Assessing Operational Risk COVID Case

Assessing Operational Risk in the COVID-19 World

Contents

Case Brief 2

Background 3

Using Analytics 4

Data and Additional Resources 5

Case Brief

Your team has been tasked with forecasting specific date ranges for the possible re-opening stores across the country at normal operations. Your client is the popular Seattle Headquartered multi-national Café styled restaurant MoonBux.

MoonBux stores across the United States have been closed for sit-down business since March 25^th, 2020. Since June, some locations allow for drive-thru or take-out orders especially those serving hospitals and first responders, but approximately 70% of their stores do not allow for indoor seating. Some locations, such as those on college campuses and at airports remain completely closed.

The company is asking for advisory services that will help them make operational decisions including, but not limited to, the potential need to close certain locations completely, decisions about employees such as extending employee benefits versus less desirable alternatives (such as lay-offs and furloughs), and in determining any anticipated issues surrounding supply chain management when resuming normal operations.

Specifically, MoonBux is asking your team to provide an assessment of the timing and risks associated with resuming normal indoor seating operations in some of their largest domestic markets. They are not optimistic about the possibility of resuming normal operations by the end of the calendar year and would like you to focus on December 2020 in your assessments and discussion of re-opening. They have approached your team as you are well known for your reputation both for verifying data and for presenting data analytics-based solutions to business problems. This is a real-time case using current data. As we are doing this case early in the quarter, the case asks you to analyze a pre-existing dashboard (rather than build the dashboard), provide an outlook, and discuss concerns with data veracity for different sources of data. The exercises discussed in the requirements section will be completed in teams, and each team will be assigned a state to analyze. The next section provides background, which is followed by a section on using analytics.

Background

Accounting Advisory Services are an important part of most accounting firms. Clients will typically seek out various advisory services from accounting firms in areas that they do not have the expertise or inclination to perform on their own. In addition, often firms can provide advisory services at a lower cost than the client would be able to achieve. Advisory services can be defined as value-adding strategies and insights for the client. They can provide services broadly from forecasting revenue or cash-flows to aid the client’s budgeting needs, to technology implementations such as enterprise risk management system implementations or blockchain implementations. Other common examples of advisory services relate to regulatory compliance advisory (i.e., meeting new regulatory standards) risk assurance services (including cybersecurity and data protection), and operational advice to improve the client’s performance and profitability.¹

Data Veracity can be defined as the degree to which the data is accurate, precise, and trusted. Whereas making decisions based on data analysis appears much better than making gut-based decisions, low quality data will lower the usefulness of insight being drawn from data. Low quality data, including manipulated data, can even result in incorrect insights and recommendations. Hence understanding data veracity is an important part of the analytics mindset and assessing data veracity is an important part of the analytics skillset. In an article by individuals at Accenture, some examples of where problems with data veracity can lead to problems in advisory services include the use of inaccurate customer contact data (i.e., addresses, emails, phone numbers) that lead to sending out promotions and offers to the wrong group of customers, and duplicate data sources leading to different measures of risk which can delay operational decisions.²

Using Analytics

To solve this case, you investigate sophisticated pre-existing analytics performed by the Institute for Health Metrics and Evaluation (IHME) and the aggregated forecast data provided on the Centers for Disease Control and Prevention (CDC).

The IHME is an independent global health research center at the University of Washington (http://www.healthdata.org/). On their website, they provide visual analytics of COVID-19 projections (https://covid19.healthdata.org/global?view=total-deaths&tab=trend) and COVID-19 data available for download. The IHME projections are based on a sophisticated model that builds in different scenarios based on social distancing (including mask-wearing) and easing of restrictions. They also provide a model that they label projections that we will consider as the most likely scenario given the tradeoffs between social distancing and easing of restrictions.

The CDC provides forecasts from over 30 different groups that have provided forecasts of Coronavirus cases and COVID-19 deaths. They also provide visualizations on their website, showing the variation across models (https://www.cdc.gov/coronavirus/2019-ncov/covid-data/mathematical-modeling.html). This page highlights one issue with data veracity – we have many competing models – or duplicate data sources that tell different stories about risk! Thus, making decisions based on this data will be difficult. At the time of writing this case, the one-week ahead case forecasts for Washington State from the 19 different providers available in the CDC’s file, has a standard deviation of nearly 600. This file, which is based on forecasts collected by the CDC is dated 9/7/2020.

The standard deviation provides us with a measure of dispersion, or uncertainty, in the forecasts of the case numbers. The higher the standard deviation of the forecasts, the less certain we can be about relying on any single forecast. This measure will help us understand data veracity.

Data and Additional Resources

The following data and resources are available in the case supplement:

1. The underlying data set from the CDC as of 09/07/20: “2020-09-07-all-forecasted-cases-model-data.xlsx”

2. The underlying data from the IHME and variable descriptions in the folder IHME.