Homework 12 approach
docx
keyboard_arrow_up
School
Georgia Institute Of Technology *
*We aren’t endorsed by this school
Course
6501
Subject
Industrial Engineering
Date
Dec 6, 2023
Type
docx
Pages
2
Uploaded by CaptainCoyoteMaster1037
Homework 12 approach
Power Company Case Study
For this scenario, if I was a consultant for the power company, I would sit down with the client
and any subject matter experts within the organization to learn how payments are made and
how the current processes work. From there I would be able to either develop models or plans
that would align with current processes or make the decision to completely replace the process
and develop a full structured plan. For this case study, we do not have that luxury to have
conversation with a given company. Throughout my thought process, I will be making
assumptions on the type of data we would work with and the processes that I believe would
work best (I also consulted with classmates and friends).
Step one:
Use an analytic model to determine the potential clients who can pay the bill but
chooses not to pay to prioritize the shut off, of power.
Given:
Past payment history (On-time, late)
Type of payment method
Type of payment plan
Price of payment plan
Account age
Customer income level
Credit score
Customer area code (Zip Code)
Current Payment
Use:
Utilize a classification model that will predict whether someone can or cannot make payments
to the Power Company. With this result, we can take those who can make payments and build
out another model to classify whether they make or don’t make payments. We can use an SVM
model to make these distinctions.
The initial SVM model will need to consider the past payment history, type of payment plan,
payment methods, customer income level, and credit score. These data points will allow the
company to classify whether they can cover the price of the payments.
The second SVM model will utilize the resulting customers all data points to classify whether
customers are paying for the service.
To:
Identify a list of customers who can make payments, but do not actually make the payments.
Step two:
Determine the logistics behind turning off the power to customers.
We should use an optimization model to determine the process for turning off power. We need
to consider the overall cost of turning off power versus continuing to provide power, the length
of time at which the customers who are able to make payments do not pay, and work force
available.
Another approach I was thinking of, is to take the result from the first SVM model, then rank the
results based on the duration of time payments are not made and by zip code. We can assign
given employees a list of the top customers not paying and attempt to cover as many homes as
possible. Optimize the overall process by priority and location base.
Your preview ends here
Eager to read complete document? Join bartleby learn and gain access to the full version
- Access to all documents
- Unlimited textbook solutions
- 24/7 expert homework help