Forecasts are always wrong! So why should we bother?
Even APICS says so: Forecasts can never be 100% correct. (Unless you are really lucky, but that wouldn’t last forever, would it?) Might that be the reason why companies are often confronted with poor forecasts – because they accept the idea that they can never be correct anyway? Read further to find out what the operational impact of poor forecasts can be – and what aspects you can work on to improve those forecasts.
An operational forecast aims to make statements about future demand in the right level of detail such that it can drive the production and distribution planning and the inventory levels in the supply chain. In the end it comes down to improving the company’s ability to fulfil the customer needs on time. And of course to do that in a cost-effective way.
Since forecasts serve as an input to production planning, a poor forecast will lead to an inaccurate production-planning and production schedule. And there the effects become clear:
- If your forecast is too high you will produce more than the demand. Your finished goods stock will build up, and in the longer term, warehousing costs may start to rise. Besides the stock increase, your people and machines will be deployed in producing more than required. Capacity is misused, productivity drops, and from there your operating costs start to go in the wrong direction.
- If your forecast is too low, your production will not satisfy the demand. Sometimes, this can still be covered by safety stocks. But safety stock costs money as well! If safety stocks cannot make up for the production deficit, you’ll end up with out-of-stocks and expedited or late deliveries. And there you go with your freight costs rising and your service levels dropping.
- In both cases you can still try to balance by changing your production schedules to react better to the demand. The risk is of getting stuck in continuous firefighting mode, depending on how flexible your workforce and your machines are. But you’ll soon end up with either idle time or overtime. Besides, if you want to adjust your production schedule upwards, you need to make sure you have enough raw materials and packaging materials available. Now, how is the procurement of these materials steered? Right – by the forecast!
Okay, so you should bother about your forecast performance! Because a better forecast leads to lower operating costs, lower stock levels, and higher service levels. But how do you make sure you have a good forecast? How can you improve it?
Forecasting is a business function. Every business function needs a clear process definition. The forecasting process needs to be embedded in your organization with clearly defined roles and responsibilities. Do you need a centralized organization? Is the set-up of your forecasting process aligned with your network structure and organizational structure? How about your product, market and customer landscape? Have you taken those into account? And by the way, someone needs to be accountable for that forecast!
Then comes the more tangible part. Data. You need to store data, harmonize it, interpret it, and know how to use it to your forecast’s advantage. Some kind of tool is necessary to use that data and get a forecast out of it. And a spreadsheet is not a tool. Of course you don’t necessarily need the most complex tool on the market. It depends on what your business needs. Then when you finally got that forecast, you’ll need to measure it! If you don’t know how good or bad your forecast is, how will you be able to improve on it?
At bluecrux we are convinced that a good forecast leads to multiple operational benefits. Within our Supply-Chain Planning value stream, forecasting is a key point. We can help you improve your forecasts by analyzing, designing and implementing the above-mentioned dimensions. For more information, please contact us.