avoided what may have been years of traffic crossing an unsafe bridge?

What did the 2014 analysis say?

A soup-to-nuts evaluation of the bridge conducted by AECOM in 2014 and published in January 2015 found that several parts of the bridge’s structure would need to be strengthened to meet current federal guidelines, and found that the ends of some bridge beams "to be inadequate to resist shear" from stresses delivered by heavy vehicles. 

"Until repairs can be performed and members strengthened, the bridge should continue to be monitored," AECOM stated in the report. The firm found that:

• "In the existing condition ... the structure requires load posting for 11 of the 15 applicable Legal and Permit trucks."

• "Approximately one third of the drop‐in beams will require repairs to eliminate the need for load posting"

• 35 beam-end locations will require "permanent steel jacketing"

• "For both interior and exterior beams the vertical tension tie reinforcement (stirrups) provided at the ends of the beams was found to be inadequate"

• "Several locations" in the tops of the concrete vertical piers that support the superstructure "do not have sufficient strength to meet" the current design standard

The firm estimated the cost of repairs at $7.7 million, and would eliminate the need for load posting – the limiting heavy loads on a bridge to a specific weight.

The repairs were planned with AECOM's help and put out to bid by the Rhode Island Department of Transportation, but a 2016 repair contract was canceled before it was finished. A new bridge repair project started in 2021, but the consortium hired to design and perform that work was just getting to the crumbling concrete beams in question when the highway was shut down and work halted in December.

Between the lines: Reports contradict narrative that there was no warning

Despite the warnings in AECOM's evaluation, and concerns from different engineers in 2009 and 2012, there was seemingly little alarm about the bridge among state officials, according to a Journal review of hundreds of bridge-related documents, including emails, inspection reports, technical analyses, photos, bids, contracts, change orders and lawsuits.

The documents, obtained through multiple Access to Public Records requests and other means, show warnings about the structure going back years that clash with the narrative that the bridge had been fine before a sudden, violent incident damaged it late last year.

They also contradict the claim that no one had suggested the bridge should have a weight limit for heavy trucks.

In fact, contractors working on the bridge when it closed in December had been questioning the DOT's structural assessment of the bridge for more than a year before the emergency.

A closer look: AECOM revised its finding, despite damage

Around three years after finding significant deterioration on the bridge, AECOM engineers re-evaluated the old beams they had raised red flags about and appear to have changed their tune.

AECOM approved a new load rating for the bridge under updated design standards for modern trucks. The "existing [weight limit] posting signs for the structure can be removed," AECOM wrote in 2017.

In the annual inspection reports posted on the DOT's website, the rating for the bridge jumped from 34 tons in 2016 to 52 tons the following year, despite no evidence that major repairs had been completed.

What went wrong with the bridge that triggered its shutdown?

The discovery of broken tie-down roads in December closed the bridge, but the discovery in January of severe problems with the bridge's post-tensioning system turned what would have been a repair project into a demolition job.

Most of the bridge’s spans are made of pre-stressed concrete, bolstered by an internal web of steel rods and cables to add strength.

In "post-tensioning," tendon cables are run through ducts in the concrete, pulled tight and anchored, squeezing the beams into compression to make them more resistant to bending or cracking under heavy loads. The ducts are filled with grout to secure and protect them.

When experts looked closer at the bridge, they found active corrosion of exposed tendon anchors, voids within the ducts, soft grout, corroded tendons and cracked, unsound concrete, according to a February report from VN Engineers.  

Without the post-tensioning system working as intended, the bridge might become unstable under heavy load even if the tie-down rods are fixed, VN said.

Even back in 2012, when Michael Baker Engineering evaluated the bridge, inspectors found four "exposed and broken prestressing strands" at the end of one of the beams.

Behind the scenes: The repair saga

To give the bridge another 25 years of use, AECOM recommended repairing damaged concrete at the ends of the beams and then wrapping them, and the tops of the piers they sit on, in Fiber Reinforced Polymer.

Damage to the beam ends was happening "where existing deck joints have leaked," and AECOM proposed to eliminate up to two-thirds of the deck joints on the bridge.

Zoom out: Political jockeying around bridge repair

The AECOM analysis was performed in the waning days of Gov. Lincoln Chafee's administration and published just as Gina Raimondo was sworn in as Rhode Island governor.

Peter Alviti Jr. was quickly tapped to overhaul the Department of Transportation and embark on a massive statewide bridge reconstruction campaign.

Using the evaluation, the DOT went out to bid in January 2016, expecting the project to cost $17.2 million, $10 million of that construction and the rest design and soft costs.

By October that jumped to an estimated $23.5 million, on $16.3 million construction costs.

And in January, 2017 the state entered into a $14,697,281 contract with Cardi Corp. The work included lifting up and repairing bridge beam ends, repairs to structural concrete masonry and fiber reinforced wraps of beams and pier caps.

At the same time, AECOM re-evaluated the bridge’s load rating, despite computer modeling still showing the bridge’s strength inadequate. A "live load test" found it passed statutory muster.

AECOM referred all Journal questions on the Washington Bridge to the DOT.

The extent of the work Cardi completed is difficult to pin down and was the subject of a 2021 lawsuit by joint venture of Barletta Heavy Division and Aetna, who bid against Cardi to finish the job.

The reasons why Cardi never finished the initial bridge contract are also murky.

'Kinda sketchy': Barletta claims 'industrial espionage' in contaminated soil case. Here's the lawsuit.

Another force in play: Political points to be scored over 'The Gina Jam'

AECOM also developed a traffic plan to keep cars moving across the Seekonk River, recommending closing the Taunton Avenue on-ramp in East Providence and the Gano Street off-ramp in Providence.

People on both sides of the river balked at the congestion those closings would cause in their neighborhoods and the DOT changed plans.

The ramps stayed open in the summer of 2018 when travel lanes on the bridge were reduced to make space for workers and the resulting vehicle weaving induced paralyzing backups on I-195.

The traffic was dubbed “The Gina Jam" by Cranston Mayor Allan Fung, Raimondo's GOP opponent for governor in the 2018 election, and within several days the traffic pattern was scrapped and work disrupted.

A little less than a year after "The Gina Jam," on July 18, 2019, the DOT told Cardi it would terminate the Washington Bridge contract, at no fault to Cardi, with the contractor responsible for finishing a punch list of work by June 13, 2020.

When the bridge closed in December 2023, Alviti told reporters the traffic jam kept the bridge work from being completed, but also said Cardi was "not performing that contract in accordance with the agreement that we had with them."

The termination agreement written by the DOT blames the "temporary traffic control" arrangement for causing "unacceptable levels of congestion during the rush hour periods."

The price of Cardi developing an alternate traffic plan was also deemed “unacceptable.”

Meanwhile: The bridge’s condition worsens

The termination also stated that the repairs Cardi was undertaking would be “insufficient.”

In other words, the bridge was in bad enough shape in 2018 and 2019 that the repairs engineers had outlined five years earlier – when they found it structurally inadequate – would not be enough to fix it.

A 2020 inspection by AECOM found, unsurprisingly, that the unaddressed problems in the bridge had only gotten worse.

“A number of new hollow areas and spalls were found during this inspection," Corey Richard, Northeast transportation operations manager for AECOM, wrote in a 2020 email.

The DOT agreed to pay Cardi $13.7 million of the $14.7-million contract for the unfinished bridge job and went out to bid again on a much larger project to complete the work.

Behind the scenes: Cardi vs. Barletta

The Washington Bridge "Phase 2" project would be done as a design-build contract, where a single bidder handles most of the design and engineering as well as construction.

The project also included:

• Widening the bridge

• Adding a new Waterfront Drive off-ramp in East Providence

• Fixing the Gano Street ramps in Providence.

Of the three bidders on the Phase 2 contract, Cardi was the low bidder and selected by the DOT.

But Barletta-Aetna challenged the award and then the Federal Highway Administration, which was picking up most of the tab for the bridge work, stepped in. The FHWA said Cardi's bid didn't meet the requirements of the RFP and refused to fund it. After more legal wrangling, the job went to Barletta-Aetna.

Bridge concerns come to light

In the spring of 2022, the engineers working under the Barletta-Aetna venture became concerned by what they were seeing now that they finally had access to the bridge.

Their bid was based on the evaluations done years earlier by AECOM, and AECOM was also the owner's representative on the project, advising the state on technical questions beyond the capacity of in-house engineering staff.

VHB, which was doing engineering for Barletta-Aetna, hired VN Engineers as a subcontractor to study the bridge's condition.

• In May 2022, VN raised concerns with the bridge's structural rating to AECOM

• In Feb. 2023, VN made a presentation to the DOT arguing the bridge needed much more strengthening than originally assumed in the bidding process – at a higher cost

VN calculated that it would take much more Fiber Reinforced Polymer – the same repair material AECOM had prescribed in 2014-2015 – to achieve the structural capacity needed to avoid a weight limit on the bridge.

It recommended double the amount of Fiber Reinforced Polymer for vertical strengthening and 4.5 times the amount for horizontal strength.

The VN presentation said the 2017 AECOM load rating that gave the bridge a clean bill of health and said weight limits could be removed had "significantly overestimated" the strength provided by the post-tensioning system.

State says it won’t pay for extra repair work

The DOT would not approve the changes. If Barletta-Aetna thought it needed to add more fiber to the bridge, it would have to cover the cost, estimated at around $4.5 million according to a Barletta spokesperson.

The agency and its contractor would argue about who would pick up the tab for the proposed change through much of last year, according to emails obtained by the Journal.

On Oct. 3, about two months before the emergency shutdown, Anthony Pompei, DOT project manager for the Washington Bridge, told representatives of Barletta, VHB and AECOM that the state "disagree(s) any of the efforts required to design and install the necessary strengthening would be considered extra work" the state would pay for.

Regardless of who was on the hook for the extra work, Pompei's response does not dispute the contractor's argument that the 2016 AECOM analysis was flawed or that the bridge was not as strong as it was supposed to be.

What to watch: DOT refuses to answer questions about previous concerns

At a General Assembly Joint Oversight Committee hearing at the State House in February, Alviti said he didn't "believe we have ever had to post a load rating on the Washington Bridge."

Rep. Joseph Solomon Jr., a Warwick Democrat, asked: Was a load rating ever recommended?

"I have asked my engineers that," Alviti said. "They are not aware of any recommendations that came to them for posting the bridge to less than what its capacity is."

Sitting next to Alviti was Jeff Klein, director of structural engineering at VHB, part of the team that had been arguing for months that the AECOM load rating was bad.

He acknowledged that "our subconsultant VN" had done an analysis that was "getting different results" from the existing load rating, but said he could "not speak to the details of it."

"I don't feel like I got a satisfactory answer on that," Solomon said last Wednesday about whether the bridge should have been under a weight limit before it was suddenly deemed a collapse risk.

In the weeks since The Journal first asked about the history of structural concerns about the Washington Bridge – including AECOM's 2014-2015 concerns, the 2017 load rating upgrade and last year's VN presentation on – the DOT has refused to answer any questions about them.

In an April 5 news conference announcing the hiring of a legal team to recover money from those involved in mistakes made on the Washington Bridge, Gov. Dan McKee denied there would be any "gag order" on information about the bridge, including at the DOT, while the litigation campaign ramps up.

"No. I'm going to be out in front of this mic multiple times and you're going to ask questions and I'm going to answer," McKee said.

A few hours later that day, DOT spokesman Charles St. Martin responded to questions about the AECOM reports with this email: "We are awaiting the results of the forensic review. Please know the forensic team has been provided all the relevant documents including information relative to your inquiry."

The DOT not only declined to explain how the bridge's structural inadequacies went away, but whether the weight-limit signs AECOM said could be taken off the bridge ever existed.

This article originally appeared on The Providence Journal: Report warned of Washington Bridge's problems years before closure