top of page
  • Writer's pictureJunghyun (Andy) Kim

Surrogate-based optimization for a mega-hub location problem under uncertainty

Sponsored by CJ logistics


Research Motivation

In recent era, the urbanization and the growth of E-Commerce have led to a significant increase in parcel delivery service network complexity. Especially, South Korea is an exemplary market where next-day parcel delivery service is so dominant that various companies are committed to improve the current service networks. Although many companies in South Korea strive to minimize the amount of time for delivery, it is seemingly inevitable to confront network traffic saturation issues occurring in various fields such as limitation of hub capacity. In response to these concerns, the CJ logistics has started to consider opening a new mega hub in order to ensure timely delivery service.


Key Idea

I propose a surrogate-based optimization methodology employing state-of-the-art surrogate models for a new mega hub location problem in a complex next-day parcel delivery service network in South Korea. The proposed methodology is to utilize the synergism of the collaboration between supervised machine learning techniques and advanced design methods such as Design of Experiment and Monte-Carlo simulation.


Overview of the proposed methodology


Results

The results show that the optimum location would reduce transportation costs by approximately 14% compared to the current hub network system operated by CJ Logistics.



Publication

[1] Transportation Research Part E (Under Review)

bottom of page