Improving Demand Forecasting

Every major company decision, from financial planning to project execution, starts with a prediction of future sales—so demand forecast accuracy matters. Under-estimating demand means running out of product when customer demand is at its highest, costing the company immediate revenue AND hurting your relationship with your customer base. Over-estimating demand means companies have to invest upfront in a lot of extra inventory, which then can’t be quickly turned around into a profit. With inventory typically comprising between 25% and 40% of assets, demand uncertainty is also often the single largest influence on stock levels.

Out-of-stock situations create unnecessary costs.

Manufacturers, when faced with out-of-stock situations because of poor demand forecasting, have to deal with unplanned production changeovers in order to keep up, lose manufacturing capacity, and might incur supply issues for other products as a side-effect. Unfortunately, you can’t make something out of nothing, and when faced with an unplanned surge in demand, you really shouldn’t be stealing from Peter to pay Paul if you want to keep up.

Overstock impacts warehousing operations.

On the flip side, having a ton of extra product in your warehouse isn’t an effective way to manage your supply either. While that may mean you can handle unplanned surges in demand, that inventory is costing your company money as it sits.  Accurate forecasts actually allow companies to unlock capital and free cash that is otherwise tied in inventory.

But forecasting is an uncertain process. Greg White of Blue Ridge Inventory made a great point;

Probability theory tells us that when you flip a coin, the chance of it coming up as “heads” is 50%. So you’d be smart to predict 1,000 heads and 1,000 tails in 2,000 flips.

But we all know from personal experience that in 10 flips it’s possible for a coin to come up heads eight times. And it’s only a little more likely to come up heads exactly five times. This is why statistical forecasting tends to be wrong, especially when the number of occurrences is small.

Statistical demand forecasting systems use statistics and probability theory to predict future demand. They do so by projecting demand forward, based on the history of prior demand. But statistical forecasting methods are blind to the effects of the many factors that may deviate from history.

Manufacturers may not be able to predict the future with 100% accuracy, but they need to get as close to perfect as possible if they want to ensure the right amount of inventory and production rates at all times. With business intelligence software, manufacturers can use a planning application and statistical forecasting engine that leverages all of their operational data to better accommodate seasonal demand, product hierarchies, product promotions, slow-moving items, causal variables, outliers and much more. Many business intelligence systems also come with outlier detection and correction capabilities that give you statistically calculated outliers in historical data to ensure that inputs to the forecasting engine are cleaner and more accurate.