Supply Chain Network Design Case Study

Found optimal network configuration for highly constrained network

Project Summary:

• Determined optimal network configuration for unconstrained scenario, plus two more scenarios with network constraints and pressures.

• Compared optimal network configuration costs to calculated baseline (current-state) costs.

• Our model reduced network costs by 45%, for a savings of $59 million!

• Made data-driven business recommendations based on our findings.

Files:

• View Final Report here

• Download Excel spreadsheet w/ Solver model (.xlsx)

  • Preview only (.html) - does not render perfectly
  • Uses Solver, OpenSolver - due to large number of variables, doesn’t work with built-in Solver
• GroupBy.html and Q10-Revenue.html
  • HTML outputs. Used Python (Pandas, Numpy) to prep data

Sample Findings:

Map of customers (blue) and DCs (orange) (More)

Our model reduced network costs by 45%, for a savings of $59 million!

By allowing for split quantities, i.e. relaxing binary constraints on Decision Variables, our model allowed all customers to be served and all demand to be met - even when distribution centers had limited capacity, three distribution centers were closed, and total network demand was increased by 15%. (More)

Tools Used: Regex, Python (Pandas, Numpy), Excel, Solver, OpenSolver

Course: DSB 6200 Supply Chain Analytics, Wayne State University

Assignment: Final Term Project (XRetailer Supply Chain Network Design Problem Case Study)

Submitted: May 7, 2018

Last Updated: May 9, 2018

Back to MS in Data Science Coursework