IE 11 Not Supported

For optimal browsing, we recommend Chrome, Firefox or Safari browsers.

Remote Control

City agencies look to onboard monitoring and data analysis systems to lower fleet costs and keep vehicles in service.

When that ominous "check engine" light appears on a vehicle's dashboard, not everyone rushes to a mechanic.

Police officers, for instance, are likely to keep driving, said John Menke, street and equipment maintenance superintendent for Glendora, Calif. "The cops never want to give up their favorite cars."

But that can spell trouble -- neglecting the problem that sets off the light today may lead to a breakdown and expensive repair tomorrow. Hoping to catch problems while they're small, some government agencies are embracing technology that remotely monitors engine performance.


Eyes on Performance
In 2004, Glendora installed Networkcar's Networkfleet monitoring system in 47 vehicles -- about one-third of the city's fleet. Elsewhere, officials at the Metro St. Louis transit system, previously the Bi-State Development Agency, are working with developers at Accenture to detect trouble in bus engines and prevent problems before expensive repairs sideline buses.

Glendora uses the software to diagnose engine performance, fulfill state emissions testing requirements and track vehicles operated by several city departments. The system's onboard equipment includes a small device that plugs into the vehicle's computer to obtain data from engine sensors, and the company also provides GPS technology to track vehicles.

The system uses a mobile data network to transmit vehicle data to the company's operations center. Fleet managers access a Web site to view information about their vehicles.

Users can view data about engine performance, monitor driver activities, such as travel speeds and time spent idling, determine when the car was running or stopped, and trace the travel path of any vehicle in the fleet. Managers receive e-mail alerts in certain cases -- for example, to tell them the check engine light has come on or the manufacturer has issued a vehicle recall.

Menke said he's surprised at how many automotive problems the system uncovers.

"On a daily basis, I get at least one e-mail from Networkcar saying, 'There's something wrong on this car,'" he elaborated. "Most of the time these things go completely unreported."


Continuous Smog Checks
The software also allows Glendora to participate in California's Continuous Testing Pilot program -- it continuously measures each vehicle's emissions and sends an alarm if the emissions rise above allowable levels. As a result, Glendora officials don't need to take vehicles equipped with the software to a service station every two years for a state-mandated smog check.

That saves the city money, Menke said.

"Your typical smog check is around $70, then you've got the employee time to take it to the smog check, plus all the paperwork that goes along with that," he explained. "When all is said and done, you probably spend a couple hundred dollars."

The system's vehicle-tracking function makes it easier for dispatchers to locate police officers and helps keep better control over other mobile workers. Because employees know managers are watching their activities, they will more likely put in a full day's work, Menke said. "It's helping to keep an honest person honest."

It cost the city about $500 and 45 minutes in installation time to equip each vehicle for the service, Menke said, and the service currently costs $20 per vehicle per month.

Because the city hasn't done a lot of cost tracking, it's hard to compute how much Glendora has saved by using the software, but the system is paying for itself, Menke said.

"We're saving money not having to take vehicles in for smog checks, and we save money if we bring in something for repair when we've detected that a vehicle needs service, rather than waiting."

The city also realizes soft benefits in the form of better employee performance, and Glendora plans to add more units as money becomes available, until the city equips most of the fleet, Menke said. The exceptions will be older vehicles lacking the electronic components needed to connect with the system.


One Step Further
Technologies for monitoring engine performance remotely aren't new.

The Cummins engines in Metro St. Louis' buses already come linked to a satellite communications system. Cummins uses this to monitor performance and contacts the transit agency when onboard sensors detect trouble.

Accenture and Metro St. Louis are working to take the concept further, developing a predictive monitoring system that not only spots existing problems, but also forecasts future equipment failures. The system also uses data about the buses to develop individually tailored maintenance schedules.

The consulting firm developed similar predictive systems for the nuclear power and aviation industries.

"Accenture tells us we are the first in the transit industry to do this," said Tom Dutton, director of IT operations at Metro St. Louis.

When it started working with the company, Metro St. Louis was already benefiting from a new preventive maintenance program. Bringing buses in for service on a regular schedule helped keep each bus on the road more than 90 percent of the time, compared with 77 percent in the past, Dutton said.

But when you operate both old and new buses, some mainly on the highway and some mainly on city streets, a one-size-fits-all maintenance program carries hidden costs. Not every bus needs its oil changed every 3,000 miles, Dutton said, and Metro St. Louis wanted to improve vehicle uptime and save money by performing maintenance exactly when required -- not too soon and not too late.

Early this year, Metro St. Louis equipped 20 buses in its 600-vehicle fleet for the Accenture system. Each bus received a small computer, which takes an electronic feed from the engine and periodically transmits that data via satellite to Accenture's data center in Chicago.

There, the data is incorporated into a computer model that describes each bus's normal performance. When the system receives a reading from a bus that doesn't match its profile, it posts information about the aberration on a Web site.

"The mechanics and maintenance people look at it and see if they can determine what the causes might be," Dutton said.

If Metro St. Louis rolls out the system after the six-month pilot, the company will add alerts to supervisors about urgent problems, Dutton said. "If it's in the middle of the summer and a bus starts to run excessively hot, that may be something they want to deal with that day instead of waiting until it comes back at the end of the shift."

Alerts could be delivered via e-mail, cell phone or pager, he said.


More Data, Better Forecasts
The more data Accenture collects, the better it can predict when buses' systems are likely to fail and the more precisely it can tailor a maintenance program for each vehicle, Dutton said. When the transit agency catches a problem before a bus breaks down on the road, it saves money and increases customer satisfaction, he said.

It can also save money by creating parts replacement schedules based on actual performance. "If I can save a radiator from being replaced every 100,000 miles, and instead do it every 120,000 miles, I can avoid replacing 600 radiators in the fleet over a year's time," Dutton said, adding that doing so could save the city $250,000.

Evaluating the system's performance so far, Accenture and Metro St. Louis are trying to calculate the system's return on investment, Dutton said, by examining how much a radiator costs or how much an oil change costs and also tracking costs the agency avoided by catching problems early or postponing maintenance.

Metro St. Louis will then evaluate the business case for implementing the system. The agency is also exploring less expensive alternatives to satellite communications for transmitting data from the buses.

"Overall, it's working very well," Dutton said, adding that maintenance supervisors and mechanics believe the data is timely and accurate. "We need to introduce more variables and make the model more robust, but we've definitely come upon something we think can add value to the fleet."