APICS 10 Step Forecasting Process

At APICS Milwaukee, we are often asked advice about forecasting. While forecasting is extremely important to effective supply chain management, it's still an imperfect science that is challenging to get right. We recommend the following 10 step process.

03/23/2020 | Marketing | 7

LINK TO FULL ARTICLE

 


 

Breakfast Roundtable Blog

In light of recent developments not allowing us to continue our Breakfast Roundtable discussions in person, we are starting this new blog forum for sharing about today's hottest topics.

Two times per month, we will start the discussion with relevant supply chain information. We encourage you to comment with your thoughts, share best practices, ask questions, etc. 

Looking for certification maintenance points? Every 5 comments will earn you 1 point!!

 

APICS 10 Step Forecasting Process

At APICS Milwaukee, we are often asked advice about forecasting. While forecasting is extremely important to effective supply chain management, it's still an imperfect science that is challenging to get right.

APICS recommends the following 10 step process to assist you with your forecasting efforts.

1.Determine purpose. For example, demand for production, capacity requirements, or staffing levels.

2.Set level of aggregation and units of measure. Specify sales in units or dollars, families, products, or SKUs (stock-keeping units).

3.Select horizon and planning bucket. Horizon, aggregation, and units are interrelated. A long-term horizon (years or quarters) usually uses total sales in dollars; medium-term (quarters or months) uses product families in units; and short-term (weeks or days) uses products or SKUs in units.

4.Gather and visualize the data (chart form). Visualizing helps in selecting the right forecasting technique.

5.Choose the forecasting technique. One or a combination of methods may work best for the combination of purpose, aggregation, time horizon, data availability, trend, and where the data fall on the stable-dynamic continuum.

6.Prepare the data for the technique. If there is seasonality, it should be temporarily removed prior to forecasting.

7.Test the forecast using historical data. Since periods in the past already have actual results, you can forecast using, say, June data to produce a July forecast and compare it to July actuals.

8. Forecast. After making adjustments, start using the forecast. If seasonality was removed for forecasting, add it back in before preparing reports that include confidence levels.

9. Achieve consensus on the forecast. For example, the sales and operations planning process gets agreement on one forecast.

10.Continuously improve. Monitor error levels and set policies for when error levels are too high. When they are, refine communications, data, or processes and techniques.

What process does your company use to forecast? What advice would you have for others on forecasting best practices? 

Engage in the discussion! Every 5 comments will earn you 1 certification maintenance point! 

Read More