Using advanced analytics to forecast demand in the life sciences market during COVID-19
The COVID-19 pandemic has revealed the vulnerability of pharmaceutical supply chains. Pharma companies are focusing on risk management to improve the resilience of their networks. Most of the measures they will take, including on-shoring, over capacities and redundancies, will lead to higher costs. To decrease inventory levels across these new supply chains and control costs, pharma companies should also focus on improving their demand planning.
Demand planning challenges
The life sciences industry faces a variety of unique challenges in demand planning during the COVID-19 pandemic. Life sciences organizations are struggling to meet supply, from raw materials through to the pharmacy, as a result of COVID-19 related disruptions. The crisis is changing the demand for over-the-counter (OTC) medications and therapeutic devices and how people shop for them.
Meanwhile, demand for supplies for clinical research is a moving target. Some trials are delayed due to distancing measures. And new research is popping up to explore the efficacy of existing drugs on the treatment of COVID-19.
For high-selling OTC medications and seasonal treatments like flu vaccines, fully automated traditional time series forecasting works well. However, for very specialized treatments and sudden occurrences like COVID-19, machine learning models have proven to be much more accurate to forecast demand.
Demand sensing, forecasting and planning are critical to understanding the many variables at play, especially during a pandemic. These are complex problems, requiring advanced analytics to best understand how demand is changing on the ground and to predict future changes as a result of the pandemic. For example, we can take data for demand spikes and lulls in regions first affected by the pandemic and use it to make predictions in areas that have yet to experience their peak. Similarly, we can use data from regions that are reopening to forecast changes to demand in other regions as they reopen. Organizations can then use this insight about demand to inform manufacturing and supply chain decisions as they work to mitigate disruptions.
Author: Alexander Daehne