Top Stories from the Special Reports pages of the International Herald Tribune,
Thursday, September 23, 1999


Test Dummies Get Smarter

By Brad Spurgeon International Herald Tribune
PARIS - As the job of smashing cars into walls to test their crashworthiness has grown from a kind of backyard craft into a science, its primary measuring tool -- the crash-test dummy -- has developed from little more than a man-sized doll into a family of highly sensitive humanoids of synthetic flesh and bone.

''When this business started some 40 years ago, people were happy if we just saved their chest and head from being injured, because that's how people get killed,'' said Nagarajan Rangarajan, president of a crash-test dummy company called GESAC Inc., in Boonsboro, Maryland. ''Now people are doing a good job of preventing serious injuries in the normal course of an accident, but we have injuries to the lower leg and to the face, which are not life-threatening, but that cost a lot of money to fix.''

The economic cost of car accidents and a recent realization by car manufacturers that safety sells has led to a worldwide proliferation of crash-test programs and research funded by both governments and private industry.

About 400,000 people are killed and 12 million injured in car accidents around the world every year. In the United States alone, car injuries cost $150 billion a year, according to the Department of Transportation.

But one of the drawbacks to this crash-testing boom is an imbroglio of test standards and dummy designs that has led to an uneven level of safety and a higher cost for cars as manufacturers spend money to meet a multiplicity of crash-test standards.

''There are a lot of different crash dummies around the world,'' said Gordie Morgan, director technical marketing of First Technology Safety Systems Inc., a Plymouth, Michigan-based, British-owned company that makes 80 percent of the world's dummies. ''But human beings are basically the same. So there's a big push to harmonize.''

With names like Hybrid III, CRABI, BioSID, BioRID, EuroSID-1, SID-II and THOR, the dummy population may be an indecipherable jungle of bureaucratic acronyms, but behind it all is a highly complex science called biomechanics, the goal of which is simple: to save lives.

In the early days, a crash test basically involved a car moving along a sled into a wall. The steering wheel was not supposed to move back farther than a certain distance. The first dummy was a primitive number called Sierra Sam, designed in 1949 for the U.S. Air Force to test ejection seats.

Today's plethora of tests frontal, rear, offset, roll-over, and side-impact, among others measure how much a car's structure deforms or collapses, and how much intrusion there is into the passenger compartment. The tests measure the effectiveness of the restraint systems, such as seatbelts and airbags, to see if the dummy hits its head on the A-pillar, the B-pillar, the roof rail, or the door frame, or if the dummy slides under the airbag or hits its knees, chest or other parts of its body.

The crash barrier at the test hall of the Insurance Institute for Highway Safety, an independent organization based in Virginia and funded mostly by car insurance companies, is 320,000 pounds (145,000 kilograms) of concrete and steel. It sits on a cushion of air and may be moved around for different crash configurations. An aluminum honeycomb facing is sometimes used to mimic the front end of another vehicle. The speed of the propulsion system that drives the vehicles may be varied within a 10th of a mile an hour. A total of 750,000 watts of lighting allow for high-speed, frame-by-frame filming to analyze the crash digitally.

Dummies are fully clothed in order to slide on the car seat as a human might. Rear impact dummies have complex spinal columns; side impact dummies have realistic rib cages; dummies for measuring foot and lower leg accidents have Achilles' tendons.


ELECTRONIC sensors providing sometimes more than 180 data channels are scattered throughout the dummy's body and attached through an umbilical cord to a black box, like an airplane's flight recorder, which stays inside the car during the crash. The box provides the sensors with power and records the signals at up to 20,000 times per second, with up to nearly 40,000 items of data for later analysis in a computer. This data is then compared to data established beforehand from work on human corpses by biomechanical researchers on what a bone can take before it breaks.

Today, a fully instrumented dummy costs about $150,000. And there is now a whole family of them: a 1-year-old, a 3-year-old, a 6-year-old, a female dummy that is also used for teenagers and a male dummy. First Technology Safety Systems is now working on a pregnant dummy. The most common male adult is what they call a ''50th percentile'' male, representing an average male, 178 centimeters (5 feet 10 inches) tall and weighing 76.5 kilograms (170 pounds). But there's also a 95th percentile male representing the heavyweights.

''What you're always looking for in next generation dummies is to make them more representative of human beings,'' said Julie Rochman, vice president of communications for the Insurance Institute for Highway Safety.

Two projects aim to create the most lifelike dummies yet. One is by GESAC, which is under contract to the National Highway Traffic Safety Administration to build the next generation frontal impact dummy called THOR (Test Device for Human Occupant Restraint). The other is the OCATD (Occupant Classification Anthropomorphic Test Device) being developed by First Technology Safety Systems with funding provided by the Alliance of Automobile Manufacturers, comprised of 11 of the world's leading car manufacturers.

THOR has a spine and pelvis that allow it to sit in different positions, slouching or upright. Its sensors provide more injury measurements than those available on Hybrid III.

OCATD uses a replica of a human skeleton made out of plastic. Encompassing that is a flesh-like material, also made of a kind of plastic but designed on the basis of several tissue and muscular studies to ensure its stiffness and compliance properties are humanlike. One of the main features is the buttocks bone and profile, which will be used in a new car-seat sensor system that will tell the airbag what size and shape of human is sitting on the seat and whether or not, or precisely how to deploy in an accident.

IN PART to solve the problem of the dangers of airbags and small occupants, legislation is beginning to move in the United States as the National Highway Traffic Safety Administration adds the whole family of dummies to the crash-test regulations over the next few months.

In Europe, after regulations remained static for two decades, the International Automobile Federation (FIA) a Geneva-based organization composed of most of the world's automobile associations and representing 100 million motorists lobbied for better legislation over the past few years. Then the FIA and the governments of Britain, Sweden, the Netherlands and later France created the European New Car Assessment Program (Euro NCAP), which tests the best-selling new model cars in Europe with a frontal and side-impact test and a pedestrian safety test.

But in the most ambitious project yet to harmonize dummies, the International Standardization Organization in Switzerland is overseeing a worldwide effort, called WorldSID, to design a single side-impact dummy.

There are now at least four different side-impact dummies in use. A prototype dummy is expected next autumn.

Marc Beusenberg, director of a Canadian company called Biokinetics and Associates Ltd., is project manager for WorldSID's design team. The team is made up of half a dozen of the world's top dummy builders. ''We are trying to build a dummy that is more humanlike than any of the existing dummies,'' Mr. Beusenberg said. ''So it is not just the political issue or harmonization of standards issue to this, but also an improved safety aspect to the development.''

WorldSID owes its existence partly to a flurry of changes in legislation as governments began to recognize the need for tighter automobile safety regulations.

In Europe, the FIA began studying government crash tests to learn how to save racing car drivers' lives after the death of Ayrton Senna, the Formula One world champion, in 1994.

''We decided to look at road cars,'' said Max Mosley, president of the FIA, ''because all the governments had been working on this and we thought there must be a great deal we can learn from it. We discovered to our horror that there had been absolutely nothing since 1974.''

The New Car Assessment Program has existed in the United States since the late 1970s, and in Australia since the late 1980s. Japan started a program in the mid-1990s, but Europe has lagged.

Unfortunately real car crashes happen in myriad ways that no number of tests can mimic, which makes some critics say they are a waste of time.

''There are two things you can do,'' said Steve Mitchell, an engineer at Kayser-Threde, a German company that makes the industry's most popular crash-test black box.

''You can say there's no way I can know what kind of crash Mr. X is going to have, at what speed, what angle, what he's going to hit, or what is going to hit him. And you can throw your hands in the air and say it's all too difficult and do nothing. Or you can say, we understand that this may not be representing Mr. X's crash, but it's a start.''

BRAD SPURGEON is on the staff of the International Herald Tribune.

Back to Samples Index