KU Leuven Logo

Jordy Moors - PhD Research


Welcome to my research page! My name is Jordy Moors, and I am a PhD researcher focusing on improving the injection moulding process through advanced data acquisition and machine learning techniques. My goal is to enhance the quality and efficiency of this widely-used manufacturing method.

What is Injection Moulding?

Injection moulding is a manufacturing process for producing parts by injecting molten material into a mould. It is widely used for producing a variety of parts, from the smallest components to entire body panels of cars.

The injection moulding process consists of several key steps:

  1. Clamping: The mould is securely closed by a clamping unit.
  2. Injection: Molten material is injected into the mould cavity under high pressure.
  3. Cooling: The molten material cools and solidifies into the shape of the mould.
  4. Ejection: The finished part is ejected from the mould.

Injection moulding is crucial due to its ability to produce complex shapes with high precision and repeatability. It is extensively used in various industries, including automotive, consumer goods, electronics, and medical devices.



Production Overview

Figure 1: Production Flow Overview

Context & Motivation

The main motivation behind my research is to reduce waste, recalls, and energy consumption in the injection moulding process. There are several existing issues in the industry that this new system aims to improve upon:

By stabilizing the process control and introducing advanced data acquisition and machine learning techniques, it is possible to address these issues. The goal is to enhance the quality and efficiency of the injection moulding process, making it more sustainable and cost-effective.

Challenges for Industrial Adoption

The adoption of advanced data acquisition and machine learning techniques in the injection moulding industry faces several significant challenges. These challenges are problematic because they hinder the ability to achieve consistent product quality, increase production efficiency, and reduce waste. Addressing these issues is crucial for the successful implementation of new technologies and the overall improvement of the injection moulding process. The primary challenges include:

Plan of Action

The research is structured into several work packages (WPs), each addressing a specific challenge within the injection moulding process. These work packages are designed to build on each other, creating a comprehensive approach to improving the process through advanced data acquisition and machine learning techniques. The detailed plan of action is as follows:

Work Package Overview

Figure 2: Overview of Work Packages


Significant strides have been made in the early stages of the research, particularly in developing a proof-of-concept data acquisition platform and integrating essential hardware components. The following milestones highlight the key achievements so far:

These achievements lay a solid foundation for the next phases of the research. The focus will now shift to expanding the dataset through extensive experimentation, developing advanced machine learning models, and implementing adaptive control systems. The ultimate goal is to create a robust, adaptable system that can be widely adopted in the injection moulding industry, driving improvements in quality, efficiency, and sustainability.


The research has made significant progress in developing and testing an advanced data acquisition and control system for the injection moulding process. Key outcomes of the research include:

To illustrate the effectiveness of the developed systems, a video demonstration has been created. The video showcases the complete injection moulding process, highlighting the integration of data acquisition and automated control:

Video: System producing a part, labeling it, and performing automatic inline quality control

The video demonstrates the following steps:

  1. Part Removal: The machine automatically removes the parts from the mould, ensuring precise handling to avoid damage.
  2. Labeling: An inkjet printer labels each part with a unique identifier, facilitating traceability throughout the production process.
  3. Flipping and Positioning: The parts are flipped and moved into position for the next stage, ensuring correct orientation for subsequent quality checks.
  4. Quality Control: High-resolution cameras capture images of each part, which are analyzed to detect any defects or anomalies. This real-time quality control step ensures that only parts meeting the quality criteria proceed further.

This integrated approach enhances the quality and efficiency of the injection moulding process and provides a scalable solution that can be adapted to various manufacturing scenarios. The successful implementation of these preliminary systems marks a significant step forward, laying the groundwork for future enhancements and more sophisticated data analysis techniques.


The future work will focus on building upon the foundation laid by the current system, with a particular emphasis on expanding data collection, developing advanced data analysis techniques, and implementing adaptive control systems. The detailed plans for the next stages of research are as follows:

The ultimate goal is to create a robust, adaptable system that can be widely adopted in the injection moulding industry. By leveraging advanced data acquisition, machine learning, and adaptive control techniques, this research aims to revolutionize the injection moulding process, making it more efficient, reliable, and sustainable.

Future Goals

Figure 6: Future Goals - Dynamic Adjustments Using Machine Learning


The research has made substantial progress in laying the groundwork for a revolutionary approach to the injection moulding process. The integration of advanced data acquisition and automated quality control systems has shown promising results in preliminary tests. These initial successes provide a strong foundation for future work aimed at further enhancing the efficiency, reliability, and sustainability of injection moulding.

Several key conclusions can be drawn from the research conducted so far:

The research highlights the potential of integrating modern data-driven techniques into traditional manufacturing processes. By leveraging data acquisition, machine learning, and adaptive control, it is possible to transform the injection moulding industry, making it more efficient, sustainable, and capable of producing higher-quality products.

Future work will build on these foundations, with the aim of developing a comprehensive, adaptive system that can be widely adopted across the industry. The anticipated outcomes include reduced waste, improved product quality, and enhanced use of recycled materials, contributing to more sustainable manufacturing practices. The successful completion of this research will not only advance the field of injection moulding but also set a precedent for the application of similar techniques in other manufacturing processes.

Dissemination & Administration

The results of this research are being disseminated through various channels to share findings with the global scientific community and industry stakeholders. Key dissemination activities include:

By disseminating findings through international conferences, publications, workshops, and the final dissertation, this research aims to contribute significantly to the advancement of injection moulding technology. The goal is to foster innovation, improve industry practices, and promote sustainable manufacturing solutions worldwide.