<img height="1" width="1" style="display:none;" alt="" src="https://dc.ads.linkedin.com/collect/?pid=358169&amp;fmt=gif">

How to Build Supply Chain Resilient Against Mounting Covid-19 Pressures

Every business on earth is struggling to come to terms with the scale and swiftness of the COVID-19 pandemic, and none are escaping unscathed. In a time of such dramatic change, traditional forecasting techniques are proving too slow to offer reliable predictions for any business or government. However, there are steps you can take today to go beyond standard forecasting techniques and respond to the massive changes brought about by the coronavirus.

Here are some insights on how to cope with these uncertain times, as presented in John Galt’s webinar “Forecasting in the Pandemic” with Dr. Barry Keating of the University of Notre Dame.

 

Data cleansing

Relying on common forecasting practices will not suffice for most businesses and governments in a time of Covid-19. Little historical data exists to take into account such a momentous anomaly. We therefore will have to look at the use of data beyond our own historical data, to look at similar situations and explanatory data. Some vital insights can be obtained from the availability of blood supplies following 9/11. Blood is not a commodity yet consists of many types and variants that one may compare to different product types.. A strong surge in blood supplies in the months following 9/11 had to be corrected and cleansed in the following years. . Data cleansing can help you return to normal demand once the pandemic has passed.

Event modeling can also help model specific events such as a promotion or a pandemic.. Some businesses will see a negative impact to their business from Covid-19 while other businesses will barely be able to keep up with sales. Flagging these events can help exclude them from history once business returns to normal and will be a valuable source of data for any future disruptions.

 

Regression and  “Nowcasting” 

While the current pandemic has largely been impossible to foresee, this storm is far from over. We should therefore collect and analyze all available data to monitor its impact.

Besides event modeling, multiple and stepwise regression can help you explain the impact of the pandemic by incorporating causal variables in the forecast model. As an analogy, sales of snowblowers can be predicted with variables such as snowfall, promotional strategy, and price. If you can find variables that are relevant to your sales and that are impacted by coronavirus, you will be able to respond rapidly to changes in those variables and forecast

New product forecasting offers useful insights in how to manage Covid-19. Most epidemics roll out as an S-curve, such as the adoption of cell phones. A strong, exponential growth is followed by a flattening of the curve.The turning point in an S-curve can be predicted quite accurately both by the Logistic Curve or the Gompertz Curve in ForecastX and Atlas Planning. These are examples of ”phenomenological” models, which are often used in medicine to predict the progress of epidemics. The results of these models can then be used as a causal variable in a regression model, which can suggest the impact of the virus on demand.

Another useful technique is “nowcasting” - the use of very short-term data for near term forecasting purposes. Google Trends will allow you to access recent search history for terms related to your business, which you can then use as a regression variable even for very near-term shifts in demand. tool that will offer . Google searches have proven to predict even apparently unpredictable events. The outbreak of the regular flu, for example, can be traced and predicted regionally by searches for terms like fever, headache or cough medicine. Mentions on social media and website traffic can also give you early warning of demand shifts, giving you time to respond.

By using data beyond simple history, you will be able to produce forecasts that keep up with the tumult and change of today’s pandemic. You may also find that these techniques will help you improve your forecasting even once business returns to normal.


For more information about supply chain management best practices and what steps you need to take to best build your forecasting resilience in a time of crisis, schedule a free consultation with one of our Demand Planning Xperts today. Our Atlas Planning and ForecastX solutions can help you build an end-to-end planning process that will help you handle today’s changes and beyond.

 

About John Galt Solutions

More than ever, companies must be able to sense and respond to the dynamics of a complex supply chain. John Galt's Atlas Planning is a unified end-to-end supply chain planning platform that helps you increase forecast accuracy, optimize inventory levels and maximize supply chain performance. Since its founding in 1996, John Galt Solutions has built a proven track record of providing affordable, automated demand and inventory management services for consumer-driven supply chains. We have an unmatched ability to configure tailored solutions for customers, regardless of size, industry, or business challenge, that save both time and money by compressing implementation periods and delivering intelligent information that positively impact your bottom line.

To learn more about John Galt Solutions, contact our press office at 312-701-9026 or visit www.johngalt.com.




Tags:     Blog Demand Planning S&OP Inventory Management Food & Beverage Scenario Analysis Food Service Atlas Planning Suite For Supply Chain Nerds Automotive Manufacturing For Prospective Customers Agribusiness Aviation Chemical Consumer Goods Healthcare & Pharma Hi-tech Non-profit Retail ForecastX

Subscribe