Supply Chain Digitization: What COVID-19 Revealed

In late 2021, container ships were sitting offshore outside Los Angeles and Long Beach for weeks. The ports were backed up. Warehouses were full. Truckers and dock workers couldn't process cargo fast enough to clear the backlog. The images of dozens of ships anchored offshore were striking because they made visible something that is usually invisible: the supply chain. Most of the time, goods just arrive. When they stop arriving, people suddenly pay attention to the machinery that normally runs in the background.

The semiconductor shortage that began in 2020 and stretched into 2022 told a similar story, but from the manufacturing side. Automakers that had spent decades refining just-in-time inventory management were operating with days of chip inventory rather than weeks. When chip supply dropped, assembly lines stopped. Ford, GM, and others reportedly had to halt production of high-margin vehicles and leave partially assembled trucks sitting in lots waiting for parts. By most accounts, the automotive industry alone lost billions in revenue during this period. This isn't a story about a sudden shock. It is a story about decades of optimization toward efficiency that left almost no buffer for disruption.

Just-in-time inventory management is rational under stable conditions. You hold less inventory, which means less capital tied up in warehouses, less waste from spoilage or obsolescence, and more predictable cash flow. The assumption, usually implicit, is that supply is reliable. When suppliers are reliable, customers are reasonably predictable, and logistics networks are stable, holding minimum inventory is smart. COVID-19 violated all three of those assumptions simultaneously. Demand shifted abruptly. Suppliers faced their own shutdowns. Logistics networks were disrupted by port closures, labor shortages, and changed routing patterns. The optimization that had been making supply chains more efficient made them less resilient.

What this has to do with information systems is the part that doesn't get enough attention. The discussion in business press focused on diversifying suppliers, reshoring manufacturing, and building buffer inventory. Those are legitimate responses. But the underlying problem that made the disruption so damaging was an information problem. Most organizations didn't know where their inventory was in real time. They had good visibility into their own warehouses. They had limited visibility into their direct suppliers' inventory positions. And they had almost no visibility into their suppliers' suppliers. When a chip shortage develops somewhere in a supply chain two or three tiers back from the final assembler, the information about that shortage travels slowly, by phone call and email, through contract manufacturers and tier-one suppliers, before it reaches the organization that will eventually stop production because of it.

This is a data governance and interoperability problem. The systems exist to track inventory, to monitor supplier order books, to flag risk in logistics networks. But those systems are owned by different organizations with different incentives to share data. A tier-one supplier has some incentive to tell its largest customer that inventory is constrained, because the customer will want to plan ahead. That same tier-one supplier has less incentive to reveal the detailed operational state of its own supply chain, because that information is competitively sensitive. Every link in the chain faces some version of this calculation. The result is that information moves more slowly than goods, and by the time a problem is visible, the lead time to respond is already short.

The concept that has emerged to address this is supply chain visibility. The idea is that organizations should be able to see not just their own inventory position but the position and risk profile of their extended supply chain, multiple tiers back. Some organizations have been building digital platforms to aggregate this information across supplier networks, sometimes through data-sharing agreements, sometimes through third-party risk monitoring services. Digital twins for supply chains, virtual models that can simulate how a disruption at one node propagates through the rest of the network, have been increasingly discussed as a way to make this kind of risk visible before a crisis hits. My read of what's available in industry reporting is that adoption of these tools is still uneven and early-stage in most sectors.

The IS argument here isn't that technology will solve supply chain fragility. Better technology helps, but the structural problem isn't technological. It is that supply chain visibility requires data sharing across organizational boundaries, and data sharing across organizational boundaries requires trust, governance frameworks, and aligned incentives. Two competing suppliers don't naturally want to share inventory positions with each other or with the same customer. Getting information to flow through a complex multi-tier supply chain is as much a governance problem as a technology problem. You have to design who sees what, under what terms, with what protections, and who has authority to act on the information when a risk is flagged.

I think this is why the "digital transformation of supply chains" conversation often stalls after the technology pitch. The technology for real-time inventory tracking, supplier risk monitoring, and logistics visibility exists. The barrier isn't building the systems. It is agreeing on who owns the data, who bears the cost of sharing it, and how the benefits of better visibility get distributed across a supply chain that involves dozens of independent organizations with different sizes, margins, and negotiating positions. A large automaker can mandate that its tier-one suppliers participate in a visibility platform. It has much less leverage over tier-two and tier-three suppliers that don't sell directly to it and may not even know who the final customer is.

COVID-19 didn't create this problem. It made it impossible to ignore.