At UPS, there’s data everywhere: on the packages that get delivered. On the drivers, who carry handheld computers to record customer interactions. And on those ubiquitous brown trucks.
MORE ON CIO.com
At UPS, New Wireless Networks and Devices Create a More Productive Workforce
Nothing Succeeds Like Succession Planning at UPS
The Perils and Promise of Real-Time Data
It’s only recently, however, that the company found value in connecting those data sets to each other.
Those brown UPS vehicles actually contain a wealth of data drawn from more than 200 sources housed inside the trucks: sensors in the engines gathering data on vehicle speeds, RPMs, oil pressure and engine temperature. In addition, other sensors track the number of times a truck goes in reverse, what doors are open and when, the time the truck spends idling, and how and when the seatbelt is being used. Just to name a few.
“There’s just a slew of data coming out all the time,” says Jack Levis, a manager in UPS’s industrial engineering group. For a while, the company used this mechanical vehicle data to help schedule routine maintenance checks. (See “The Perils and Promise of Real-Time Data” and “How Master Data Management Unified Financial Reporting at Nationwide Insurance” for more on data management.)
Then there is data about customer interactions. UPS is known for religiously tracking all kinds of customer interaction data (captured via the drivers’ handheld DIAD devices) and has accumulated tons of historical data over the years. This includes such things as: addresses that have been delivered to, where and when pick ups have occurred, and any types of customer interactions.
In addition, GPS devices installed in UPS’s fleet of trucks a couple of years ago, which let dispatchers more efficiently route deliveries and helps lost drivers, records precise mapping data, such as street names, addresses and latitudinal and longitudinal information. (To read more on UPS’s GPS rollout, see “New Wireless Networks and Devices Create More Productive Workforce.”)
While certainly vast, these individual data sets, however, had no real connection to each other. “There’s just this pile of data that means nothing to anybody,” Levis says.
But, Levis says, as these things typically happen at UPS, several years ago two UPS employees wondered what automotive and operational insights could be derived from marrying the three disparate data sets.
After months and months of experimentation and research, UPS’s telematics program was born. (“Telematics is a big word,” Levis says, “but it just means a computer is gathering data.”) To the researchers, the telematics application would paint a very detailed picture of a truck’s and its driver’s day together.
How the Telematics System Works
Here’s how it works at UPS: The system relies on off-the-shelf telematics software to help gather and compile the data. At the end of each driver’s shift, all the information is uploaded to a data center in Mahwah, N.J. Then, proprietary applications using in-house developed algorithms allow UPS automotive and operations personnel to query and analyze the data and, ultimately, draw some conclusions about UPS’s vehicle-maintenance and logistics processes.
In 2007, UPS piloted its telematics program on 334 delivery trucks in the Roswell, Ga., and Athens, Ga., areas. Employees in the automotive and operations groups began sifting the data. For starters, analysis of the telematics program data helped cut the amount of time delivery trucks idled by 24 minutes per driver per day, a fuel savings that UPS estimates at $188 per driver, per year. (UPS has more than 90,000 U.S. package drivers, so the potential savings could be huge.) “That adds up to a lot of wasted fuel,” Levis says, “and a lot of carbons being emitted into air that don’t need to be.”
The results made UPS executives take notice. In meetings with executives, Levis says that telematics “is on their radar screen” as a priority. (To read an in-depth profile of how UPS’s current CIO got to where he is today, see “Nothing Succeeds Like Succession Planning.”)
In addition, many of the insights gained from the telematics system have been eye-opening and somewhat counter-intuitive for the engineers in the automotive group. UPS has typically scheduled its fleet’s maintenance by time-dependent factors.
“We’re talking about an alternator or a belt, and we would replace that just because the allotted time had passed,” Levis says. “Telematics started telling us: Don’t replace things that don’t need to be replaced, [which means that] we can reduce preventative-based maintenance and go to condition-based maintenance. That came from just analyzing all that information.”
UPS Rightsizes Repairs, and Cuts Costs
For example, engineers and other “data miners,” as Levis calls them, discovered that UPS was replacing large components and parts on its delivery trucks when telematics showed that what actually needed to be replaced was just, say, an O-ring. “So rather than a thousand-dollar job, it was a $20 or $30 job,” Levis says. “There’s this constant querying of data and finding another nugget. And what happens is that when we find another nugget, then we update the algorithm, and there’s one more way that telematics becomes part of the daily routine.”
The operations analysts comb through the data to see where, for example, UPS delivery personnel may be driving unnecessary miles on their routes, in addition to finding other efficiency patterns and safety trends. Naturally, Levis and his team need to ensure that when they present the telematics analysis to the drivers, they do it in a way that is helpful and educational.
“Nobody wants to hear that a computer is going to tell them what to do or point out every fault they have,” Levis says. “And if you approach it like that you’ll have resistance.”
Levis says as the pilot expands out this year to 10 more areas in the United States and one in Canada, the telematics program will keep on learning and offering up new insights. “We’ve just scratched the surface on finding things,” he says. In the future, Levis can see the program expanding to UPS’s fleet of tractor trailers.
“We have enough now that there’s an excellent business case for telematics,” Levis says. “But there’s patterns we haven’t found in the data, like what’s the impact of tire pressure: Is tire pressure a leading indicator of something else that we haven’t thought about? These are the things we’re doing today.”