The Twilight System is an example created for the study of controller synthesis and performance analysis of Manufacturing Systems. This manufacturing system is a simplification of the product handling model that has been created at ASML, the world-leading manufacturer of lithography systems, using similar kinds of peripherals and resources. Our example system contains four resources. Two robots to transport balls. The other two resources are processing stations, to conditioner to ensure a right ball temperature, and a drill to drill a hole in a ball. Each ball processed by the system follows the same life cycle. First, a ball is picked up at the input buffer by the load robot. Then it is brought to the conditioner and processed. Next, the item needs to be transported before the ball cools down too much by either one of the robots to the drill, where it is drilled. Finally, the drilled ball is transported to the output buffer by the unload robot. More details can be found in the attached description. Model-based specification, analysis and synthesis techniques can be used with a model of the system to automate its design trajectory. The models, and their formal semantics and the algorithms to manipulate these models are subjects of research.


The model is specified in terms of resources, activities, and dependencies between them. Such models can currently be annotated with timing information to specify the minimum time difference required between two actions. We want to extend the modelling language to enable the specification of a maximum time between two activities. You will contribute to the semantics and implementation of the extension of the modelling language and tools for the specification of such maximum time differences. To show that this enables correct modelling of the minimum and maximum time differences, the existing Twilight model needs to be extended. From past research on production printers, we know that the analysis tool-kits need to be extended to deal with such maximum time differences. The generalization will enable the tool-set to analyse a new class of problems.

Required Skills and Expertise

We are looking for a student that has at least some affinity with:

  • Programming in: Java (extending the modelling language), C++ (analysis toolkits)
  • Model-based design and semantics of models
  • General interest in Cyber-Physical Systems

Contact information

The project is hosted by the Electronic Systems (ES) group, Electrical Engineering department. If you are interested in this assignment, contact Joost van Pinxten( or Marc Geilen ( for more information.