How Can Data Analysis Be Key in Dealing with Today’s Bullwhip Effect in Supply Chain?

Joshua Zable | 31 March, 2020

When the way the world works is constantly changing now more than ever, demand spikes for certain products (think N95 masks, toilet paper and hand sanitizer) and it dwindles or downright disappears for others (how often did you use the word "non-essential" before the past few weeks?). In this business climate, companies need to act quickly to deal with fluctuations in both supply and demand. 

Amazon actually stated last week it is delaying delivery of non-essential products and won't accept third-party sellers' products to their warehouses for fulfillment unless they are "household staples, medical supplies or other high-demand products." 

When you see rapid developments like this frequently alter well-established patterns of supply and demand, you can expect a “bullwhip” effect on the supply chain. When businesses do not take steps to mitigate the bullwhip effect, supply chain inefficiencies occur. Swings in inventory in response to shifts in customer demand become more and more exaggerated as one moves further up the supply chain.  

Get the Free eBook:  Dealing with Today's Bullwhip Effect in Supply Chain

Savvy supply chain management involves the leveraging of channel-wide integration to better serve customer needs. By coordinating quality management activities, productivity and efficiencies can increase. While many are employing classic statistical process control techniques, extraordinary times like these require the use of additional tools. 

Most statistical process control methodologies assume a steady state process behavior where the influence of dynamic behavior is ignored. Control charts, a popular and effective method, have one major drawback: they only consider the last data point and do not carry a memory of the previous data. As a result, small changes in the mean of a random variable are less likely to be detected rapidly.

In a dynamic business environment, getting ahead of the bullwhip is critical. Combining classic statistical process control, like control charts, along with additional methods can be extremely helpful. For example, exponentially weighted moving average (EWMA) control charts, a much less popular method, improve upon the detection of small process shifts. Rapid detection of small changes in the quality characteristic of interest and ease of computations through recursive equations are some of the many good properties of the EWMA chart that make it attractive.

Unfortunately, challenging times like these require us to think differently. It also affords us an opportunity to explore additional proven statistical quality methods. This will not only help us proactively deal with today’s dynamic environment, but build an even stronger foundation for the future.