Variable labor optimization

In businesses that rely on part-time employees - such as entertainment venues, retail, ecommerce warehouses, etc. - being able to optimally staff the business is key to maximizing profitability while providing quality customer service.

Overview

Identify the optimal staffing plan to enable the business to serve the customers at minimum cost.

TARGET

Did the staffing level adequately serve the customers.

Challenge

Most companies have some idea of likely demand but turning that demand into an actual staffing plan can be challenging. Historical timecard data needs to be combined with the historical demand forecasts and information about the weather and events. Next, some measure of customer service needs to be extracted, either from sales, customer support or inventory systems. All this data needs to be cleaned, joined and transformed into valuable ML features before going into model training. This pre-modeling prep process can be frustrating and time consuming. We are here to help.

Modeling techniques and libraries

Machine learning demand

Build machine learning models to predict the staff required as a function of the independent variables. Use model interpretability packages to evaluate the impact of the independent variables on the prediction.

Packages:
  • Sklearn
  • ELI5
  • LIME
  • SHAP

Schedule optimization

Build and solve the optimization problem to find the optimal number of employees to schedule to maximize the customer service and timing to maximize sales.

Package:
  • CVOXPT
  • pyOpt
  • scipy.optimize

Data features

Commute distance
HR Database
Data Type
Continuous
Target
No
Yes
Day of Week
Calendar
Data Type
Categorical
Target
No
Yes
Discounts
POS
Data Type
Continuous
Target
No
Yes
Event
Calendar
Data Type
Binary
Target
No
Yes
Holiday
Calendar
Data Type
Binary
Target
No
Yes
Items
POS
Data Type
Continuous
Target
No
Yes
Number of different schedules
Timecard
Data Type
Continuous
Target
No
Yes
Overtime in last payperiod
Timecard
Data Type
Continuous
Target
No
Yes
Overtime in last year
Timecard
Data Type
Continuous
Target
No
Yes
Price
POS
Data Type
Continuous
Target
No
Yes
Rain
Weather
Data Type
Binary
Target
No
Yes
Severe Weather
Weather
Data Type
Binary
Target
No
Yes

Related accelerators

No items found.

Get your data science on.