Production demand forecasting: How to do it right?

In manufacturing world, several activities, if not all, are based on forecasting for future demands. For this reason, manufacturers have always treated demand forecasting as one of the most important foundations for accurate, timely and effective production.

In our last post “How ERP revolutionises manufacturers”, we mentioned about a recent survey by Gatepoint, in which manufacturing executives were interviewed and 49% of them see the need to improve demand forecast capability in order to achieve their revenue target.

Why is demand forecasting important?

To maintain competitive position in the market, manufacturers should adapt to fast changing world as well as continuously deliver innovative products and service to customers. Therefore, they need demand forecasting.

Demand forecasting is a part of Enterprise Resource Planning (ERP) system to forecast raw material requirements for production, and predict future customer buying habits to optimise inventory levels while meeting customer expectations. Forecasts based on time series, is one of the most widely used, it is the data passed to make future predictions, the fact is that this method does not consider exogenous variables in the production process with the data demands that occurred in past periods.

It allows manufacturing engineers to anticipate future situations in their planning of production processes and take actions to better meet organisational goals. The actions developed has its startup through forecasts, which directly affects the expected performance of the organisational functional planning and controlling of many processes and systems such as production, sales, finance, supplying and distributing.

Demand forecasting is the primary tool for manufacturers to accurately determine the optimal supply rate and build adequate resources accordingly, henceforth, minimise expenses. Furthermore, it enables the collaboration between outbound and inbound process of the manufacturing process, such as sales and production. For instance, one important activity for any manufacturing firms is relating to production and inventory planning for uncertain market demands. Production departments always need sales and marketing accurate forecast in order to produce enough to meet up with the market’s demand and therefore, balance the supply and demand of the products.

Read more: APS - a key integration for modern manufacturing ERP

The ingredients of accurate demand forecasts

Here are the typical inputs for demand forecasting:

• Historical sales trends: 2-5 year period is used to analyse sales activities, normally.

• Supplier forecasts: know supplier trends to flexibly adjust to any situation.

• Seasonal changes: there are some seasons that sales are higher, or lower than other periods in year, hence manufacturers need these information to develop suitable production plan. Furthermore, other factors such as raw material lifecycle should also be included in the analysis.

• Constraints or business rules: re-examine and re-determine the constraints that impede the manufacturing process, such as warehouse space limitation to consider what level of production is suit best.

Read more: Data Analytics for Manufacturing: the Tesla’s Case Study

The accuracy of forecasts will mostly depend on the accuracy of these inputs – information. Although manufacturers can use different forecasting techniques based on objective methods or subjective methods, there is always a variance between forecasted number and actual number. The higher the variance, the more inaccurate the forecasts.

So, the question is “How to improve forecast accuracy?”

The key to accurate demand forecasting: Improving collaborative forecasting

A collaborative process is well-suited to creating a demand forecast that considers multiple, and sometimes competing factors such as:

• Historical demand, including trends, similar products, and seasonality
• Macro and micro economic trends
• Promotions and advertising
• New product introduction and competitor activities
• Unique insight and judgment of demand and supply chain planning participants

By enhancing collaboration between departments, employees will be provided with relevant data listed above, and through discussions and iterations manufacturers can create forecasts for each product line you sell.

However, most companies use Excel to process data for demand forecasting, but the tool itself is not designed to collect and handle large numbers of data, sharing data amongst numerous participants or tracking individual inputs to the process. This efficiency will eventually force people to make compromises and assumptions in the process and cause critical problems.

Read more: Manufacturers adapt to Industry 4.0

Therefore, improved collaborations by implementing an ERP system that enables both internal and external collaboration will be the key to achieve accurate demand forecasting. Internal collaboration provides any historical information as well as market insights; and external collaboration provides better insights about outside information such as purchase orders, periodic forecasts or forecasts acceptance and negotiation for better demand forecasting. (See table)

Conclusion

All manufacturers need a richer, broader collaboration architecture that incorporates deeper business functions and develop teamwork to the next level, connect both internal and external resource to your inner system and work as a united individual to optimise demand forecasting, and ultimately, succeed in a rapidly changing environment.

Wondering what else you can benefit from having a collaborative working culture and how do get it right? Read our full whitepaper “5 levels of manufacturing collaboration” now to find out more!