Nothing better illustrates the encroachment of data science and analytics on the older “economy of tangible things” than the business of delivering packages.
The use of analytics in package delivery is not new. Companies like UPS and Fedex are longtime users of operations research methods like optimization and simulation to route inter-city shipments, site new depots, allocate shipments among different modes, and simulate capacity utilization.
UPS’s Jack Levis championed ORION, “On-Road Integrated Optimization and Navigation,” the company’s ambitious and ultimately successful software project that facilitates optimal planning of individual driver routes. Routings that, at the moment, seem counter-intuitive to drivers (e.g. delivering to house #1 in an area but skipping houses 2 and 3 in the same neighborhood and coming back to them later) are suggested by the software in the service of optimizing the total day’s tasks. Levis, Director of Process Management, is active in the analytics community and has a TED talk recounting the travails he and ORION went through before getting it accepted and implemented at UPS.
UPS has a lead in the analytics/logistics race owing to its “embedded knowledge” in the form of accumulated detailed driver wisdom that can be leveraged for analytic gain, if it can be extracted in forms useful for modeling. But Amazon is trying to catch up, throwing considerable resources into all aspects of the delivery chain, to gain knowledge, experience and data quickly.
Amazon uses traditional operations research analytics, as well as more recent predictive modeling methods to go beyond optimizing a defined logistical task to actually predict what you might order in the future, and pre-position goods for faster shipping. It has made headlines with its drone delivery pilot and its acquisition of delivery vehicles, from the 767 cargo plane that recently crashed in Texas to the fleet of rented vans that carry packages the “last mile” to residential recipients.
Through its “Amazon Hub” locker services for apartment buildings, the company joins other concierge services, with a difference: Amazon itself will be the biggest shipper to its hubs, and it will be in the best position to make use of the shipping data (both Amazon and non-Amazon) generated at the locker centers.