“Quick response forecasting (QRF) techniques are forecasting processes that can incorporate information quickly enough to act upon by agile supply chains” explained Dr. Larry Lapide, in a recent Journal of Business Forecasting column. The concept of QRF is based on updating demand forecasts to reflect real and rapid changes in demand, both during and between planning cycles.

Demand challenges for companies 

The length of time it takes for the demand forecasting process to incorporate rapid changes, or short-term spikes in demand is a challenge for most companies. Short-term spikes in demand can occur with retail store promotions, sudden changes in weather conditions and social media sensations.

For example, if a famous person is seen wearing a certain garment by a fashion designer, or a unique color of nail polish, this gets circulated rapidly to others through Facebook, Twitter, and other social media. Then, people start purchasing the product online, from their mobile devices, as well as brick-and-mortar stores. Since these short-term spikes are not reflected in sales orders, they are not included in demand forecasts, creating challenges for the supply chain to meet those spikes in demand.

Another challenge is enabling QRF, given the explosion of digital data and the enormous amount of information available regarding automated consumer engagement. This shift from active engagement to automated engagement takes place when technology takes over tasks, from information gathering to actual execution, like purchasing the latest designer shoes worn by a famous person at the Grammys online.  When data expands, as a result of the Internet of Things (IoT), streaming data from devices can make it difficult to decipher demand signals from the noise. In many cases, these short-term spikes exceed the projected baseline demand, and are not always a result of planned sales promotions or marketing events.

The goals of supply chain forecasting are not always about minimizing operating costs and inventories, but more about maximizing inventory availability in order to capture potential upside revenue. The traditional “efficient supply response” philosophy targets mature products with the goal of minimizing operating costs and inventories, and focusing less on lost sales for those products.

Many supply chains are still too sluggish, and not agile enough to take full advantage of demand forecasts that include sort-term spikes. As a result, manufacturing managers tend to complain about getting whipsawed by rapid changes, despite resulting increases in accuracy.

Downstream data, like point of sale (POS) and syndicated scanner data, is often dismissed as too detailed and cumbersome to work with, especially for the lion’s share of a company’s product line. To most, it appears to be useful only for products that sometimes experience significant changes in demand, such as during promotions and new product launches. Others feel that POS data is not especially predictive, or that it can't be accurately forecasted. Thus, POS and syndicated scanner data is rarely, if ever, incorporated in a meaningfully way into the demand forecasting and operations planning processes.

Misconceptions of downstream data 

These misconceptions about downstream data are due to a lack of understanding, as most demand planners are too far removed from marketing reporting upstream into operations planning. However, according to Oliver Wight (2003), downstream data is the true demand signal.

In my personal experience, I've found downstream data to be more stable and easier to forecast than sales orders or shipments. The main reason is that there is no “bull whip” effect associated with downstream data. In fact, POS/Syndicated Scanner data can be forecasted much more accurately, providing meaningful results.

Today, downstream data is even more accurate due to retail census data through digital scanners for all markets, channels, stores and key retail customers.

Forecasting at the edge may be the solution for QRF

Quick Response Forecasting (QRF) is a concept that coincides with a recent post I wrote, “Forecasting at the Edge for Real-Time Demand Execution.” In the post, I propose using event stream processing combined with machine learning to analyze data as it is generated at the store device, which decreases latency in the decision-making process.

The key benefit of forecasting at the edge is the ability to analyze data as it is generated. For example, if sensor data from mobile devices points to increased sales for a particular product, analyses at the network edge can automatically alert manufacturers to increase production of that product to meet the short-term spike in demand.

Using event stream processing combined with machine learning, that information can save time and lower costs compared with transmitting the data to a central location for processing and analysis, potentially enabling companies to minimize or eliminate back orders.

Pushing analytics algorithms to sensors and network devices alleviates the processing strain on enterprise data and analytics systems. Instead of demand planning, we can accomplish real-time demand execution across the supply chain.

QRF is possible now

Today, the technology exists to support Quick Response Forecasting using event stream processing and machine learning.  There are no longer any data or technology hurdles to overcome. Only our own inhibitions and the desire to take the leap toward achieving the end goal of an autonomous supply chain are stopping us.  So, why not do everything at the edge, including forecasting short-term spikes in demand.

Is “Quick Response Forecasting” really "forecasting at the edge for real-time demand execution", or something completely different? Let me know your thoughts regarding QRF.

Learn more about using streaming data for competitive advantage.