Internship Proposal: Root-Cause Analysis for Emulation Projects

  • TBA Group
  • Den Haag, Nederland
  • dec 12, 2022
Internship Quantitative Logistics Operations Research

Job Description

Background

In the last 15 years TBA has been involved in a large number of emulation projects that aimed to test and tune the performance and quality of terminal operating systems (TOS). The TOS is a critical piece of software running on the container terminal being responsible for operation (e.g. vessel planning, container decking, equipment dispatching, gate and rail handling, etc.). The performance and quality of the TOS can have a huge financial (and safety) impact on the terminal. Therefore, it is essential to test and tune it when certain changes are carried out on the terminal, such as expansion or TOS updates. In order to continuously test the TOS, TBA created a virtual container terminal, called CONTROLS, that aims to act as a real terminal that is able to communicate with TOS using the same interface as in reality. During our emulation projects we connected the real TOS with the virtual container terminal and by running large number of experiments we succeed to find TOS software issues, TOS planning issues, and advice on TOS settings that led to higher performance. In order to analyse the results of these experiments we are using a KPI (Key Performance Indicator) tool that creates a large number of charts, each focusing on certain aspect of the terminal (crane or truck performance, crane or truck status, yard performance, etc.) and log files produced by CONTROLS and TOS. Identifying all the issues, and finding the root causes in the system involves checking all KPIs and log files, which is a very intensive and time consuming activity.

Research goal

The goal of this research is to provide an approach and tool that improves and supports the root cause analysis for emulation projects. The tool should be able to find the root causes of the issues, and address the root causes that have been identified.

Methodology

This research follows a data-driven methodology, which is an exploratory approach that analyses data produced by emulation experiment and it makes strategic decisions based on data analysis and interpretation.

Length and type of research

6 months, Master thesis

Academic background

Computer Science or Econometrics and Operational Research

More info

For more info please contact Csaba Boer (csaba.boer@tba.group)