Optimizing Global Supply Chains by Leveraging Risk Metrics: Case Studies

Over the past century, the cost of natural disasters have increased along with their frequency. (Source: EM-DAT). Severe weather events materially impact the global supply chain. For example, in 2000, lightning during thunderstorms caused a small fire in Philips’ chip factory in Albuquerque, NM. The fire was put out within 10 minutes. Downstream in the Phillips supply chain is Ericsson, a Swedish telecom company 4,000 miles away. Ericsson depended on the factory as the sole source of many radio frequency chips. Ericsson experienced a production loss of $400 million and a share price loss of 14 percent—all due to a 10-minute fire.

A more recent example is the Polar Vortex winter of 2013 and the severe cold in Chicago, a major logistics hub for North American shippers. Weeks of severe cold caused rail carriers to cut their train lengths in half, rail switches froze and were inoperable, and rail operations through Chicago suffered major delayed to a point where products froze. Intermodal operations came to a grinding halt. This chokepoint put hundreds of millions of dollars worth of inventory at risk. It also caused stockouts and disappointment during the critical holiday season. Disruptions led to tens-of-millions of dollars in losses for major global shippers such as Unilever, General Mills, ABInBev, Miller Coors, etc.

The extreme winter of 2017-18 in the Northeast U.S. is another example. Multiple Nor’easters stalled logistics in areas where temperatures were far below zero. Tens of millions of dollars worth of temperature-sensitive cargo sat at-risk at the Port of New Jersey, awaiting drayage.

In 2018, Hurricane Florence was the most severe landfalling storm in Wilmington NC since Hurricane Helene in 1958 and had dramatic economic impact on corporate supply chains. 25% of the truck-loads moving into Maryland, Delaware, Virginia, the Carolinas, and Georgia were disrupted and delayed. Our study identified over 2,500 high-risk loads inside the storm representing nearly $100-million in total shipment value. The Port of Norfolk is a major import/export hub handling 7,000 moves per day. A significant percentage of these moves were impacted for two days in advance of the storm even though the port suffered no damage to its infrastructure or operations. Chicken and hog operations in the eastern Carolinas were severely impacted. North Carolina is the second largest producer of hogs and pigs in the US with a total economic impact estimated at $9 billion.

There has been a shift in supply chain risk management practices over the last 5-years. Supply chain managers are moving from "Reactive Management" to "Proactive Planning". Scott Thompson, a contributing writer for Chron.com defines this shift: “Reactive business strategies are those that respond to some unanticipated event only after it occurs, while proactive strategies are designed to anticipate possible challenges.” The keyword to focus on here is “unanticipated.” While every risk to the enterprise cannot be anticipated, there are plenty of events that suit themselves to continual, data-driven planning.

The first step toward a proactive supply chain is to make a transition in how risk is identified. Manual processes frequently miss many of the risks that haven’t historically been a threat and then they also overemphasize other risks that drive media attention but often end-up with "business as usual" outcomes for a supply chain operator. Risks must be assessed upstream and downstream of every node in the supply chain and manual processes simply aren't efficient at this scale. Any process that delays the identification of risks will also delay the mitigation. An automated and dynamic risk assessment/analysis allows for quick response and proactive planning, often before the risk has matured into a real threat. Integrated solutions continually collect real-time supply chain data can be configured to provide notifications of risks across all nodes in the supply chain. This intelligence drives decisions, empowering supply chain managers with information they can immediately use to proactively plan for all types of risks, even when those risks are unfamiliar.

The second step is to focus on the "right" data and to ensure that the data is properly cleaned. Forbes and KPMG found 84 percent of CEOs are concerned about the quality of the data they’re basing their decisions on. Once a shipment is in route, risk mitigation options are limited. Insights into key performance metrics prior to tendering or pickup, at a week or more, give supply chain managers the opportunity to make changes. Multiple modes, lanes, carriers, equipment and more must be modelled simultaneously and must include datasets that reflect operational scenarios, costs, customer relationships, etc. Integrated solutions collect and present data continually and provide reliable risk scoring for multi-modes, including global, multimodal, distribution centers and ports. Risk scoring provides instant visibility into the severity and likelihood of the risk occurring, as well as recommendations for mitigation of each risk. Integrated supply chain risk analysis uses machine learning, artificial intelligence and multi-factor prescriptive analytics to automatically identify risks across the supply chain.

For example, when weather data is combined with operational data (i.e. routing, timing, order-flow, carrier, equipment-type, etc), prescriptive analytics may reveal a simple change in the mode of transportation and may not only ensure the shipment makes it to its destination on time but also at a lower cost. Reefer trucks, for example, are expensive and potentially unnecessary if temperatures along the route have a high likelihood of remaining within tolerance levels.

This presentation focuses on case-studies with economic and operational data from major food and beverage shippers as well as some of the nation's largest road/rail carriers (anonymized to protect confidentiality).