E.J. Antoni of the Heritage Foundation testifies before a U.S. Senate Judiciary subcommittee on Oct. 24, 2023. (Screenshot from C-SPAN)
WASHINGTON — President Donald Trump nominated conservative economist E.J. Antoni to fill the top spot at the Bureau of Labor Statistics after abruptly firing the previous statistician following a disappointing jobs report earlier this month.
Trump announced the nominee late Monday on his Truth Social platform, stating that “Our Economy is booming, and E.J. will ensure that the Numbers released are HONEST and ACCURATE.”
Antoni, an economist at the far-right Heritage Foundation, has harshly criticized the previous BLS commissioner, Erika McEntarfer, who was nominated by former President Joe Biden in 2023 and confirmed by the U.S. Senate 86-6 in January 2024. The bureau tracks national economic data, including employment figures.
Without providing evidence, Trump slammed the latest jobs report, released Aug. 1, as “RIGGED” and fired McEntarfer hours later.
The economy gained just 73,000 jobs in July, according to the monthly report. BLS also significantly adjusted May and June figures, to 33,000 for both months, down from the previously reported 291,000. Revisions to past reports often happen after the bureau receives updated data from businesses and federal agencies.
U.S. economic data collection is often referred to as the “gold standard,” as Federal Reserve Chair Jerome Powell, a Trump appointee, said last month.
Trump faced backlash for firing McEntarfer, including from his own former BLS commissioner.
William Beach, whom Trump tapped in 2017 to lead BLS, told CNN McEntarfer’s firing was “groundless.”
BLS data is “more accurate now than they were 30 years ago,” Beach said during the Aug. 3 interview.
In an Aug. 4 appearance on Steve Bannon’s WarRoom podcast, Antoni said BLS data collection is “outdated.”
“You need somebody who is willing to overhaul the entire thing,” he told Bannon.
Shortly after Trump’s November win, Antoni posted on X that “DOGE needs to take a chainsaw to BLS.”
Kevin Roberts, Heritage Foundation’s president, said Tuesday that Trump made a “stellar choice” in nominating Antoni.
“EJ Antoni is one of the sharpest economic minds in the nation—a fearless truth-teller who grasps that sound economics must serve the interests of American families, not globalist elites,” Roberts said in a statement. “His leadership as chief economist at The Heritage Foundation has been instrumental in advancing our mission to protect American families and rebuild a resilient economy rooted in free enterprise.”
Antoni contributed to the Heritage Foundation’s Project 2025, a roughly 900-page far-right blueprint to overhaul government institutions published ahead of Trump’s election win.
Antoni will need approval from the Senate, which currently has a 53-47 Republican majority.
Sen. Patty Murray, a senior member and former chair of the Senate Committee on Health, Education, Labor and Pensions, slammed Antoni as an “unqualified right-wing extremist who won’t think twice about manipulating BLS data and degrading the credibility of the agency to make Trump happy.”
“Any Senator who votes to confirm this partisan hack is voting to shred the integrity of our nation’s best economic and jobs data, which underpin our entire economy. If E.J. Antoni gets confirmed, I hope Republicans like playing make-believe, because that’s all BLS data will become,” the Washington state Democrat said in a statement Tuesday.
Sen. Bill Cassidy chairs the committee, which will be tasked with advancing Antoni’s nomination to the full Senate. Cassidy, of Louisiana, did not have a statement on Antoni posted on his website or X feed as of Tuesday at 3 p.m. Eastern.
Photo by Architect of the Capitol | U.S. government work via Flickr
The July jobs report released last Friday wasn’t pretty. It showed weaker than anticipated U.S. job growth in July, and there were substantial downward revisions of jobs numbers for May and June as well. Economists predicted a slowdown. The chaos of tariff threats has created substantial uncertainty, which is bad for the economy, and the tariffs that have gone into effect have raised prices. It’s no surprise, then, that we’re seeing a slowdown in jobs.
Moody’s chief economist Mark Zandi noted on social media, “It’s no mystery why the economy is struggling; blame increasing U.S. tariffs and highly restrictive immigration policy. The tariffs are cutting increasingly deeply into the profits of American companies and the purchasing power of American households. Fewer immigrant workers means a smaller economy.”
But instead of reflecting on mistakes in economic policy or offering some austerity suggestion, like limiting U.S. children to two dolls each , President Donald Trump blamed the messenger, firing the government official in charge of the data release, commissioner of the Bureau of Labor Statistics (BLS) Erika McEntarfer. He baselessly asserted that the bad news was “concocted” and suggested that he knows better than the data. The economy is great, according to him, and he will find a commissioner to tell him so.
Trump’s approach is a disaster for economic decision making and for public trust. The BLS is an independent agency with a strong legacy of providing the data that businesses, analysts and policymakers need. Good economic decisions require reliable data. As the American Economics Association wrote: “The BLS has long had a well-deserved reputation for professional excellence and nonpartisan integrity. Safeguarding this tradition is vital for the continued health of the U.S. economy and public trust in our institutions.”
The BLS monthly jobs report provides a timely snapshot of labor market dynamics which inform investing and hiring decisions as well as policy choices. BLS data also measures the rate of inflation through the consumer price index. The rising price of goods is not only a key economic indicator but also the scale by which Social Security payments are adjusted and a point of reference in private and union wage negotiations.
BLS data are essential to understanding what is going on in the economy, when a slowdown is emerging, and the cost of daily life. The independence and integrity of the agency, long assumed and supported by both parties, is now under attack.
Wisconsinites lived through something like this more than a decade ago. Former Republican Gov. Scott Walker promised to create 250,000 jobs in his first term. He focused on the goal relentlessly, at least until it became clear that he would not meet it. (In fact, the Wisconsin economy didn’t even meet Walker’s first term goal across his two terms – adding just 233,000 jobs by the time he left office after serving for eight years.)
In the first years of Walker’s “relentless focus on jobs” under his administration’s tagline “Wisconsin is Open for Business,” the monthly numbers showed that Wisconsin’s economy was growing more slowly than the national labor market and neighboring states.
Walker blamed the data. He insisted that we wait instead for a federal source which was more reliable, but had a substantial time lag. As someone who watches this data, I can assure that this was the only time in my three-decade career when differences between monthly and quarterly sources of federal jobs data were a policy talking point.
But in the end, the data issue was just a distraction from the truth. Wisconsin was growing more slowly, and no amount of complaining about the data or waiting for another source on jobs could change that fact. Eventually, the Walker administration went silent on both the data and the promised 250,000 jobs.
Trump’s approach is worse than waiting for another source of data. His firing of the commissioner suggests that he’ll only accept data that confirms his narrative. And that makes it harder for any of us to trust any data the federal government is willing to release.
That’s bad for the economy and bad for democracy. As narrow and nerdy as this topic may seem, we all have an interest in facts and reliable data. We have had a government infrastructure capable of producing it. We lose it at our own peril.
On May 6, MIT AgeLab’s Advanced Vehicle Technology (AVT) Consortium, part of the MIT Center for Transportation and Logistics, celebrated 10 years of its global academic-industry collaboration. AVT was founded with the aim of developing new data that contribute to automotive manufacturers, suppliers, and insurers’ real-world understanding of how drivers use and respond to increasingly sophisticated vehicle technologies, such as assistive and automated driving, while accelerating the applied insight needed to advance design and development. The celebration event brought together stakeholders from across the industry for a set of keynote addresses and panel discussions on critical topics significant to the industry and its future, including artificial intelligence, automotive technology, collision repair, consumer behavior, sustainability, vehicle safety policy, and global competitiveness.
Bryan Reimer, founder and co-director of the AVT Consortium, opened the event by remarking that over the decade AVT has collected hundreds of terabytes of data, presented and discussed research with its over 25 member organizations, supported members’ strategic and policy initiatives, published select outcomes, and built AVT into a global influencer with tremendous impact in the automotive industry. He noted that current opportunities and challenges for the industry include distracted driving, a lack of consumer trust and concerns around transparency in assistive and automated driving features, and high consumer expectations for vehicle technology, safety, and affordability. How will industry respond? Major players in attendance weighed in.
In a powerful exchange on vehicle safety regulation, John Bozzella, president and CEO of the Alliance for Automotive Innovation, and Mark Rosekind, former chief safety innovation officer of Zoox, former administrator of the National Highway Traffic Safety Administration, and former member of the National Transportation Safety Board, challenged industry and government to adopt a more strategic, data-driven, and collaborative approach to safety. They asserted that regulation must evolve alongside innovation, not lag behind it by decades. Appealing to the automakers in attendance, Bozzella cited the success of voluntary commitments on automatic emergency braking as a model for future progress. “That’s a way to do something important and impactful ahead of regulation.” They advocated for shared data platforms, anonymous reporting, and a common regulatory vision that sets safety baselines while allowing room for experimentation. The 40,000 annual road fatalities demand urgency — what’s needed is a move away from tactical fixes and toward a systemic safety strategy. “Safety delayed is safety denied,” Rosekind stated. “Tell me how you’re going to improve safety. Let’s be explicit.”
Drawing inspiration from aviation’s exemplary safety record, Kathy Abbott, chief scientific and technical advisor for the Federal Aviation Administration, pointed to a culture of rigorous regulation, continuous improvement, and cross-sectoral data sharing. Aviation’s model, built on highly trained personnel and strict predictability standards, contrasts sharply with the fragmented approach in the automotive industry. The keynote emphasized that a foundation of safety culture — one that recognizes that technological ability alone isn’t justification for deployment — must guide the auto industry forward. Just as aviation doesn’t equate absence of failure with success, vehicle safety must be measured holistically and proactively.
With assistive and automated driving top of mind in the industry, Pete Bigelow of Automotive News offered a pragmatic diagnosis. With companies like Ford and Volkswagen stepping back from full autonomy projects like Argo AI, the industry is now focused on Level 2 and 3 technologies, which refer to assisted and automated driving, respectively. Tesla, GM, and Mercedes are experimenting with subscription models for driver assistance systems, yet consumer confusion remains high. JD Power reports that many drivers do not grasp the differences between L2 and L2+, or whether these technologies offer safety or convenience features. Safety benefits have yet to manifest in reduced traffic deaths, which have risen by 20 percent since 2020. The recurring challenge: L3 systems demand that human drivers take over during technical difficulties, despite driver disengagement being their primary benefit, potentially worsening outcomes. Bigelow cited a quote from Bryan Reimer as one of the best he’s received in his career: “Level 3 systems are an engineer’s dream and a plaintiff attorney’s next yacht,” highlighting the legal and design complexity of systems that demand handoffs between machine and human.
In terms of the impact of AI on the automotive industry, Mauricio Muñoz, senior research engineer at AI Sweden, underscored that despite AI’s transformative potential, the automotive industry cannot rely on general AI megatrends to solve domain-specific challenges. While landmark achievements like AlphaFold demonstrate AI’s prowess, automotive applications require domain expertise, data sovereignty, and targeted collaboration. Energy constraints, data firewalls, and the high costs of AI infrastructure all pose limitations, making it critical that companies fund purpose-driven research that can reduce costs and improve implementation fidelity. Muñoz warned that while excitement abounds — with some predicting artificial superintelligence by 2028 — real progress demands organizational alignment and a deep understanding of the automotive context, not just computational power.
Turning the focus to consumers, a collision repair panel drawing Richard Billyeald from Thatcham Research, Hami Ebrahimi from Caliber Collision, and Mike Nelson from Nelson Law explored the unintended consequences of vehicle technology advances: spiraling repair costs, labor shortages, and a lack of repairability standards. Panelists warned that even minor repairs for advanced vehicles now require costly and complex sensor recalibrations — compounded by inconsistent manufacturer guidance and no clear consumer alerts when systems are out of calibration. The panel called for greater standardization, consumer education, and repair-friendly design. As insurance premiums climb and more people forgo insurance claims, the lack of coordination between automakers, regulators, and service providers threatens consumer safety and undermines trust. The group warned that until Level 2 systems function reliably and affordably, moving toward Level 3 autonomy is premature and risky.
While the repair panel emphasized today’s urgent challenges, other speakers looked to the future. Honda’s Ryan Harty, for example, highlighted the company’s aggressive push toward sustainability and safety. Honda aims for zero environmental impact and zero traffic fatalities, with plans to be 100 percent electric by 2040 and to lead in energy storage and clean power integration. The company has developed tools to coach young drivers and is investing in charging infrastructure, grid-aware battery usage, and green hydrogen storage. “What consumers buy in the market dictates what the manufacturers make,” Harty noted, underscoring the importance of aligning product strategy with user demand and environmental responsibility. He stressed that manufacturers can only decarbonize as fast as the industry allows, and emphasized the need to shift from cost-based to life-cycle-based product strategies.
Finally, a panel involving Laura Chace of ITS America, Jon Demerly of Qualcomm, Brad Stertz of Audi/VW Group, and Anant Thaker of Aptiv covered the near-, mid-, and long-term future of vehicle technology. Panelists emphasized that consumer expectations, infrastructure investment, and regulatory modernization must evolve together. Despite record bicycle fatality rates and persistent distracted driving, features like school bus detection and stop sign alerts remain underutilized due to skepticism and cost. Panelists stressed that we must design systems for proactive safety rather than reactive response. The slow integration of digital infrastructure — sensors, edge computing, data analytics — stems not only from technical hurdles, but procurement and policy challenges as well.
Reimer concluded the event by urging industry leaders to re-center the consumer in all conversations — from affordability to maintenance and repair. With the rising costs of ownership, growing gaps in trust in technology, and misalignment between innovation and consumer value, the future of mobility depends on rebuilding trust and reshaping industry economics. He called for global collaboration, greater standardization, and transparent innovation that consumers can understand and afford. He highlighted that global competitiveness and public safety both hang in the balance. As Reimer noted, “success will come through partnerships” — between industry, academia, and government — that work toward shared investment, cultural change, and a collective willingness to prioritize the public good.
Besides thanking the various individuals involved in putting together the 17th National Congress on School Transportation last month in Des Moines, Iowa, and providing copies of the proceedings to the National Highway Traffic Safety Administration and other organizations, resolutions focused on increased safety and data keeping.
The most time-sensitive resolution is No. 6, which requests clarification on FMVSS 213a and 213b final rules related to the performance and use of child safety restraint systems (CSRS). NCST submitted the resolution to the National Highway Traffic Safety Administration immediately. It notes that the school transportation industry takes great pride in providing the safest form of transportation available and that preschool and special needs transportation are a sizable component of the industry.
The resolution states the importance of further engaging NHTSA “to ensure children requiring securement based on age and weight are carried safely and securely, child safety restraint systems are attached to the seatback to ensure a secure fit for the child. It is believed that there are approximately 310,000 to 335,000 [child safety restraint systems] specifically designed for school buses on the road.”
NHTSA is currently accepting comments for a notice of proposed rulemaking initiated as a result of the final rules for FMVSS 213a and 213b that would exempt CSRS from side-impact protection requirements and lower anchorage attachment requirements (due to being designed for school buses specifically). The NPRM also states that the CRABI-12MO test dummy is no longer being used to test forward-facing CSRS for side impact, and that labels on school bus CRSs will also be updated to reflect their installation method, versus referencing vehicle belts or child restraint anchorage systems.
The NPRM also seeks to delay the implementation of FMVSS 213a and 213b to Dec. 5, 2026 from June 30, 2025, giving more time to manufacturers to test and certify their products.
Resolution 1 expressed appreciation to Patrick McManamon for serving as NCST Chair from 2015 to 2024. He stepped down as chair earlier this year citing professional and personal reasons.
Resolution 3 recognized the following individuals for serving as on-site officials and for their dedication and service to NCST.
– Mike LaRocco, conference chair
– Charlie Hood, on-site chair
– Susan Miller, on-site coordinator
– Lori Wille, editor
– Laura Meade, parliamentarian
– Rene Dawson & Reginald White, timekeepers
– Samantha Kobussen, National School Transportation Specifications and Procedures artwork
– Zander Press, printer
– Ronna Weber and NASDPTS leadership for “making the Congress a success in the manner it was organized and concluded.”
However, the NCST resolution asks NHTSA if CSRS specifically designed for school buses — such as the IMMI Star, BESI ProTech, and HSM Portable Child Restraint — are exempt from the side-impact requirements under FMVSS 213a, as of the effective date of June 30. If they are not exempt, NCST questioned if devices manufactured prior to June 30 will remain permissible for continued use beyond the implementation deadline. The resolution also asks, in the event the specified CSRS are not exempt and in consideration of maintaining a high standard of safety, what alternative CSRS models or types would be deemed acceptable for continued use on school buses.
The resolution seeks clarification from NHTSA if it will be issuing any additional guidance or initiating rulemaking specifically addressing the use and approval of CSRSs for school bus applications prior to the June 30 effective date. It also asks NHTSA if it will be updating the curriculum for the Child Passenger Safety on School Buses training courses to reflect the forthcoming changes, particularly those involving add-on school bus securement systems.
“The NCST respectfully urges NHTSA to provide a formal response and guidance at the earliest possible opportunity, mindful of the June, 30, 2025 implementation date to support informed decision-making, training readiness, and procurement planning by school transportation providers nationwide,” the resolution states.
NHTSA mandates transportation equipment design and safety performance requirements but does not regulate use. States establish requirements for each type of CSRS based on a child’s age and weight as well as the vehicle. NHTSA did publish Guideline for the Safe Transportation of Pre-school Age Children in School Buses, which essentially recommends using CSRS for the appropriate weight and height of children and following CSRS manufacturer installation instructions. That guideline, which is not binding for states, came out in February 1999 and no updates have been made since.
Additionally, the NHTSA-sponsored Child Passenger Safety on School Buses, taught at TSD Conference, is also best-practice guidance and not a regulation. It was already updated in 2023 by the National Safety Council. The organization develops and maintains the curriculum. The NHTSA website also includes a School Bus Safety page that links to more information on the eight-hour, hands-on securement training.
Meanwhile, Resolution 2 referenced a March 2024 School Transportation News article that identified a student passenger reporting challenge that indicates school bus ridership is disappearing. The Editor’s Take column by Ryan Gray noted that the National Household Travel Survey (NHTS) sponsored by the Federal Highway Administration indicates the number of students transported nationwide by the yellow school bus is about one-third less than the figure used by the industry.
The resolution recognizes the need to develop a standardized reporting system for collecting school bus ridership data and “requests the interim steering committee of the 18th NCST to appoint a focus group to research and develop recommendations for standardization of data collection relative to ridership on school bus and make periodic reports to the Interim Committee.”
Data collection for the 2024 survey is expected to be completed this fall.
Resolution 5 “encourages transportation professionals to plan bus stops that are not in proximity to known registered sexual offenders when made aware, when possible. Training programs should be provided to all transportation personnel on recognizing and reporting suspected or known human trafficking.”
The resolution states that the NCST is aware of the safety concerns associated with sexual predators and offenders as well as human trafficking, noting an increase these crimes occurring across the U.S.
All NCST resolution proposals presented to the state delegations passed.
Wisconsin Watch, a nonprofit news organization that uses journalism to make Wisconsin communities strong, informed and connected, seeks a data investigative reporter to expand our capacity to provide vital evidence and context to our reporting.
Working with other journalists in our statewide and Milwaukee newsrooms, you will use data to help us better understand Wisconsin communities and hold leaders to account. We believe that access to truthful local news is critical to a healthy democracy and to finding solutions to the most pressing problems of everyday life.
Job duties
Reporting to the managing editor, you will:
Find, compile and clean data that powers our journalism.
Plan and execute quantitative analyses — and interpret results — to support stories and visualizations.
Design and build creative static and interactive graphics to visualize findings.
Pitch and develop your own stories with support from editors.
Help other journalists advance their data journalism skills, teaching and encouraging best practices across the newsrooms.
At Wisconsin Watch we make sure that we are producing quality journalism and give our reporters the time they need to make sure the job is done well. Rather than chasing clicks, we measure success through the impact we deliver to those we serve.
Required qualifications: The ideal candidate will bring a public service mindset and a demonstrated commitment to nonpartisan journalism ethics, including a commitment to abide by Wisconsin Watch’s ethics policies. More specifically, we’re looking for a reporter who:
Has worked on data projects in a newsroom or has performed statistical analysis in a research setting.
Demonstrates ability to analyze data in Python, R, SQL or a similar high-level language.
Has experience with off-the-shelf data visualization tools like Datawrapper or Flourish.
Demonstrates ability to formulate compelling story pitches to editors.
Aches to report and support stories that explore solutions to challenges.
Has experience with or ideas about the many ways newsrooms can inform the public.
Has experience working with others. Wisconsin Watch is a deeply collaborative organization. Our journalists frequently team up with each other or with colleagues at other news outlets to maximize the potential impact of our reporting.
Bonus skills:
Familiarity with Wisconsin, its history and its politics.
Beat reporting experience.
Spanish-language proficiency.
Don’t check off every box in the requirements listed above? Please apply anyway! Wisconsin Watch is dedicated to building an inclusive, diverse, equitable, and accessible workplace that fosters a sense of belonging – so if you’re excited about this role but your past experience doesn’t align perfectly with every qualification in the job description, we encourage you to still consider submitting an application. You may be just the right candidate for this role or another one of our openings!
Location: The reporter must be located in Wisconsin. Wisconsin Watch is a statewide news organization with staff based in Madison, Milwaukee and Green Bay.
Salary and benefits: The salary range is $50,000-$70,000. Benefits include five weeks of vacation; paid sick leave and family and caregiver leave; 75% reimbursement for silver-tier health and dental insurance on the federal exchange; 100% vision insurance coverage; $100 per paycheck automatic employer contribution to a 403(b) retirement plan (no match required) after 90 days.
Final offer amounts will carefully consider multiple factors, and higher compensation may be available for someone with advanced skills and/or experience.
Deadline: Applications will be accepted until the position is filled. For best consideration, apply by June 2.
To apply: Please submit a single PDF of your resume and work samples and answer some brief questions in this application form. If you’d like to chat about the job before applying, contact Managing Editor Jim Malewitz at jmalewitz@wisconsinwatch.org.
Wisconsin Watch is dedicated to improving our newsroom by better reflecting the people we cover. We are committed to diversity and building an inclusive environment for people of all backgrounds and ages. We especially encourage members of traditionally underrepresented communities to apply, including women, people of color, LGBTQ+ people, and people with disabilities. We are an equal-opportunity employer and prohibit discrimination and harassment of any kind. All employment decisions are made without regard to race, color, religion, sex, sexual orientation, national origin, age, or any other status protected under applicable law.
About Wisconsin Watch and Milwaukee Neighborhood News Service
Founded in 2009, Wisconsin Watch is a nonprofit news organization dedicated to producing independent, nonpartisan journalism that makes the communities of Wisconsin strong, informed and connected. We believe that access to truthful local news is critical to a healthy democracy and to finding solutions to the most pressing problems of everyday life. Under the Wisconsin Watch umbrella, we have multiple news departments including a statewide investigative and explanatory projects team, a Capitol bureau, a regional collaboration in northeast Wisconsin called the NEW News Lab, and Milwaukee Neighborhood News Service (NNS).
NNS was founded in 2011 as a mission-driven newsroom that reports on and celebrates Milwaukee’s central city neighborhoods. Through its reporting, website, e-newsletters and News414 texting service, NNS covers ordinary people who do extraordinary things, connects readers with resources and serves as a watchdog for their neighbors. Together, Wisconsin Watch’s state team and NNS reporters collaborate to produce solutions-oriented investigative and explanatory stories highlighting issues affecting communities in Milwaukee.
Car design is an iterative and proprietary process. Carmakers can spend several years on the design phase for a car, tweaking 3D forms in simulations before building out the most promising designs for physical testing. The details and specs of these tests, including the aerodynamics of a given car design, are typically not made public. Significant advances in performance, such as in fuel efficiency or electric vehicle range, can therefore be slow and siloed from company to company.
MIT engineers say that the search for better car designs can speed up exponentially with the use of generative artificial intelligence tools that can plow through huge amounts of data in seconds and find connections to generate a novel design. While such AI tools exist, the data they would need to learn from have not been available, at least in any sort of accessible, centralized form.
But now, the engineers have made just such a dataset available to the public for the first time. Dubbed DrivAerNet++, the dataset encompasses more than 8,000 car designs, which the engineers generated based on the most common types of cars in the world today. Each design is represented in 3D form and includes information on the car’s aerodynamics — the way air would flow around a given design, based on simulations of fluid dynamics that the group carried out for each design.
Each of the dataset’s 8,000 designs is available in several representations, such as mesh, point cloud, or a simple list of the design’s parameters and dimensions. As such, the dataset can be used by different AI models that are tuned to process data in a particular modality.
DrivAerNet++ is the largest open-source dataset for car aerodynamics that has been developed to date. The engineers envision it being used as an extensive library of realistic car designs, with detailed aerodynamics data that can be used to quickly train any AI model. These models can then just as quickly generate novel designs that could potentially lead to more fuel-efficient cars and electric vehicles with longer range, in a fraction of the time that it takes the automotive industry today.
“This dataset lays the foundation for the next generation of AI applications in engineering, promoting efficient design processes, cutting R&D costs, and driving advancements toward a more sustainable automotive future,” says Mohamed Elrefaie, a mechanical engineering graduate student at MIT.
Elrefaie and his colleagues will present a paper detailing the new dataset, and AI methods that could be applied to it, at the NeurIPS conference in December. His co-authors are Faez Ahmed, assistant professor of mechanical engineering at MIT, along with Angela Dai, associate professor of computer science at the Technical University of Munich, and Florin Marar of BETA CAE Systems.
Filling the data gap
Ahmed leads the Design Computation and Digital Engineering Lab (DeCoDE) at MIT, where his group explores ways in which AI and machine-learning tools can be used to enhance the design of complex engineering systems and products, including car technology.
“Often when designing a car, the forward process is so expensive that manufacturers can only tweak a car a little bit from one version to the next,” Ahmed says. “But if you have larger datasets where you know the performance of each design, now you can train machine-learning models to iterate fast so you are more likely to get a better design.”
And speed, particularly for advancing car technology, is particularly pressing now.
“This is the best time for accelerating car innovations, as automobiles are one of the largest polluters in the world, and the faster we can shave off that contribution, the more we can help the climate,” Elrefaie says.
In looking at the process of new car design, the researchers found that, while there are AI models that could crank through many car designs to generate optimal designs, the car data that is actually available is limited. Some researchers had previously assembled small datasets of simulated car designs, while car manufacturers rarely release the specs of the actual designs they explore, test, and ultimately manufacture.
The team sought to fill the data gap, particularly with respect to a car’s aerodynamics, which plays a key role in setting the range of an electric vehicle, and the fuel efficiency of an internal combustion engine. The challenge, they realized, was in assembling a dataset of thousands of car designs, each of which is physically accurate in their function and form, without the benefit of physically testing and measuring their performance.
To build a dataset of car designs with physically accurate representations of their aerodynamics, the researchers started with several baseline 3D models that were provided by Audi and BMW in 2014. These models represent three major categories of passenger cars: fastback (sedans with a sloped back end), notchback (sedans or coupes with a slight dip in their rear profile) and estateback (such as station wagons with more blunt, flat backs). The baseline models are thought to bridge the gap between simple designs and more complicated proprietary designs, and have been used by other groups as a starting point for exploring new car designs.
Library of cars
In their new study, the team applied a morphing operation to each of the baseline car models. This operation systematically made a slight change to each of 26 parameters in a given car design, such as its length, underbody features, windshield slope, and wheel tread, which it then labeled as a distinct car design, which was then added to the growing dataset. Meanwhile, the team ran an optimization algorithm to ensure that each new design was indeed distinct, and not a copy of an already-generated design. They then translated each 3D design into different modalities, such that a given design can be represented as a mesh, a point cloud, or a list of dimensions and specs.
The researchers also ran complex, computational fluid dynamics simulations to calculate how air would flow around each generated car design. In the end, this effort produced more than 8,000 distinct, physically accurate 3D car forms, encompassing the most common types of passenger cars on the road today.
To produce this comprehensive dataset, the researchers spent over 3 million CPU hours using the MIT SuperCloud, and generated 39 terabytes of data. (For comparison, it’s estimated that the entire printed collection of the Library of Congress would amount to about 10 terabytes of data.)
The engineers say that researchers can now use the dataset to train a particular AI model. For instance, an AI model could be trained on a part of the dataset to learn car configurations that have certain desirable aerodynamics. Within seconds, the model could then generate a new car design with optimized aerodynamics, based on what it has learned from the dataset’s thousands of physically accurate designs.
The researchers say the dataset could also be used for the inverse goal. For instance, after training an AI model on the dataset, designers could feed the model a specific car design and have it quickly estimate the design’s aerodynamics, which can then be used to compute the car’s potential fuel efficiency or electric range — all without carrying out expensive building and testing of a physical car.
“What this dataset allows you to do is train generative AI models to do things in seconds rather than hours,” Ahmed says. “These models can help lower fuel consumption for internal combustion vehicles and increase the range of electric cars — ultimately paving the way for more sustainable, environmentally friendly vehicles.”
“The dataset is very comprehensive and consists of a diverse set of modalities that are valuable to understand both styling and performance,” says Yanxia Zhang, a senior machine learning research scientist at Toyota Research Institute, who was not involved in the study.
This work was supported, in part, by the German Academic Exchange Service and the Department of Mechanical Engineering at MIT.
In a new dataset that includes more than 8,000 car designs, MIT engineers simulated the aerodynamics for a given car shape, which they represent in various modalities, including “surface fields.”
Login
