New Inventory Problems Expose Old Supply Chain Weaknesses

How did companies find themselves in this inventory mess?

Several factors caused these problems:

• Delays in fulfillment from Asian factories due to congestion in ports and other parts of the transportation system meant that many retailers got stuck with items that could not be sold because they missed their selling season.
• As supply chains became more volatile and uncertain, companies ordered more of everything “just to be sure.” This practice is a variation of the consumer hoarding habits that became evident during the height of the pandemic. As a result, when deliveries arrived, inventories ballooned.
• Consumers changed their preferences. In the face of rising inflation and the fear of a coming recession, consumers started buying less branded products and focused on value instead. Since this shift was unexpected at the time orders were placed, it landed companies with the wrong inventory.

Some of the drivers behind these changes — notably the fallout from the Covid-19 pandemic and Russia’s invasion of Ukraine — could not be predicted, and the scale and duration of the resultant disruptions were unprecedented. However, some of the causes are well-known and point to ways in which similar situations in the future might be mitigated.

Emergency measures for accommodating excess inventory

A recent article in the Wall Street Journal highlights the predicament companies now find themselves in. As warehouse space across the US has become scarce, some enterprises are storing goods in parking lots and on truck trailers as they desperately seek places to park excess inventory.

As the Journal reports, while the strategy may bring some short-term relief, it can exacerbate the overall supply chain situation. Trailers laden with goods cannot be used to transport products moving through supply chains, further hindering goods flows already hampered by other blockages.

Meanwhile, retailers such as Macy’s, Kohl's, and Target are using age-old practices such as heavy price discounts in stores to move surplus inventory off their books. Another tried and tested approach is to send surfeit stock to off-price retailers such as Ross Stores. The results are diminished margins.

Demand roller-coaster driven by disruptions

As explained earlier, shocks caused by the Covid-19 pandemic and other global disruptions are partly to blame for the inventory mountain that companies are now struggling to contain. These systematic shocks triggered huge fluctuations in demand and supply and hence the re-emergence of the bullwhip effect (for more on this see my blog post It’s Time to Confront the Prospect of a Global Recession).

Related: Confronting the Prospect of a Global Recession

But some causes are very familiar to supply chain professionals. As the Wall Street Journal reports, the inventory overflow strategy “is the latest sign of how retailers and manufacturers are continuing to reset their distribution operations on the fly to keep supply chains running amid disruption in transportation networks and difficulties in forecasting demand.”

Forecasting demand has always been an inexact science. Accurately predicting the capricious buying habits of consumers is a monumental task. Adjusting the supply, especially when performed for highly complex global supply chains attuned to just-in-time and make-to-order strategies, is even harder.

Artificial Intelligence and Machine Learning

Admittedly, companies have gotten much better at learning to take advantage of demand variability. Postponement (delaying final assembly of a product as late as possible so demand forecasts are more accurate) is an example of a technique that tempers the inaccuracies of demand forecasting. More recently, artificial intelligence and machine learning (AI and ML) are being deployed to analyze historic buying patterns and improve the accuracy of sales projections.

However, all statistical forecasting models are based on the past. When disruptions are large enough to introduce a structural change in demand no statistical model, whether it is based on ML or some other methodology, can provide a good forecast. In fact, during the pandemic, many companies resorted to manual order-setting as automated algorithms were making wrong order decisions.

The sheer speed at which demand patterns shifted over the last couple of years is another spoiler. Even if they were able to foresee demand with precision, companies find it difficult to pivot instantly in response to lightning-fast switches in buying behavior.

Impaired Vision—Lack of Visibility in Upstream Supply Chain

Sometimes supply is even more variable than demand, a phenomenon that companies have had to deal with recently and which hampers their ability to manage inventory effectively. In this situation, many companies are caught unawares when a shipment is late or does not turn up. In the short term, this leads to failed customer service or, at best, to expedited deliveries at a much higher cost. In the long run, the results are higher safety stocks and, again, higher costs.

The one defense against supply variability is visibility into the upstream supply chain. Unfortunately, while achieving perfect end-to-end visibility is something of an industry Holy Grail, almost no company can claim it has conquered this prodigious challenge. The main problem is that while companies can force their immediate (Tier 1) suppliers to share data, supply chains are usually multi-tiered.

Companies do not know who the deep-tier suppliers are (for example, a Tier 1 supplier would not reveal to the OEM who the Tier 2 suppliers are since it is a trade secret and also assurance against the OEM bypassing the Tier 1 supplier). And even if they know the identity of remote suppliers, OEMs have no leverage since they have no relationship with these distant enterprises. Such ignorance obstructs the flow of timely, accurate product information that is essential to 20/20, end-to-end visibility.

The good news is that substantial improvements have been made over recent years. For example, the introduction of sophisticated sensing technology has significantly improved shipment visibility. However, as helpful as this is, it’s only a partial solution that covers the last stage of the supply chain — delivery to the OEM or retailer.

Causes for optimism and learning from uncertainty

Still, developing inventory management practices capable of averting the excesses currently experienced by companies is not a lost cause.

Enterprises are learning much about doing business in extraordinary times, especially in relation to managing the supply chains that support world commerce. As shipments fail to arrive in a timely fashion, companies can pinpoint where the failures are and take remedial action. Some companies bypass congested ports or shift to other transportation modes such as air for critically important parts.

Even a failure of a deep-tier supplier at least means that the OEM or retailer can identify that broken link. And as companies accumulate more data on the impact of severe disruptions on supply and demand patterns, advanced data science models, including the use of AI and ML, will continue to improve.

It is unlikely that companies can eliminate the extreme variability in demand and supply that has been so difficult to manage over recent years. But more enterprises are learning how to live with these inconsistencies and even use them to competitive advantage by, for example, installing visibility systems and establishing closer relationships with suppliers and customers. Thus, companies will get better at riding the extreme ebbs and flows in supply and demand that can engorge supply chains with inventory.

Supply Chain 24/7 Education Resource Center
Find the latest educational resources, degrees, and programs.
Visit: Supply Chain 24/7 Education Resource Center

Related: Global Economic Outlook Uncertain due to War in Ukraine