Dive Brief:

• C.H. Robinson has partnered with Microsoft to provide increased inventory visibility for shippers using its Navisphere platform, the companies announced Tuesday. The two companies created Navisphere Vision, which uses Azure IoT technology, machine learning and predictive analytics to forecast potential disruptions. Microsoft has also begun using the platform within its supply chain.
• "Our platform is providing visibility to all shipments in the supply chain, whether it's sitting or in transit," Jordan Kass, the president of managed services at C.H. Robinson, said in an interview with Supply Chain Dive. The inventory data pairs with outside sources of information, including weather and traffic, and the system will use predictive analytics in an attempt to anticipate supply chain disruptions, Kass said.
• Suppliers can connect to the system either directly or through the cloud and send in purchase orders to the Navisphere system as they come in, which will begin the process of tracking the inventory. On the outbound side, customer orders will be sent to C.H. Robinson's ERP system and picked up directly by Navisphere, which will begin the process of route optimization and carrier selection management.

Dive Insight:

The partnership with Microsoft will allow shippers to connect to the capabilities included in the Azure ecosystem, which includes IoT technology. C.H. Robinson will own a pool of IoT devices and customers can contract with the company to use them for shipments.

C.H. Robinson has been able to track the physical location of its shipments using Navisphere for a while, but the addition of IoT technology will allow shippers to see additional variables such as temperature, shock, humidity, light and pressure of the shipment.

"It's, in the case of Microsoft, monitoring tilt," Kass said. "So for them think about things like servers and the idea that you want them loaded to ride in an upright position."

If the IoT device detects one of these variables slipping into a range that is considered dangerous for the shipment, the system can send an alert in real-time to the shipper.

"You can set thresholds for any one of these dimensions when a threshold is crossed it will proactively alert any of the users," he said.

Shippers can work with C.H. Robinson to configure alerts specifically for their supply chain. And the system allows them to choose the frequency of the alerts (monthly report, real-time, etc.) along with where the alerts will be sent (text, email, etc.).

The master view for a shipper will show all of its inventory across the globe and the system will highlight segments that could be late or at risk.

The system will use machine learning to determine if any of the shipments are at risk by pairing the location and IoT data with the weather and traffic data. The data C.H. Robinson collects on the shipments will be used to train its machine learning algorithm in hopes of being able to better predict supply chain disruption in the future, Kass said.

But not all supply chain disruptions can be avoided with inventory visibility or weather or traffic data. A recent survey found the majority of shippers consider the lack of clarity on consumer demand to be the biggest supply chain bottleneck they are seeing as a result of the COVID-19 pandemic. Many models for consumer demand have had to be thrown out because making predictions based on historical data in the current environment doesn't make much sense.

"It's not just say, an increasing trend; it's not just fairly predictable seasonal lifts," John Aloysius, a professor in the supply chain management department at the University of Arkansas, told Supply Chain Dive last month. "You know, everything's just like a completely different story and the historic data is no longer as relevant."

Laws surrounding data are different from country to country and not all of it can be archived and store in the system. "If you're in Europe, and you do a home delivery ... you're not going to be able to store that home address, so things like that are going to be removed from the system," Kass said. "The remaining data elements will be and that obviously is what is feeding and teaching and training the computer itself."