How Walmart Applies Data Science

December 10, 2018
How Walmart Applies Data Science

Data science is applied at Walmart Labs to sourcing, order/shipment preparation, transportation, last mile/routing/scheduling, and last mile order pick up. Applications include queuing and linear programming.

Check out this Medium article for an informative and high-level overview of how data science applications improve the efficiency and effectiveness of complex supply chain management at Walmart.

This post is based on the Medium article, Data Science in Walmart Supply Chain Technology, by Mingang Fu, November 1, 2018. Image source: Prasit photo/Getty Images.

Discussion Questions:

1. Describe how queuing theory is applied to Walmart’s supply chain management.

Guidance: Many students have not considered queuing applications beyond the typical retail service examples. Note that queuing theory is being applied by Walmart to solve last mile order pickup problems.  The instructor could illustrate queuing concepts and terms from the textbook by further discussing the queuing example provided in the article.

2. Describe the application of linear programming in Walmart’s transportation management.

Guidance: Students may not be familiar with more complex linear programming problems such as mixed integer applications.  The instructor may choose to introduce more advanced linear programming methods or simply illustrate with this article the utility of linear programming.

3. Now that you have explored supply chain management data science applications from Walmart Labs, how could data science applications improve operations in other industries?

Guidance:  Below is an activity which instructors can use to help students consider the various data science tools available, and link these tools to potential projects for an industrial application.

  1. The instructor lists three different industries on the whiteboard.
  2. Under each industry listed, the students describe data science applications that would improve operations management.  (Recommended: 5 or more applications per industry.)
  3. Students rank order the priority of applications, individually or in teams.
  4. Discuss the results as a class.
  5. The instructor can then highlight various data science methods required for each application, the challenges of developing each application, etc.
print
Facebook
Twitter

Leave a Reply

avatar
  Subscribe  
Notify of