Of all the data in the world, 90% of it was created in the last two years, according to IBM. Supply chains, by their nature, are particularly abundant with data. Reefer sensors relay temperature levels for cold chain shippers and carriers, electronic logging devices (ELDs) automatically track truck drivers' hours and warehouse management systems (WMS) measure and analyze a number of components within a facility.
Analytics turns this vast amount of information into insight, according to the 2020 Third-Party Logistics Study by Infosys Consulting, Penn State University and Penske Logistics presented at the CSCMP Edge conference in Anaheim, California. And with this insight, "you stand a much better chance of improving your operations," John Langley, professor of Supply Chain Management at Penn State University, and one of the study authors, told Supply Chain Dive.
In some aspects, the logistics industry is already leveraging real-time data and analytics. Langley named dynamic pricing in freight as an example, where real-time information helps to match supply with demand. But he said data and analytics can help to solve a far greater range of common shipper challenges.
In the 3PL study, a majority of shippers agreed the use of analytics could improve tracking challenging metrics and KPIs such as on-time and in-full (OTIF), freight costs per shipment and order-to-delivery cycle times.
|Type of problem||% of shippers who said analytics would be helpful|
|On-time and complete order fulfillment||69%|
|Freight costs per shipment||60%|
|Cost to serve||58%|
|Order-to-delivery cycle time||58%|
2020 Third-Party Logistics study
Langley said analytics have "great applicability" for a KPI like OTIF because of its nature as a compound metric. "When there is failure in the compound measure, to what extent is this lack of being on time or lack of completeness?" he said.
OTIF, popularized by Walmart but certainly not unique to the retail giant, has challenged many shippers to adjust aspects of their operations, such as inventory management and on-time delivery metrics, to ensure purchase orders arrived to retailers in full and when they were promised. Monitoring such a metric often demands a data-driven approach.
Shippers may learn the most from real-time data on shipments that did not meet either the on-time or the in-full requirements, Langley said. Data and analytics can help determine cause and effect relationships to find the root of a service failure, distinguishing causation from correlation or coincidence, according to Langley. "If you can measure it, capture it, analyze it, you can use it to your advantage in terms of knowing more about your own processes," he said.
Various types of analytics deliver different insights. Diagnostic analytics using real-time data can identify service failures, Langley said, and more advanced predictive analytics can anticipate the likelihood of the failure occurring again.
Currently, few shippers use highly advanced analytics technologies such as artificial intelligence (AI). Only 16% in the study said they use cognitive analytics or AI for planning or operations. On the other end of the spectrum, 78% use descriptive analytics, and 69% use diagnostics analytics.
Langley compared jumping into prescriptive analytics or AI without first using diagnostic and descriptive analytics as going down the black diamond ski slope before the green circle ones. The less mature forms of analytics "tend to fuel the progress made in the more sophisticated approaches," he said.
In the report, Langley outlines a series of steps to develop an analytics strategy, including identifying problems and objectives, clean data collection and feedback and continuous improvement. He sees many more shippers adopting mature forms of analytics over the next three to five years. "It's just exploding in terms of interest," Langley said.