San Francisco-Oakland Bay Bridge Research Articles

Optimizing Mixed Autonomy Traffic Flow with Decentralized Autonomous Vehicles and Multi-Agent Reinforcement Learning

We study the ability of autonomous vehicles to improve the throughput of a bottleneck using a fully decentralized control scheme in a mixed autonomy setting. We consider the problem of improving the throughput of a scaled model of the San Francisco–Oakland Bay Bridge: a two-stage bottleneck where four lanes reduce to two and then reduce to one. Although there is extensive work examining variants of bottleneck control in a centralized setting, there is less study of the challenging multi-agent setting where the large number of interacting AVs leads to significant optimization difficulties for reinforcement learning methods. We apply multi-agent reinforcement algorithms to this problem and demonstrate that significant improvements in bottleneck throughput, from 20% at a 5% penetration rate to 33% at a 40% penetration rate, can be achieved. We compare our results to a hand-designed feedback controller and demonstrate that our results sharply outperform the feedback controller despite extensive tuning. Additionally, we demonstrate that the RL-based controllers adopt a robust strategy that works across penetration rates whereas the feedback controllers degrade immediately upon penetration rate variation. We investigate the feasibility of both action and observation decentralization and demonstrate that effective strategies are possible using purely local sensing. Finally, we open-source our code at https://github.com/eugenevinitsky/decentralized_bottlenecks .

The San Francisco–Oakland Bay Bridge, CA, USA – eastern span

With daily traffic of 260 000 vehicles, the San Francisco–Oakland Bay Bridge is a major connection in the San Francisco Bay area (CA, USA). The new bridge is a designated lifeline structure (to remain open for emergency traffic after a major seismic event) with a design life of 150 years. The new bridge is 3.6 km long and consists of four distinct structures: a low-rise post-tensioned concrete box girder near the Oakland shore; a 2.4 km long segmental concrete box girder (skyway); a first-of-its-kind self-anchored suspension (SAS) bridge with a 385 m main span over the navigational channel; and a post-tensioned concrete box girder that connects to the east portal of the Yerba Buena Island tunnel. Opened in 2013, the signature span of the bridge is the SAS bridge with a length of 624 m and a total deck width of 79 m accommodating ten lanes of traffic in addition to a bike/pedestrian path. The US$6.4 billion megaproject was procured under multiple contracts and was delivered using the traditional design–bid–build method. This paper describes the key design innovations and construction methods which address the unique challenges on this project.

Underwater implosion of re-enforced concrete piers

From 2015 through 2018, 18 reinforced concrete piers formally supporting the east span of the San Francisco-Oakland Bay Bridge were removed using underwater demolition. The events ranged from one pier to up three piers at one time with the number of charges ranging from 92 to 688 and weights from 12 to 35 pounds. In all cases, a bubble curtain, blast attenuation system was used to reduce the blast pressures. Although the blast plans varied, more variation in the monitored peak sound pressure levels and sound exposure levels was observed than was expected from initial predictions. Potential causes of this variability were examined including varying charge confinement, directionality due to self-shielding of the structures as the individual blasts progressed, the effect of varying distance to the monitoring location for the individual charges, obscuration of the monitoring location due to intervening structures, and varying effectiveness of the blast attenuation system. Each of these potential causes are examined in this presentation.

Dynamic Displacement Estimation for Long-Span Bridges Using Acceleration and Heuristically Enhanced Displacement Measurements of Real-Time Kinematic Global Navigation System.

In this paper, we propose a dynamic displacement estimation method for large-scale civil infrastructures based on a two-stage Kalman filter and modified heuristic drift reduction method. When measuring displacement at large-scale infrastructures, a non-contact displacement sensor is placed on a limited number of spots such as foundations of the structures, and the sensor must have a very long measurement distance (typically longer than 100 m). RTK-GNSS, therefore, has been widely used in displacement measurement on civil infrastructures. However, RTK-GNSS has a low sampling frequency of 10–20 Hz and often suffers from its low stability due to the number of satellites and the surrounding environment. The proposed method combines data from an RTK-GNSS receiver and an accelerometer to estimate the dynamic displacement of the structure with higher precision and accuracy than those of RTK-GNSS and 100 Hz sampling frequency. In the proposed method, a heuristic drift reduction method estimates displacement with better accuracy employing a low-pass-filtered acceleration measurement by an accelerometer and a displacement measurement by an RTK-GNSS receiver. Then, the displacement estimated by the heuristic drift reduction method, the velocity measured by a single GNSS receiver, and the acceleration measured by the accelerometer are combined in a two-stage Kalman filter to estimate the dynamic displacement. The effectiveness of the proposed dynamic displacement estimation method was validated through three field application tests at Yeongjong Grand Bridge in Korea, San Francisco–Oakland Bay Bridge in California, and Qingfeng Bridge in China. In the field tests, the root-mean-square error of RTK-GNSS displacement measurement reduces by 55–78 percent after applying the proposed method.

Virtual Reality Helps Secure Public Approval

A design team used virtual reality to generate public approval of a proposed bike and pedestrian path for the west span of the San Francisco-Oakland Bay Bridge in California.

Challenges of civil engineering megaprojects in North America

In many respects, megaprojects define the civil engineering profession, going back to the Panama Canal and the Hoover dam. They present many challenges. By definition, they are large, innovative projects that have never been done before. The Boston Central Artery/Tunnel and the San Francisco–Oakland Bay Bridge are now in service and by all accounts are performing well. However, these both had large cost overruns and concerns about quality and lengthy construction delays. The Central Artery/Tunnel also suffered a fatal ceiling collapse. Megaprojects require careful management. Owners and builders must be transparent and responsive to open record requests. Public employees should be empowered to bring forward concerns about quality and safety issues. The agencies should establish reliable open channels of communication with the public. Cost estimates should consider risk and provide for both best and worst cases. Independent inspection agencies should be established to provide oversight of large projects.

Environmental Hydrogen Embrittlement of G41400 and G43400 Steel Bolting in Atmospheric Versus Immersion Services

This article is based on a review of data in the literature and those released by the California Department of Transportation and reported by the Bureau of Safety and Environmental Enforcement (BSEE). In 2013, hot dip galvanized (HDG) threaded rods failed due to environmental hydrogen embrittlement (EHE) cracking on the San Francisco–Oakland Bay Bridge. These rods were made of G41400 steel, conforming to ASTM A354 Grade BD (39 HRC max). All 32 EHE failures occurred in the bottom threads that had been immersed in pools of water, inadvertently, under no stress, for over four years before they were pretensioned. Several failure analysis reports stated that they failed due to EHE cracking because they were pretensioned while being immersed in water. This is erroneous, however, because one or more of the failed rods were not immersed in water during pretensioning and afterward. Conversely, hydrogen from metal corrosion of the HDG BD rods can diffuse into them under no stress while corroding in standing water for several years. The EHE cracking occurred when subsequently pretensioned. None of about 4000 threaded ends of HDG BD rods that have been exposed only to marine atmosphere (but not immersed in water) has failed in three to four years in service under a pretension load. These data are compared with recent EHE cracking of G43400 bolting in subsea oil production facilities in the Gulf of Mexico, reported by the BSEE.

Earthquake Resilience of Reinforced Concrete Structural Walls with Replaceable “Fuses”

ABSTRACTThe concept of providing a fuse in a structural system has been both developed and investigated over the past few years (e.g., the use of replaceable links in eccentrically braced frames or the use of replaceable links in the new San Francisco-Oakland Bay Bridge). This paper presents a new coupling beam with a creative fuse installed in the mid-span of a steel-concrete composite beam via an end plate and high-strength bolts. First, a practical design methodology of a replaceable coupling beam is presented. Next, the seismic behavior of the proposed fuse is tested and evaluated, and two structural wall specimens with or without a fuse are designed and fabricated according to the proposed design methods. The two specimens possess similar shear capacities under large-scale cyclic loading, and the walls of the two specimens show similar failure modes; however, the new walls exhibit slightly lower levels of damage than the conventional walls. In particular, the inelastic deformation and damage of replaceable coupling beams are mainly concentrated in the fuse, whereas the non-yield segment and the beam-wall pier interfaces remain nearly intact and produce only slight damage, which is beneficial regarding the replacement of the fuse in post-earthquake events.

Prediction and Monitoring of Underwater Sound Levels from the Implosion of a Reinforced Concrete Bridge Pier

After the completion of the new east span of the San Francisco–Oakland Bay Bridge in California, large concrete piers of the old span needed to be demolished. To consider using controlled blasting for this action, hydroacoustic levels were predicted and monitored with regard to specified fish and marine mammal criteria. The metrics included peak pressure and sound exposure levels at distances from 25 to more than 4,000 ft from the pier. For peak pressure, the measured levels were slightly higher than estimated, although for sound exposure levels, the measured levels were somewhat lower than estimated because of the effect of surface reflection. A blast attenuation system consisting of a wide bubble stream was used to minimize the hydroacoustic levels in the water surrounding the pier. The implosion event consisted of 588 individual charge detonations ranging from 35 to 21 lb (15.9 to 9.5 kg/delay) spaced 9 ms apart. Although there were 135 individual detonations of the larger 35-lb charges, the highest peak pressures varied in level by 10 to 15 dB during the course of the implosion. The methods for predicting the levels, measured results, analysis of data, and performance of the blast attenuation system are reviewed.

Crossing the bridge: The effects of time-varying tolls on curbing congestion

This paper estimates the traffic volume and travel time effects of the road congestion pricing implemented on the San Francisco-Oakland Bay Bridge. I employ both difference-in-differences and regression discontinuity approaches to analyze previously unexploited data for the two years spanning the price change and obtain causal estimates of the hourly average treatment effects of the policy. I find evidence of peak spreading in traffic volume and decreases in travel time during peak hours. I also find suggestive evidence of substitution to a nearby bridge and decreases in travel time variability. In addition, I calculate own- and cross-price elasticities.

Modeling Magnetic Fields from a DC Power Cable Buried Beneath San Francisco Bay Based on Empirical Measurements.

The Trans Bay Cable (TBC) is a ±200-kilovolt (kV), 400 MW 85-km long High Voltage Direct Current (DC) buried transmission line linking Pittsburg, CA with San Francisco, CA (SF) beneath the San Francisco Estuary. The TBC runs parallel to the migratory route of various marine species, including green sturgeon, Chinook salmon, and steelhead trout. In July and August 2014, an extensive series of magnetic field measurements were taken using a pair of submerged Geometrics magnetometers towed behind a survey vessel in four locations in the San Francisco estuary along profiles that cross the cable’s path; these included the San Francisco-Oakland Bay Bridge (BB), the Richmond-San Rafael Bridge (RSR), the Benicia-Martinez Bridge (Ben) and an area in San Pablo Bay (SP) in which a bridge is not present. In this paper, we apply basic formulas that ideally describe the magnetic field from a DC cable summed vectorially with the background geomagnetic field (in the absence of other sources that would perturb the ambient field) to derive characteristics of the cable that are otherwise not immediately observable. Magnetic field profiles from measurements taken along 170 survey lines were inspected visually for evidence of a distinct pattern representing the presence of the cable. Many profiles were dominated by field distortions unrelated to the cable caused by bridge structures or other submerged objects, and the cable’s contribution to the field was not detectable. BB, with 40 of the survey lines, did not yield usable data for these reasons. The unrelated anomalies could be up to 100 times greater than those from the cable. In total, discernible magnetic field profiles measured from 76 survey lines were regressed against the equations, representing eight days of measurement. The modeled field anomalies due to the cable (the difference between the maximum and minimum field along the survey line at the cable crossing) were virtually identical to the measured values. The modeling yielded a pooled cable depth below the bay floor of 2.06 m (±1.46 std dev), and estimated the angle to the horizontal of the imaginary line connecting the cross-sectional center of the cable’s two conductors (0.1143 m apart) as 178.9° ±61.9° (std dev) for Ben, 78.6°±37.0° (std dev) for RSR, and 139.9°±27.4° (std dev) for SP. The mean of the eight daily average currents derived from the regressions was 986 ±185 amperes (A) (std dev), as compared to 722 ±95 A (std dev) provided by Trans Bay Cable LLC. Overall, the regressions based on fundamental principles (Biot Savart law) and the vectorial summation of cable and geomagnetic fields provide estimates of cable characteristics consistent with plausible expectations.

Observed and simulated traffic impacts from the 2013 Bay Area Rapid Transit strike

ABSTRACTDespite high costs, many cities build public transit to address regional equity, environmental and economic goals. Although public transit accounts for a minority of trips (∼5%), the impact is widely felt when service is suspended during a strike through excess road demand and slower journeys. In 2013, Bay Area Rapid Transit (BART) workers participated in two brief strikes, and the resulting traffic conditions illustrate the value of transit to drivers in the San Francisco Bay Area region. This paper tests the impact of rail transit service interruption on freeway traffic conditions using volumes and travel times. During the strike, regional freeway conditions showed negligible change. However, on facilities that parallel BART service, the impacts are as bad as the worst day of a typical week. Conditions on the San Francisco–Oakland Bay Bridge showed significant impacts with travel times and volumes nearly doubling the baseline median values on the worst day.

ASTM F1624 rising step load (RSLtrademark) testing for hydrogen embrittlement threshold of threaded cut outs of A354BD rods

After tensioning high-strength steel rods installed on the lower housing of Shear Key S1 and Shear Key S2 at Pier E2 of the San Francisco-Oakland Bay Bridge, 32 of 96 threaded rods failed. The rods were tensioned to 0.7 Fu and the failures were delayed failure, from 3 days to 14 days after tensioning. Failure analysis found that failure was due to Hydrogen Embrittlement. A six part test program performed to evaluate safety of additional rods in the field. This paper reports on Test V and Test VI – Small Threaded Specimen Testing. Small (Charpy-sized) threaded specimens were used to predict the threshold Load for the onset of hydrogen induced stress corrosion cracking of large A354BD Rods. The procedure used was to measure the hydrogen embrittlement threshold load for a small specimen and use a validated model to calculate the corresponding threshold stress intensity factor from the threshold load. With this threshold stress intensity factor, a model was used to calculate threshold load of the full size rod. Three independent methods were used to validate the accuracy of using small specimens to predict full-size failures. 1) Use finite element analysis to relate the local conditions at the thread root for a sub-size specimen to the thread root for a full-size threaded rod. 2) Correlate stress intensity factor between sub-size specimens to full-size threaded rods. 3) Mechanical testing of full-size rods in tension. It was found that the analysis methods agreed with each other, thus validating the use of data from sub-size specimens to predict the failure performance of full-size threaded rods. The data generated from Test V has the additional advantage of being able to quantitatively determine the effects of variations in the applied potential from compositional differences in the galvanizing and the variation in the hardness of the rods on the threshold stress.

Dynamic Adaptive Approach to Transportation-Infrastructure Planning for Climate Change: San-Francisco-Bay-Area Case Study

Adaptation of existing infrastructure is a response to climate change that can ensure a viable, safe, and robust transportation network. However, deep uncertainties associated with climate change pose significant challenges to adaptation planning. Specifically, current transportation planning methods are ill-equipped to address deep uncertainties, as they rely on designing responses to a few predicted futures, none of which will occur exactly as envisioned. In this paper, we propose using dynamic adaptive planning (DAP), an emerging general strategic planning method, to account for deep uncertainties by building flexibility and learning mechanisms into plans that enable continuous adaptation throughout implementation. This paper first reviews uncertainty in general, introduces what is meant by deep uncertainty, and then introduces DAP. Then, DAP is applied to a case study of the Oakland approach to the San Francisco-Oakland Bay Bridge, which was initially assessed under the 2010–2011 FHWA Climate Change Vulnerability Assessment Pilot program, to illustrate how DAP could be applied as a response to climate change in the context of evolving transportation infrastructure adaptation planning practices in the United States. We conclude that DAP is well suited to account for the deep uncertainties of climate change in transportation and infrastructure planning, and provide suggestions for further research to better apply DAP in this field.

Bay Bridge anchorage construction: George Booth Post.

Bay Bridge anchorage construction: George Booth Post.

Empirical assessment of bottleneck congestion with a constant and peak toll: San Francisco–Oakland Bay Bridge

Extensive theoretical research has been conducted on the morning commute problem building on the bottleneck model (Vickrey in Am Econ Rev 59(2):251–260, 1969 ). There have been far fewer studies assessing the bottleneck model and its extensions using real data. This paper presents an analysis of empirical data from the toll plaza of the San Francisco–Oakland Bay Bridge, which is a major regional bottleneck. The Bay Bridge transitioned from a constant toll to a peak step toll on July 1, 2010, so the site is well-suited for assessing many relevant extensions to the classic bottleneck model for the morning commute, including theories for congestion pricing. This paper uses traffic data from the Bay Bridge to address two areas of theoretical work on bottleneck congestion: (1) heterogeneous schedule preferences, and (2) equilibrium response to step tolls. The results suggest that the predicted equilibrium for the morning commute is consistent with observed traffic patterns, especially at the beginning of the rush. The data can be analyzed through the lens of relevant theories to quantify commuters’ value of earliness and lateness relative to queuing delay. The pattern of queuing suggests that the peak step toll has not been implemented in a socially optimal way.

Oil Spill Source Identification by Principal Component Analysis of Electrospray Ionization Fourier Transform Ion Cyclotron Resonance Mass Spectra

One fundamental challenge with either acute or chronic oil spills is to identify the source, especially in highly polluted areas, near natural oil seeps, when the source contains more than one petroleum product or when extensive weathering has occurred. Here we focus on heavy fuel oil that spilled (~200,000 L) from two suspected fuel tanks that were ruptured on the motor vessel (M/V) Cosco Busan when it struck the San Francisco-Oakland Bay Bridge in November 2007. We highlight the utility of principal component analysis (PCA) of elemental composition data obtained by high resolution FT-ICR mass spectrometry to correctly identify the source of environmental contamination caused by the unintended release of heavy fuel oil (HFO). Using ultrahigh resolution electrospray ionization (ESI) Fourier transform ion cyclotron resonance mass spectrometry, we uniquely assigned thousands of elemental compositions of heteroatom-containing species in neat samples from both tanks and then applied principal component analysis. The components were based on double bond equivalents for constituents of elemental composition, CcHhN1S1. To determine if the fidelity of our source identification was affected by weathering, field samples were collected at various intervals up to two years after the spill. We are able to identify a suite of polar petroleum markers that are environmentally persistent, enabling us to confidently identify that only one tank was the source of the spilled oil: in fact, a single principal component could account for 98% of the variance. Although identification is unaffected by the presence of higher polarity, petrogenic oxidation (weathering) products, future studies may require removal of such species by anion exchange chromatography prior to mass spectral analysis due to their preferential ionization by ESI.

ASCE Sixth Congress on Forensic Engineering

The Sixth Congress on Forensic Engineering will take place in San Francisco from October 31 to November 3, 2012. The three-day technical program is designed with four concurrent technical tracks, committee meetings, panel discussions, and workshops. More than 200 technical papers will be presented. The opening keynote speaker will be Raymond “Paul” Giroux, P.E., M.ASCE, who will present “The Building of the Golden Gate Bridge.” Giroux is chair of this year’s 75th anniversary celebration of the opening of the city’s signature bridge. The Forensic Engineering Congresses have been organized by the ASCE Technical Council on Forensic Engineering (TCFE) every 3 years since 1997. They are recognized internationally as a premier event for the forensic civil engineering community. This year’s location is particularly inviting because San Francisco is a major destination for tourism. In conjunction with the Congress, several tours are planned to enhance the Congress theme, “Gateway to a Safer Tomorrow.” These include a project tour of the San Francisco–Oakland Bay Bridge, one of the largest construction projects currently under way in the United States; a tour of the historic and elegant base-isolated San Francisco City Hall building; an exclusive evening event at the California Academy of Sciences, including a tour of the innovative award-winning museum designed by Renzo Piano; and a postconference tour of California’s wine country. Several new editions of TCFE publications are expected to become available at the time of the Congress. The second edition of Failure Case Studies in Civil Engineering—Structural, Foundation and Geoenvironmental is a product of the TCFE Education Committee. The TCFE Forensic Practices Committee will publish the second edition of Guidelines for Forensic Engineering Practice. A third publication, Guidelines for Failure Investigation, 2nd Ed., will be presented by the TCFE Forensic Investigation Committee. Authors will be available to discuss the content of these new publications. Opportunities will also be provided to attend TCFE committee meetings and to become committee members. Congress attendees are expected to include forensic engineers, civil engineers, architects and architectural engineers, building owners and managers, attorneys, educators, and invited guests. Registrants will benefit from an exchange of practices and procedures, educational opportunities, dissemination of information, and opportunities for publishing and networking. Professional development credits will be available. For further information, visit the Congress website at www.asce .org/forensiccongress.

Eight Mile Crossing: The San Francisco-Oakland Bay Bridge

The San Francisco-Oakland Bay Bridge was the product of a team of talented contributors, each bringing expertise to bear on what would become one of the most extraordinary achievements of 20th century bridge engineering.

Traffic Impacts of Variable Pricing on the San Francisco–Oakland Bay Bridge, California

The raising of bridge tolls in the peak period on the San Francisco-Oakland Bay Bridge in California in mid-2010 provides a rare opportunity to assess traffic impacts. Carpoolers who previously traveled for free during peak hours were charged an electronic toll under this variable pricing scheme. On the basis of 29 months of time series data, the introduction of carpool charges had a stronger impact on traffic volumes than did peak period pricing of regular traffic. The estimated short-term elasticity of about −0.30 suggests significant numbers of peak hour carpoolers did not travel, switched routes, shifted to public transit, or opted to drive alone. More than half the loss in carpool traffic was estimated to be attributable to the toll increase, a far stronger influence than factors such as rising gasoline prices and unemployment. The estimated elasticity of regular traffic in response to variable pricing was −0.23, an indication that peak period motorists were fairly insensitive to pricing and a reflection of the nondiscretionary nature of many peak hour journeys. It will be important to track trends over time to gain insights into the longer term sensitivity of motorists to variable pricing on natural corridors such as the San Francisco–Oakland Bay Bridge.