UPS unpacks big data to deliver insight
"Data must yield insight. If your data doesn't do that, then it's just trivia," said Jack Levis, senior director of Process Management at UPS.
A parcel without an address is just a box, and big data without insight is just used petabytes in a cloud.
Data is largely considered an invaluable asset but if mishandled, it's just more junk in the drawer. To unlock data's potential, it needs effective guidance coupled with tools and direction from a qualified workforce.
The United Parcel Service (UPS) has been creating predictive and prescriptive tools for more than a decade, Jack Levis, senior director of Process Management at UPS, told CIO Dive in an email. As early as 2003, the shipping company launched Package Flow Technologies, which included predictive models, planning, execution and descriptive analysis tools, he said.
About 10 years later, in 2013 On-Road Integrated Optimization and Navigation (ORION) was deployed and added "proprietary advanced optimization to better plan and execute drivers' routes," he said. ORION used the same data foundation of Package Flow Technologies.
Together, the programs have decreased drivers' mileage by 185 million miles a year.
UPS isn't the only company growing big data and business analytics revenue. By 2022, that number will hit about $260 billion with a compound annual growth rate of nearly 12% over 2017, according to IDC's Worldwide Semiannual Big Data and Analytics Spending Guide.
Yet only about half of organizations actually put the data they've collected to use, and only one-third say they completely trust data. If data isn't used or trusted to help make business decisions, what's the point in collecting it?
"Data must yield insight. If your data doesn't do that, then it's just trivia," he said.
It's not all packages tied up in string
UPS' first challenge in handling big data was coming up with how to create, maintain and enhance the applicable data, he explained. "Data is always a bottleneck on these types of projects, and ensuring proper data was available, valid and accurate was an important priority."
The first challenge in handling big data was coming up with how to create, maintain and enhance applicable data. Next was restructuring management.
Restructuring management was the second challenge. The company needed to work toward communicating how new tools and models were to become the new standard and normal for decision-making. "We learned that 'if you build it, don't assume they will come,'" he said. "We educated and communicated over and over."
However, he noted, the challenges of data and the changes they spurred were the most rewarding aspects. The processes have helped cement UPS as one of the leaders of big data and analytics in their domain.
UPS' CIO Juan Perez has his eyes set on an "autonomous everything" future at the company. Such ambitions have lead to the company to experimentation with autonomous vehicles, virtual assistants and drones. The company's embrace of future tech sets employees up for more comfort and familiarity while using the tech.
The use of automation is meant to augment, not replace workers' capabilities for UPS. But just like analytics, the company is conscious of considering employees and customers before delving too deeply into an emerging technology. Perez wants to see the company be "constructively dissatisfied" with the present to maintain its innovative approach to IT.
This is privacy's world and data is just living in it
Using data for business advantage is one thing, but collecting too much personal data without legal collaboration can send a company down a spiral of consumer distrust. Constraints on data collection are unavoidable, now and in the future, according to Levis, and that includes personally identifiable information (PII).
UPS bypasses data collection concerns because it works in a backward fashion by "focus[ing] on the decision that needs to be made," he explained. "After the focus on the decision, the information, data and tools become more apparent."
The shipping company doesn't need to rely too heavily on PII to impact business decisions, according to Levis. This removes much of the headache attached to privacy concerns and leaves room to focus on other initiatives.
Follow Samantha Ann Schwartz on Twitter