CASE SOLUTION: Brownspeed Health Care Employee Retention Using Predictive Analytics (Complete & Original)

BROWNSPEED HEALTH CARE: EMPLOYEE RETENTION USING
PREDICTIVE ANALYTICS CASE STUDY SOLUTION




e
pl
SYNOPSIS
m
Sa
Brownspeed Health Care (BHC), based in Orlando, Florida, provided a wide range of consultative services
to the US health care industry. Recent social and political issues in the US health care system, along with
near-record employment levels, led to a high rate of employee turnover throughout the industry. Employee
n

turnover cost the health care sector billions of dollars each year in direct replacement expenditures (e.g.,
recruiting, training, and integration). These estimates did not include the added financial impact associated
tio


with lost productivity and opportunity costs.

BHC received numerous requests from its clients for assistance, and Johnny Stonebrook, BHC’s chief
lu



analytical officer, has been asked to report on how the company can help its clients address the problem.
BHC management is hoping to help its health care clients by deploying an analytics-based solution to
So




The Case Solution Starts From page 6

, e
pl
m
ASSIGNMENT QUESTIONS
Sa
1. What are some of the key challenges associated with time series forecasting, especially as they apply
to the dynamic health care sector?
2. Download the monthly CPI-W from the US Bureau of Labor Statistics website for 2009 to 2019.5
3. Develop a time series line graph in Microsoft Excel for the CPI-W using the US Bureau of Labor
n

Statistics dataset.
4. Analyze the database using the one-, two-, and three-parameter exponential smoothing models.
tio


5. How do the forecasts compare with the actual monthly data for 2019?
6. What is the planned Social Security COLA for 2019?
7. Identify some specific employee retention strategies.
lu
So




The Case Solution Starts From page 6

,ANALYSIS

1. What are some of the key challenges associated with time series forecasting, especially as they
apply to the dynamic health care sector?

Time series forecasting is a subset of predictive analytics. It consists of a family of mathematics-based




e
techniques that can be used to predict future events using past data. Time series forecasting is based on the




pl
fundamental assumption that future patterns will be like historical trends.

Making predictions about future events based on the past is often referred to as extrapolation. The accuracy


m
of time series forecasting models tends to degrade when structural changes occur in the underlying data.
For example, the transition to national health care insurance could significantly alter patterns in the number
Sa
of new hospital admissions.6 In these situations, more sophisticated modelling techniques are often
required, such as relational predictive analytics.

Some specific challenges associated with health care forecasting include the reliability and limitations of
n
tio
lu
So




The Case Solution Starts From page 6

,EXHIBIT TN-1: STEPS TO DOWNLOAD DATABASE FROM US BUREAU OF LABOR STATISTICS
1. Go to the US Bureau of Labor Statistics website (www.bls.gov).
2. Click the Data Tools button on the toolbar and select Top Picks.
3. Under Price Indexes, check the box “CPI for Urban Wage Earners and Clerical Workers (CPI-W) 1982-84=100
(Unadjusted) - CWUR0000SA0.”
4. Click the Retrieve Data button at the bottom.
5. Click More Formatting Options in the upper-right corner.
6. In the Select View of the Data Display to the left, select Column Format.
7. Click the Retrieve Data button.
8. Click the Excel icon (next to “Download”) above the table.
9. Save the Excel file to your desktop.




e
pl
m
Sa
n
tio
lu
So




The Case Solution Starts From page 6

, EXHIBIT TN-6: CPI-W MONTHLY FORECAST (2019) THREE-PARAMETER MODEL

2019 Actual Forecasted Error (%)
January 245.133 245.129 0.00
February 246.218 245.095 –0.46
March 247.768 244.688 –1.24
April




e
pl
m
Sa
n
tio
lu
So




The Case Solution Starts From page 6