32Tank planning optimization with learning algorithms – Den Hartogh 

 Overview

Company Description

Den Hartogh Logistics is one of the leading Logistics Service Providers. The family-owned organisation was established in The Netherlands in 1920. As a bulk logistics service provider for the chemical, gas, polymer and food industry, we combine the best elements to create the optimal solution for each situation. Safety and operational excellence are embedded in our culture.

 Den Hartogh has a presence in every region of the world, with premises/offices in 50 locations within 27 countries. Our workforce consists of more than 2,100 people and our modern equipment includes more than 24,000 tank containers, 5,750 dry bulk containers and specialised dry bulk trailers, 300 tank trailers and 650 trucks.

 From our four business units, this project will take place in business unit Liquid Logistics Europe. This business unit is responsible for all the intra-European liquid chemical activities

Project Description

Every week we are loading hundreds of tank containers on the European continent. Of course, we aim to do so at the highest customer service, but also for the lowest cost.

However, the requirements of our customers become more and more restrictive. We have to take into account prior loaded product restrictions, loading place restrictions and specific tank container restrictions. Additionally, we also have to anticipate on the applicable restrictions at destination to load the next job. As you understand, we need many different type tank container to accommodate these requirements.

To allocate the right tank type to a customer order therefore becomes quite a complex puzzle. At the moment, Den Hartogh already uses algorithms to match the order restrictions to find the right tank, but these algorithms are starting to get outdated and are very rigid. E.g. the algorithms only state if the tank container is suitable, but not which of the suitable tank containers is most optimal to use.

In this project we would like to investigate if we can improve the automated tank allocation suggestion, preferably with learning algorithms, to further optimize the European tank container network

Goals of the Project

Investigate how to improve the automated tank allocation suggestion, preferably with learning algorithms, to further optimize the European tank container network.

Deliverables

Msc thesis report

Essential Student Knowledge

–

More information: escf@tue.nl