Analytics Success Story: UPS’s ORION Project
Arguably, as one of the largest prescriptive analytics and operations research projects in the world to date, UPS’s On-Road Integrated Optimization and Navigation (ORION), has become an iconic figure for exemplifying the most successful analytics applications. Using a variety of inputs (most of which originating from its fleet telematics) and advanced analytics algorithms, ORION calculates optimal routing for its drivers. Because of its tremendous success, in 2016, UPS’s ORION project won the prestigious Franz Edelman Award for Achievement in Operations Research and the Management Sciences, presented annually by INFORMS, which recognizes excellence in the execution of prescriptive analytics and operations research projects on the organizational level. (Since its inception in 1971, cumulative benefits from Edelman finalist projects well exceeded $250 billion.)
United Parcel Service, or UPS, one of the leading logistics providers in the world, has been competing in a highly competitive and rapidly changing global business environment. Securing and sustaining success with a compelling competitive edge in such an environment requires a persistent and relentless pursuit for perfection by inventing and innovating new and improved business processes and practices.
The success of UPS can partially be attributed to its long-standing culture of “constructive dissatisfaction,” which is credited to UPS founder, Jim Casey. It is the belief that companies and people should always be looking for ways to improve themselves. As is the case for most successful companies in the logistics business, UPS is committed to continuous improvement through investments in technologies—investing approximately $1 billion a year in operational efficiency and customer solution projects. In the case of ORION, the company not only had to invest in technologies to develop the desired solution, but get creative in the way it used leading-edge predictive and prescriptive analytics.
Just to create a context of the problem space, consider this: for any given business day, each UPS driver makes an average of about 120 delivery stops (Rosenbush and Stevens, 2015). The number of route combinations a driver can make is nearly infinite, a far greater number than the nanoseconds the earth has existed. Identifying the most efficient route, especially after considering variables such as special delivery times, road regulations, and the existence of private roads that don’t appear on a map, is a near-impossible endeavor for a human being. The ORION project was initiated to take on this seemingly impossible-to-solve optimization problem. With the goal to ensure that UPS drivers use the most optimized delivery routes in regard to distance, fuel, and time, UPS developed ORION.
Motivated by a perfectionist view, the ORION project is the result of a long-term operational technology investment and commitment by UPS. ORION was more than a decade in the making from the initial development of the algorithm to full deployment to nearly 55,000 routes in North America. 2013 marked the first major ORION deployment by a team of 500 dedicated resources to roll out ORION to 10,000 UPS routes. As results exceeded expectations, UPS sped up the U.S. deployment and completed it by the fall of 2016.
Development of ORION
Optimal solutions that rely on analytics need data—rich, timely, and accurate data. In 2008, UPS deployed its telematics technologies on delivery trucks to gather all kinds of transactional and locational data to understand where efficiencies can be improved. By installing GPS tracking equipment and vehicle sensors, combined with a driver’s handheld wireless mobile devices, UPS started to capture data related to traveled routes, amount of time vehicles idled, and even whether drivers were wearing seatbelts (Peterson, 2018).
Successful implementation of the real-time data-gathering modules set the stage for the development of ORION, which consists of a number of advanced analytics modules—based on optimization and other prescriptive analytics models—that could quickly and optimally solve seemingly unsolvable, tremendously complex routing problems. The resulting algorithm in ORION includes about 1,000 pages of code and turns the captured real-time data into easy-to-follow instructions for drivers to optimize their routes. The ORION algorithm was initially developed in a lab and tested at various UPS sites from 2003 to 2009. The company prototyped ORION at eight sites between 2010 and 2011 and deployed it to six beta sites in 2012. The final systemwide deployment of the project was in 2016.
Today, ORION can solve an individual route in seconds and is constantly running in the background evaluating routes before drivers even leave the facility. This level of route evaluation conducted through the ORION program requires extensive hardware and architectural provisions. Running on a bank of servers in Mahwah, New Jersey, ORION is constantly evaluating the best way for a route to run based on real-time information. While most of America is sleeping, ORION is solving tens of thousands of route optimizations per minute. In addition to architectural enhancements, the driver’s delivery information acquisition device (DIAD) is enhanced to serve as the tool for communicating optimized routes to drivers while on the road (Paterson, 2018).
Costing $250 million to build and deploy, ORION is expected to save UPS $300 to $400 million annually. By building efficient routes and reducing the miles driven and fuel consumption, ORION contributes to UPS’s sustainability efforts by reducing 100,000 metric tons of greenhouse gas emissions.
The success UPS has seen as a result of implementing ORION is the ramification of a decade-long effort that has gone into its development. The company has already seen an average daily reduction of six to eight miles per route for drivers who are using ORION routes. Once fully deployed systemwide, ORION will save UPS about 100 million miles per year. That’s a reduction of 10 million gallons of fuel consumed. It also reduces carbon dioxide emissions by about 100,000 metric tons. Initial results show miles reduced with each route using ORION; a reduction of just one mile per driver per day over one year can save UPS up to $50 million.
ORION also benefits customers because it enables more personalized services, even on peak business days. This includes the UPS My Choice service, which allows consumers to have online and mobile access to see their incoming UPS home deliveries and enables them to actively choose delivery preferences, reroute shipments, and adjust delivery locations and dates as needed. Currently, millions of customers take advantage of the UPS My Choice service, and ORION technologies will continue to make possible even more personalized services, with international service on the future roadmap (Peterson, 2018).
Analytics has become the chief enabler for modern-day businesses. With a series of innovative analytics projects in the past several decades, UPS has expanded its intelligent decision-making capabilities by using rich data sources (Big Data that comes from GPS devices, vehicle sensors, and driver handhelds as well as the transactional data that come from business practices) along with advanced modeling techniques—from descriptive to predictive to prescriptive analytics.