Automated intermodal planning– CTV B.V.

 Overview

Company Description

CTV is a logistics service provider specialized in intermodal container transport. We are trying to find the best transport route by combining train, barge (inland vessels) and truck as efficiently as possible. From our offices in Venlo, Duisburg and Basel, we are offering the intermodal service to customers throughout Europe.

Daily we transport about 500 containers via several intermodal routes. Most of the containers are entering of leaving Europe through the main ports Rotterdam, Antwerp and Hamburg. We are orchestrating the whole intermodal transport between the ports and the warehouses.

Project Description

The container planning process starts when a customer pre-announces that containers are expected to enter the port on a certain vessel. When we have this information, we start making the intermodal planning. This means that the containers should be planned on a certain train or barge. To do this we have to take certain variables into account like:

– Arrival time vessel

– Free time in port

– Lead time of the modality

– Transport schedule of the modality

– Etc

When the initial intermodal planning is created, the container passes several stages where disruptions can occur like vessels or train/barge delays. When this happens the whole planning for the container needs to be checked and possibly renewed.

Because there are a lot of irregularities in the supply chain, creating the delivery planning is dependent on a lot of information and therefore very difficult. Lacking a proper planning, can lead to a lot of issues like storing costs or shortage of warehouse personnel. Having a good planning is therefore crucial.

Up until now, CTV is creating the intermodal planning and re-planning manually. However, most of the information is available in our IT-systems and therefore it should be possible to create and update the intermodal planning automatically.

This project revolves around creating an algorithm to find an optimal intermodal planning for all containers entered in the system. Once disruptions occur, the algorithm should automatically give proposals to adjust the planning. We are especially interested in how AI and deep reinforcement learning (DRL) methods should be combined with classic algorithms from transport and logistics

Goals of the Project

• Automate the intermodal planning by developing an algorithm which can automatically create and update an intermodal planning taking all variables into account.

Deliverables 

• A master thesis report explaining and showcasing the benefit of the algorithm compared to a manual planning.
• The algorithm that continuously calculates the best intermodal planning by taking all variables of the whole intermodal flow into account.
• The algorithm should be presented to the company in a way that the IT-supplier is able to integrate this into the Transport Management System. This means that it does not have to be coded 1-1 in the systems of CTV, but it should be developed in such a way that the IT-supplier can rebuild in in CT’s Transport Management System.                                                        

Essential student knowledge

• Forecasting process
• Creating algorithms

More information: escf@tue.nl