During the outbreak of a new virus, we all inevitably have more questions than answers. Average citizens, first responders, politicians and business leaders alike are hungry for information to guide decisions that could have enormous consequences. In the relatively short time that we’ve had to fight COVID-19, one thing is apparent: Data can make a difference in how society reacts to this crisis – and it is helping us find ways to deliver better outcomes.
Over the last three months, as we’ve interacted with government agencies, public health workers and supply chain experts on their responses to the novel coronavirus, we’ve seen an intense need for more data and an increased demand to analyze the available data.
Recently, we shared a number of these stories – about data and analytics assisting with coronavirus efforts – at Virtual SAS® Global Forum 2020.
We talked about understanding the spread of the disease, forecasting and optimizing resources, and investigating treatments. And we focused a lot of our conference content on the trends we’ve seen supporting organizations during the pandemic.
SAS executive insights on demand
Now on demand, our Virtual SAS Global Forum 2020. Enjoy complimentary access and pick the sessions that you want to hear.
Optimizing hospital resources to save lives
One of the enduring headlines of this pandemic is the continued concern about shortages in hospital resources, including hospital beds, ventilators and personal protective equipment. We’ve worked with government health organizations and hospitals, including European health ministries and US private health systems, to predict patient hospitalization rates and plan resources accordingly.
For example, we’re working with Cleveland Clinic on projection-based simulations that estimate worst case, best case and most likely scenarios and can adjust in real time as the situation and data change.
Unlike some forecasts that use a single projection, the Cleveland Clinic models allow more flexibility for holding time periods constant, modeling various scenarios and updating the curves based on inputs such as changes in response measures. Cleveland Clinic and SAS are freely sharing these models through GitHub and through a scenario analysis tool to help hospitals, public health agencies and anyone looking to forecast the impact of COVID-19 to improve resource optimization and planning.
At Virtual SAS Global Forum, I spoke with Chris Donovan, Executive Director of Enterprise Analytics at Cleveland Clinic. He explained how their team used SAS to build multiple models to predict admission rates and hospital resource needs. This approach helped them evaluate best case and worst case scenarios and adapt as predictions shifted.
Cleveland Clinic responds to COVID-19 with analytics
In this short clip, Cleveland Clinic Executive Director of Analytics Chris Donovan explains the importance of building an analytically mature organization.
Understanding coronavirus and how it spreads
Epidemiologists track disease outbreaks using statistics. Cumulative frequency graphs and exponential growth curves have been shared widely to help visualize the growth of the disease and understand when the growth might be peaking.
Mapping dashboards have also been useful for seeing where clusters are happening. We’ve partnered with government and health officials in Iowa, Italy, Germany and other states and countries to develop region-specific models and dashboards for planning government and health care resources.
In Asia, we’re working with public health officials to improve their contact tracing programs with visual analytics, data management and network analysis technologies. We believe this approach can help scale the traditional process of test, trace and isolate so that more people can be notified about possible exposure.
During Virtual SAS Global Forum, Steve Bennett, Director of our Global Government Practice, showed us how SAS Contact Tracing Investigations can simplify contact tracing programs and reduce the spread of the disease.
He also showed how the Philippines and other countries are using analytics for understanding broad population mobility trends to support better public health policies. “Understanding mobility can help you make better decisions about public transportation policies, shelter in place policies and getting medical resources in place in advance,” explained Bennett.
From modeling the genetic makeup of COVID-19 to developing new treatment methods and vaccines, the medical and scientific communities are leveraging every resource available to combat the disease. Jim Goodnight CEO SAS
Forecasting supply and demand for critical goods during a pandemic
As supply chains witness unprecedented interruptions, consumers change their behaviors and some businesses slow their demand, forecasting has become a true challenge.
Working with retailers and manufacturers, we know that traditional models tend to overcorrect in unforeseen circumstances. When there’s little historical data on similar scenarios, models will be hard to stabilize. We’re seeing that it helps to make small, frequent updates to the model with slight modifications using data signals, like timing, shopping channel mix and product mix.
One consumer goods company asked us to forecast demand for multiple European countries based on different supply chain restrictions in each country. As a major food producer, this company is relied upon by millions of consumers to keep grocery shelves stocked with healthy food. The models our teams developed accounted for potential staff shortages due to sickness and other changes that could affect accessibility to sites in multiple countries.
A popular presentation at our conference, Supply Chain Demand Sensing: COVID-19 Impacts, explained how retailers can predict and plan for changes to consumer demand, including how to model a return to recovery.
Supporting critical scientific research with deep analytics
From modeling the genetic makeup of COVID-19 to developing new treatment methods and vaccines, the medical and scientific communities are leveraging every resource available to combat the disease.
In addition to working with health and life science companies developing new drugs to fight the new virus, we’re deploying sophisticated text analytics to find answers in more than 50,000 full-text documents on COVID-19 virus and other coronaviruses, which have been gathered and released to the public by the Allen Institute for AI, Semantic Scholar and other research groups.
When we look at the volume of data gathered on this disease by the research community, it’s more than any one person can consume. Applying text analytics and AI to discover trends and commonalities among research studies can lead to new insights about the disease and help answer many unanswered questions.
SAS is joining other technology companies and citizen data scientists to analyze this open data set, share what we learn about the data, and provide a collaborative platform for researchers to further study viral transmission, incubation and mutation.
The conference presentation “How COVID-19 Is Accelerating Innovation” describes tools and resources for epidemiological modeling, science research and pharmaceutical development.
Recovering and reimagining the future
I have always believed that knowledge is empowering. As we confront new waves of the virus around the globe, I am optimistic that analytics will continue to illuminate the path.
Analytics will help us understand more and more about the R0, or the rate of spread, and how each of our mitigation efforts affects it so we can work to reopen economies methodically and safely. Analytics will assist economic recovery by identifying where stimulus money and incentives can have the biggest impact.
And further into the future, I’m sure we’ll be applying analytics to the many questions that remain after the storm. We’ll be seeking to understand what lessons we have to learn from the experience.
- How can we better protect vulnerable populations?
- How can we act sooner?
- How can we be prepared for the next unexpected event?
History shows that when we are disrupted and forced to work in new ways to solve immediate problems, some of the innovations are durable. New manufacturing and supply chain practices will continue, new research will inspire future ideas, and partnerships that brought new products to the market will remain intact. All of these improvements can help us become more resilient and more innovative in the long term.
Even as we work to understand what we can do better, I hope we will remember what we did right – how we discovered a spirit of resilience, cooperation and innovation that allowed us to emerge stronger.
You can visit Virtual SAS Global Forum to watch the presentations discussed here – and many others.
- Fighting coronavirus: 4 ways analytics is making a differenceCoronavirus has separated us from family, friends, cultural and religious communities. Unfortunately, isolation is essential to slowing the spread of the virus. What else can be done? Learn how analytics is being used to improve responses to the coronavirus outbreak.
- Three steps for conquering the last mile of analyticsPutting your analytical models into production can be the most difficult part of the analytics journey. It’s no surprise that this last mile of analytics – bringing models into deployment – is the hardest part of digital transformation initiatives for organizations to master, yet it’s the most crucial.
- ModelOps: How to operationalize the model life cycleModelOps is where analytical models are cycled from the data science team to the IT production team in a regular cadence of deployment and updates. In the race to realizing value from AI models, it’s a winning ingredient that only a few companies are using.