Cut the Uncertainty Tax
The Way to Lower Taxes is Better Math
Roughly half the global inventory is a result of demand uncertainty and market volatility. The net result is what is essentially a huge “uncertainty tax”. Despite advances in technology, fundamental forecasting methods used by manufactures have not changed since the 1950s. Conventional forecasting methods based on historic sales are ill-equipped to predict the fast-changing and increasingly volatile world. For example, historic sales cannot account for economic events like the current concerns over the Euro or natural disasters like hurricanes or tsunamis.
As a result, demand forecasts are woefully wrong. In a recent study encompassing one third of the consumer goods trade in North America, forecast error was close to 50%. This means that if you forecast ten thousand cases of a particular product will sell this week, half of the time the real number is between 5,000 and 15,000 and half the time it is outside of that range.
Not to disappoint consumers, companies build excess inventory to compensate for their inability to forecast. Poor forecasts essentially equate to an uncertainty tax in the form of excess inventory. A tax with high financial, carbon and water costs. This places an enormous burden on a system that is already stressed serving our current population, let alone a projected 3 billion additional people in the next 40 years.
To reduce this uncertainty tax, what’s called for is a new breed of mathematics to sense changing demand patterns and provide better forecasts. In recent years, the industry has experienced an explosion of data in the supply chain and the growth of low-cost computing capabilities following Moore’s Law. This created the opportunity for new software that uses modern mathematics to automate the analysis of Internet-scale datasets. More importantly, to repeat this process every day, so that forecasts accurately reflect the realities of the supply chain and dynamically adjust to market conditions and volatility. By sensing demand, companies consistently reduce forecast error by 40% and are able to cut inventory levels by 15% or more.