Predicting consumer demand is big business. A robust industry of analysts and algorithms attempts to nail down who will buy how much of something in the most granular manner possible. Similar to someone who can count cards in blackjack, Las Vegas casinos spend hundreds of millions of dollars monitoring such behavior in an attempt to minimize their losses. In the business world, there is no entity monitoring those who can accurately predict consumer behavior. Every now and again they hit a hot streak, but in the aggregate, supply chains are so complex that any short-term gains are nearly always leveled out over time.
Forecasting is based on historic trends. For example, customers in segment X bought this amount, and Y bought another. When demand is steady, supply chains behave reasonably predictably. But in a pandemic, as we all witnessed, demand for certain items spikes. While supply chains did a decent job keeping pacing (not in all instances), some of the spikes were misinterpreted as fundamental shifts.
When spikes occur, a retailer might increase orders the following month which signals to the product planner in the manufacturing firm to increase production on said product. The time between when the retailer signals to the manufacturer a demand shift and when the product makes it back down to the retailer’s hands takes time. In between, a retailer might see demand quickly dry up (toilet paper or exercise bikes in the pandemic) and this produces what the industry refers to as the “bullwhip effect” – think of a crack of a bullwhip along a chain and the immediate interruption in demand. The desire to ride the demand wave is intense (and profitable), but shifts in consumer demand are lightening quick and chains aren’t nearly as agile.
Speaking of time, “just-in-time” production refers to the practice of producing only what buyers need when they need it. Toyota was an early adopter of “just-in-time” production, capitalizing on its proximity to suppliers and factories. When you can schedule the bulk of deliveries to be delivered within a day or two, the firm benefits from less money tied up in inventory. China is still famous for this model within its industrial parks.
Yet, we now live in a world with far-flung global networks. A big reason why there’s been a surge of auto parts being shipped via air is that pandemic-related demand threw the forecasts out of whack. The cost is prohibitive but had to be done because parts weren’t arriving on time otherwise. As such, firms are slowly moving from “just-in-time” to “just-in-case” production which naturally carries more inventory
Another supply chain characteristic often overlooked (or not well-communicated) by the public is the layers. McKinsey estimates that a typical auto-maker counts on 250 tier-one suppliers, an incomprehensible 18,000 suppliers, and 30,000 plus parts. Even a smartphone, for example, can have between 3,000 to 4,000 parts. The tier-one firms supply companies directly, whereas the tier three firms supply the tier two ones, and so on and so on. A large firm likely doesn’t know who its suppliers are beyond tier two.
At the beginning of the pandemic, depending on the industry, many suppliers in tiers three and below shut down. The manufacturer, however, would never know about a shut down until their tier one or two firms could not fulfill a delivery. If you’re assembling a car, one or two critical parts cannot simply be replaced. During the pandemic this caused (and continues to) delays where on the product side companies ran out of parts, and on the distribution side, labor shortages and congested ports exasperated the situation.
So what’s the answer? If most firms begin to adopt “just-in-case” production coupled with reasonable caution, that should smooth the big bumps. The problem is it’s natural to chase profits. Counting blackjack cards is an attractive proposition because when you hit, it’s a hit!
Credit: Source link