A Guide to Demand Forecasting in Supply Chain Management

March 19, 2021

Forecasting demand is tricky even under the best conditions. This week’s hot product may be in the discount bin by next week. Swift changes in customer demand make demand forecasting one of the most significant supply chain challenges professionals face.

The wide-ranging effects of COVID-19 on everything from reduced shipping capacity to sudden shifts in consumer behavior forced many supply chain professionals to throw their playbooks out the window. The pandemic accelerated the need for innovations that have percolated the industry for years. The sudden emphasis placed on certain goods upended typical demand, even in unexpected areas like the glass supply chain, which faced a shortage of glass vials as pharmaceutical companies rushed to prepare vaccines.

Effective demand forecasting requires managers to consider the roles played by vendors, suppliers, transportation providers, and the internal supply chain. It provides the best basis for rationalizing inventory decisions, setting safety stock levels, and preparing supply chains for the next financial cycle.

A qualified demand planner has a host of tools at their disposal. They can use proven techniques to predict the future with a high degree of accuracy—even though forecasts are rarely 100% accurate. Their predictions save companies money and conserve resources, creating more sustainable supply chains.

Below is a condensed guide from the experts at the University of Tennessee’s Global Supply Chain Institute covering what’s most important to know about supply chain forecasting.

What is Demand Forecasting in Supply Chain Management?

Demand forecasting is a method for predicting future demand for a product. The strength of the projection depends on the quality and amount of data, the methods for calculating the projection, and the experience used. Supply chain managers often rely on more than one type of sales forecast for accuracy.

The first step to creating a demand forecast is clearly understanding the most important questions to address. Do you need to understand seasonal demand to optimize inventory levels at different times of the year? Or do you need supply chain analytics to identify unstable links in your supply chain? Historical demand forecast can be a starting point for future demand and forward-looking supply chain operations. However, planning for future events relies on so many more inputs.

Types of Demand Forecasting in Supply Chain Management

Demand planning often relies on historical data about sales to project future demand. Other factors include seasonal demand highs and lows, economic patterns, and market trends. Demand forecasting software helps develop sales forecasts using statistical forecasting.

Demand planning for your supply chain may involve more than one technique. Short-term forecasts deliver a higher level of forecast accuracy. However, depending on the supply chain’s length of cycle, a long-term forecast may be required for inventory management. Forecasting up to a year out is considered short-term; long-term demand planning projects predict demand for up to four years.

Demand forecasts can be internal or external. An external forecast considers the effects of economic and market changes on future demand. When a demand planner creates an external sales forecast, they consider consumer demand trends, spending patterns, and the broader economic outlook.

On the other hand, internal demand planning focuses on the capacity of internal supply chains and personnel. Internal demand forecasting can answer questions about resources available to meet seasonal demand. Both internal and external demand planning is essential for creating an accurate demand forecast and avoiding supply chain challenges.

Demand forecasting can also be divided into passive and active types. Active demand forecasting is dynamic, using predictions about economic trends and the company’s growth plans to create a demand plan. Passive demand planning relies solely on historical data. In some business sectors with products that have consistent demand and a well-established market, passive forecasting may work well. For startups, active demand planning is a must, as historical data won’t be enough to give a clear picture of future demand. However, established businesses use active demand forecasting for effective inventory and supply chain management to increase their market share.

Common methods for creating demand forecasts

Demand forecasting methods can be broken down into two basic categories: quantitative and qualitative.

Quantitative forecasting relies on historical data about customer demand, supply chain performance, seasonal demand, and other data-driven metrics. Qualitative forecasting methods are less tethered to past trends. These methodologies allow the demand planner to add insights from internal and external experts. Both quantitative and qualitative methods have advantages and disadvantages. The best solution for the savvy demand planner may be a mix of both.

Quantitative forecasting often leverages big data. Some common quantitative forecasting methods include:

  • Barometric forecasting. Uses current data to project future demand. Barometric demand planning uses statistical analyses to create a demand forecast.
  • Trend projection. Uses historical data, including growth patterns, to create a sales forecast. While this may be accurate for the near future, it can be risky to rely on sales history without considering other factors, especially for a long-term demand plan.
  • Exponential smoothing. Uses historical data as an input and considers seasonal variations in sales. It’s a useful quantitative method for startups because demand planning with exponential smoothing can be based on a small dataset.
  • Regression analysis. Regressions can range from simple to complex, allowing you to incorporate internal and external data into your analysis.
  • Econometric forecasting. Uses the interplay of data about demand with information on external elements that can move demand. Requires more sophisticated statistical forecasting techniques than other methods but may also create a more accurate demand forecast.

All quantitative demand forecasting methods lend themselves to machine learning planning solutions. Demand planning software saves time and allows companies to update sales forecasts continuously. Software that uses AI to react quickly to market trends can provide vital, up-to-the-minute forecasts that evade supply chain challenges and improve inventory management.

Qualitative forecasting methods are less dependent on data and tend to include more human inputs. A qualitative demand forecasting solution leverages the knowledge base within a company as well as outside experts. Some of the most common qualitative forecasting methods include:

  • Sales force composite. Sales team members interact most with customers, often spotting trends before other sources of market information. Your sales personnel can also access market information about your company’s products and clients. Although it may not be based on data analysis or statistical forecasting, the knowledge base present in your sales force is a valuable resource. This method brings together the sales team members and others involved in creating the demand forecast, leveraging internal experience for demand planning.
  • Market research. This method uses data about market trends and opportunities to create a demand forecast. It is helpful for startups that don’t have access to historical data. Market research forecasting should factor in promotion, expansion, and information about supply chain capacity and limitations.
  • The Delphi Method. The Delphi Method for demand forecasting is sometimes called the expert method. The demand planner assembles a panel of experts to answer a set of questions that reveal the experts’ predictions about future demand. The planner takes the responses, creates a summary, and shares that with a panel. The experts then answer the questions again, considering the summary from the initial set. The demand planner repeats this process until the panel of experts reaches substantial agreement.

How demand forecasting impacts supply chain management

Supply chain management isn’t just about sourcing raw materials or coordinating transport. It reaches into each facet of modern business operations. Demand forecasting is critical to the value supply chain managers add to their organizations.

That could mean identifying a raw material shortage and a supplemental supplier. Or it could involve arranging for additional warehouse space to accommodate seasonal demand. Accurate demand forecasting allows the supply chain manager to set a reorder point that prevents products or components from being back ordered while keeping the inventory level low to free up capital for other uses.

Demand forecasting impacts almost every aspect of supply chain planning. The demand planning process allows supply chain professionals to make decisions based on projected demand rather than relying solely on historical sales data or guesswork. Because the links of the supply chain depend on each other, the demand forecast helps map out the steps needed to maintain appropriate inventory levels, meet customer demand, and bring partners into the loop.

Customer demand can shift quickly, emphasizing the importance of solid supply chain forecasting methods. Recent research illustrates how a hybrid model could deliver accurate demand forecasting when there is high uncertainty due to market volatility.

Additionally, demand planning should be recognized as an essential component of end-to-end supply chain planning. This approach to supply chain management considers the supply chain as a whole, beginning with ideation and design and concluding with the customer experience. The demand planning process allows supply chain managers to convert market research analytics into on-the-ground actions to improve operations from start to finish.

Why demand forecasting is crucial for effective supply chain management

Demand forecasting is part of a suite of tools and methodologies that has elevated the supply chain manager’s importance. It also serves an important goal of modern supply chain management: sustainability.

Sustainable management practices include reducing waste at every step in the supply chain. Because planning ties inventory levels to future demand, less waste and fewer products end up in the recycling bin. Forecasting gives supply chain managers more capacity to plan logistics and transportation for maximum efficiency, reducing fuel consumption. These savings advance the goal of sustainability while also benefiting the bottom line.

The top supply chain managers leverage demand forecasting to help set goals and deliver results. Many of the most successful supply chain managers we know developed their skills in accurate demand forecasting through advanced degree programs, such as the University of Tennessee’s highly ranked Master of Science in Supply Chain Management Online.

Components of Demand Forecasting Strategy in Supply Chain Management

To be successful at supply chain forecasting, consider incorporating the following components:

  • Clean, reliable data, including historical data and trend projections.
  • Actionable inputs, including from sales team members, outside experts, and market research.
  • Robust supply chain analytics
  • Flexibility to find and correct potential forecast errors and to change course as needed.
  • Collaboration with internal and external supply chain partners.

Amazon, the world’s largest e-commerce company and one of our Supply Chain Forum partners, is a fine example of demand forecasting done well. The company’s logistics operation is also the world’s most extensive. Amazon’s sophisticated supply chain planning allows it to anticipate demand and move products to the warehouses closest to the customers most likely to order them.

With top-notch demand forecasting, powered by AI, Amazon can offer one-hour delivery. In fact, your next Amazon order may already be in transit, thanks to Amazon’s sophisticated forecasting and understanding of customer demand. Amazon’s tech leverages big data and machine learning to take inventory optimization and operations planning to a new level.

While Amazon sets the bar high, demand forecasting technologies and supply chain analytics put a winning strategy within reach for planners at companies of all sizes. Demand forecasting in supply chain management is just one of the dynamic aspects that makes the profession exciting and challenging every day.

Learn more about the MS SCM Online program, including key dates, informational webinars, graduate testimonials, and registration information. Fill out the form below to request more information.

Request Information

By providing my information and submitting this form, I consent to be contacted via telephone (including a cell phone, if provided), email and text message about the Program selected above and additional programs under the Haslam College of Business.