Saving Lives: Delay & Denial vs. Data

“Today’s production car (sic) … are all to varying degree @#$%-box death traps regarding safety.” – Henry “Smokey” Yunick in “Best Damn Garage in Town”

Famed NASCAR mechanic Henry “Smokey” Yunick penned those comments a decade and a half ago in his autobiography (under the title noted above) prior to passing away in 2001. Yunick was an advocate and self-taught promoter of innovation at the track including everything from engines to safety to aerodynamics. (He was often accused of “cheating” or breaking NASCAR rules.)

Yunick lived long enough to see many drivers die at the track including, near the end of his own life, the tragic demise of Dale Earnhardt in a crash in the final lap of the 2001 Daytona 500. That fatal crash helped usher in NASCAR’s adoption of the HANS (head and neck support) device (in 2002) to protect drivers in head-on crashes.

NASCAR was where Yunick and his fellow mechanics learned in real time about the importance of tire configuration; fuel formulations and lubrication; suspension; engine component manipulation; and the management of air around the car to improve traction. Yunick’s world revolved around chemistry and the mechanical elements of the car.

A lot has changed since Yunick’s day, and yet maybe not as much as we think. A little known sliver of Yunick’s legacy derives from regular columns he wrote for Popular Science magazine. Yunick was proud that his regular writings ultimately surpassed in popularity those of his Florida neighbor and fellow Popular Science columnist, Werner Von Braun.

Yunick received (and answered) massive amounts of letters from readers. It was in those letters that Yunick learned, decades ago, of vehicle complaints such as “runaway” GM cars and exploding Firestone tires. To the consternation of his editors, Yunick reported what his correspondents complained about and was validated when the reports proved accurate.

A similar situation has played out throughout the recent recall crisis that plagued the auto industry in 2014 – and lingers into 2015. Minor customer complaints or isolated fatal accidents were largely overlooked with devastating results to consumers and to the auto industry.

A NYTimes commentary this weekend notes that the Virginia Tech Transport Institute has conducted research into online automotive forums in an effort to identify “smoke words” – “words or phrases that are substantially more prevalent in posts about defects.” The goal of the research is to identify problem areas in cars and other consumer products from comments in forums.

The research (detailed here: https://tinyurl.com/okuzp3u - “Can Data Stop Car Wrecks”) is important, but it misses the most important lesson of the recall travails of 2014. The challenge lies not in identifying the problems. The challenge lies in getting car companies to accept the fact that a problem exists.

The average car owner knows that their biggest bugaboo is the problem that cannot be reproduced or diagnosed at or by the dealer. It was as true in Yunick’s time as it is today - some vehicle failures resist analysis or resolution.

Toyota’s lingering unintended acceleration (UA) woes are a good case in point. In fact, the struggle to find the source of the unintended acceleration problem back in 2011 brought to light the National Highway Traffic Safety Administration’s diagnostic and, by extension, regulatory limitations. NHTSA simply lacked the technical resources to identify the source of the UA failure (finally appealing to NASA for assistance), as described in these two posts from the Safety Research & Strategies Website:

• https://tinyurl.com/nb89fb9 - NHTSA: No Evidence Prius Unintended Acceleration Linked to Known Causes
• https://tinyurl.com/ns8hhav - Toyota Unintended Acceleration and the Big Bowl of “Spaghetti” Code

So the problem is not identifying problems. The problem is getting dealers, car makers or even regulators to respond to or acknowledge the existence of the problems. It’s a simple case of denial.

Decades ago, when Yunick told his friends at General Motors about their runaway cars they laughed at him, until the cars actually reproduced the errant behavior in the GM parking garage with Yunick and a GM pal sitting in the cars.

Writes Yunick of his experience at the time: “What did they do? Surprise, next day they started a department called ‘unusual and mysterious car behavioral problems,’ six months later person in charge wanted any letter I had on items like this.”

Yunick’s correspondents told him their dealers told them they were imagining the problems or they couldn’t figure it out. I have been given the “I-give-up-shrug” by many dealers over the years. (Have you?)

The same was true for Firestone – which went so far as to send Yunick multiple sets of tires, which continued to fail catastrophically on Yunick’s own car. The Firestone failures ultimately led to a government investigation – and Firestone’s acquisition (rescue?) by Bridgestone.

It is interesting that VTTI saw fit to data mine online forums for “smoke words” when NHTSA actually administers its own Website for capturing customer complaints: www.safercar.gov. The complaints of drivers are there for the world and NHTSA to see, categorized by car maker, model and year. (Consumers can also check on that site to see if their cars have outstanding recalls.)

Drivers are the original canaries in the coal mine – whether you’re driving a NASCAR vehicle or a passenger car. Drivers report vehicle complaints on a daily basis to friends, forums, NHTSA, dealers and to the car makers. It must get tiresome for the executives tasked with responding to these complaints. But, as an industry, we ignore these complaints at our peril. They are early warning signs.

Should the government and car makers be doing more to mine complaints and crash data for insights into the causes of crashes and fatalities and to identify and correct potentially deadly vehicle defects? Of course. Is there a culture of denial and delay preventing this process from functioning in a more efficient and effective manner. Definitely.

Is the problem more severe now than ever before with tens of millions of lines of software code potentially standing between the existence of a problem and the ability to properly diagnose it? Clearly.

New tools are needed in this battle to save lives. The task may well be beyond the capability of NHTSA. But NHTSA, at least in the U.S., is all we have.

There is hope in the form of new leadership, Director Mark Rosekind. Rosekind has already expressed his belief that anything less than a 100% recall fulfillment rate is unsatisfacory. Maybe he can apply that same common sense to the culture of denial and delay that has clearly contributed to this country’s shocking annual highway fatality rate.

And maybe he can help the agency come to terms with the need for an enhanced forensic software capability. Maybe car crashes ought to be treated like airplane crashes with the correlated retrieval and analysis of black box data.