Anomaly Detection in Supply Chain Networks – Dow



Company Name / Department Dow Inc.
Contact Person

Aram De Ruiter


Terneuzen, the Netherlands

Study programme(s)

Operations Management

Community ESCF
Start Date

September 1st, 2021

Housing arranged by company


? euro per month

Company Description

Dow Inc.’s ambition is to become the most innovative, customer-centric, inclusive and sustainable materials science company. Our goal is to deliver value growth and best-in-class performance. The Company’s portfolio is comprised of six global business units, organized into three operating segments: Performance Materials & Coatings, Industrial Intermediates & Infrastructure and Packaging & Specialty Plastics. Its products serve different applications, including coatings, home and personal care, durable goods, adhesives and sealants, and food and specialty packaging.


Project Description

As a global materials science company, Dow has a complex Supply Chain Network. Due to natural and technological disruptions, and changes taking place in the global marketplace and changes that are internal to Dow, our supply chain can become sub optimal. In this project, by using anomaly detection algorithms, we would like to be able to quickly identify when our network performance has degraded. Monitoring the performance will allow the Company to intervene via network redesign, new policies, and so forth, to re-optimize the Company’s operations given to the new environment. For this project, Dow seeks to find a set of anomaly detection algorithms and techniques to monitor and identify performance based on dynamic supply chain data and compare their effectiveness. As a stretch goal, the Company would like to see the capability to recommend interventions or new policies in response to market changes.


Goals of the Project

  • Establish definition of anomalies in supply chain models
  • Literature review on anomaly detection algorithms
  • Apply anomaly detection algorithms on Supply Chain Data
  • Compare results of different models


  • Effective anomaly detection models that successfully identify anomalies in Dow’s Supply Chain
  • Code and Model Repository
  • Model comparison analysis
  • Thesis report

Essential Student Knowledge

  • Supply Chain Management
  • Coding Skills (Python, R, MATLAB, etc.)
  • Data analysis
  • Anomaly Detection Algorithms
  • Graph models and algorithms




 More information: