Compositional optimization of large-scale discrete event systems

Licentiate thesis, 2018

Optimization of industrial processes such as manufacturing cells can have great impact on their performance. Finding optimal solutions to these large-scale systems is, however, a complex problem. They typically include multiple subsystems, and the search space generally grows exponentially with each subsystem. This is usually referred to as the state explosion problem and is a well-known problem within the control and optimization of automation systems.

This thesis proposes a new method of solving these optimization problems using a compositional optimization approach. This integrates optimization with techniques from compositional supervisory control, dividing the optimization of subsystems into separate sub-problems.

The key to this approach is the identification of local behavior in subsystems, behavior that is independent of all other subsystems. It is proven in this thesis that this local behavior can be optimized individually without affecting the global optimal solution. This is used by the approach, to reduce the state space in each subsystem, and then to utilize these reduced models compositionally when the global optimal solution is computed.

Results in this thesis show that compositional optimization efficiently can generate global optimal solutions to large-scale optimization problems, too big to solve based on traditional monolithic models. It is also shown that these techniques can be applied to several industrial applications, e.g. in logistics, manufacturing etc.