One of largest and most data-intensive companies in the world, United Parcel Service relies on advanced analytics to save hundreds of millions of dollars per year in time, fuel, and maintenance costs – all while improving service and rolling out new features that allow a level of customer self-service that rivals even the back-end access of company managers.
Arguably one of the very first companies to employ analytics to improve operations, over the last 15 years UPS has moved from simply describing what happened yesterday to predicting and prescribing how it will operate tomorrow.
“Everywhere I go, people are chasing big data,” UPS Director of Process Management Jack Levis said at the University of Cincinnati’s Analytics Summit 2014. “To me, what’s the process for that? You better know what to do with it when you catch it. That’s my job. Big data doesn’t mean much to me. I care about big impact.”
At UPS, that impact is huge. If Levis and his team can reduce the distance that the company’s drivers traverse every day by one mile, he said, that equates to a cost savings of $50 million per year. If they can save one minute of drivers’ time, that equals $15 million per year in savings. If they can reduce by one minute the amount time a truck is idling, that saves $500,000 a year.
To achieve these results, UPS has been using cutting-edge technologies since 1990s, when it introduced the first handheld devices used by drivers. Called Delivery Information and Acquisition Devices, or DIADs, these devices form the backbone of the company’s data gathering, package delivery, and route optimization efforts.
Levis calls these devices “hungry animals” because they love to consume data. Today, the company has 80,000 fourth-generation DIADs in the field, allowing drivers to do everything from receiving messages and inputting exceptions to optimizing delivery routes. This information is combined with data from more than 200 sensors on the vehicle itself to give UPS “a very good idea of what yesterday looked like.”
Already using predictive analytics to make sense of what yesterday means for today, Levis said, the company’s next step was to move to a prescriptive model that allows it to react on the fly for exception management as well as roll out new services based on up-to-the-minute information. Predictive analytics has already saved the company 85 million miles per year since 2003 and saved drivers 8 billion entries into their DIADs.
Over the past few years, UPS has been rolling out On Road Integrated Optimization and Navigation (ORION), a route optimization and package delivery analytics platform that allows the company to optimize routes right up to the point when a driver leaves for the day and offer customers new services like guaranteed delivery windows that were impossible just a few years ago.
“We’ve really gone from a trucking company with technology to a technology company with trucks,” said Levis.
Combined with hyper-accurate maps developed and continually updated by UPS and GPS data, ORION not only gets packages where they need to be more efficiently, it helps prevent drivers from delivering packages to the wrong address. This is important given that, on average, drivers have about 120 stops on their routes that must be accounted for every day.
Levis said that an unexpected benefit of ORION has been that, just when the company thought it knew everything there was to know about route optimization and package delivery, the software showed them they were wrong. Because of that, the company’s naysayers realized that ORION was the future of the business.
“I always say that projects have three levels: ridicule, followed by violent opposition, followed by acceptance as if self-evident,” said Levis. “ORION did that. No one believed it for a long time.”
As important as projects that provide reliable data are, Levis said, the right information and the vision of how to proceed is only half the battle.
“There is a difference between vision and execution,” he said. “Vision is hard, but execution is harder. As the leaders in the room, we have to set the vision but we also have to give the folks a chance to execute.”
Now a freelance writer, in a former, not-too-distant life, Allen Bernard was the managing editor of CIOUpdate.com and numerous other technology websites. Since 2000, Allen has written, assigned and edited thousands of articles that focus on intersection of technology and business. As well as content marketing and PR, he now writes for Data Informed.com, Ziff Davis B2B, CIO.com, the Economist Intelligence Unit and other high-quality publications. Originally from the Boston area, Allen now calls Columbus, Ohio, home. He can be reached at 614-937-2316 or firstname.lastname@example.org. Please follow him on Twitter at @allen_bernard1, on Google+ or on Linked In.