The Limits of Analytics During Black Swan Events A Case Study of the Covid-19 Global Pandemic

During statistically unlikely events (Black Swan events) analytical models fail to provide their expected level of fidelity: their error can increase by several hundred percent. The modern economy is built on such analytical models, which are intended to provide useful results during routine conditions. All models eventually fail to provide the expected level of fidelity under extreme conditions. This thesis investigates the critical limitations of analytical methods during Black Swan events. Specifically, we study the space of possible model errors for statistical forecasting models and their respective implications for supply chain systems. We explore the forecast errors through numerical simulation and a real-world case study of a global manufacturing company experiencing the Covid-19 pandemic, a Black Swan event. We demonstrate that in some cases demand can shift by over 60%, leading to the bifurcation of the forecast error space, and resulting in an 500% increase in forecast error. This new regime causes supply chain planning systems to grind to a halt as existing inventory models become irrelevant. Such a drastic change in a company’s operational environment requires urgent action to ensure continued operations. For management to make correct decisions, it is critical for them to understand the limits of analytics during Black Swan events.

Name thesis:degree_name

Master

Department dc:contributor.department

Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science

Grantor dc:publisher

Massachusetts Institute of Technology

Year dc:date.issued

2021

Author dc:creator

• Kaminski, Erez

Advisors dc:contributor.advisor

• Kraska, Tim
• Willems, Sean

Statement dc:rights

• In Copyright - Educational Use Permitted
• Copyright retained by author(s)

Licence dc:rights.uri

• https://rightsstatements.org/page/InC-EDU/1.0/

Handle dc:identifier.uri

https://hdl.handle.net/1721.1/139124

OAI identifier oai:identifier

oai:dspace.mit.edu:1721.1/139124