Making optimisation systems work in practice
I work with optimisation systems for energy and operational assets — with a focus on making them reliable, understandable and robust in real-world operation.
What I do
I work with optimisation systems for energy and operational assets — with a focus on making them reliable in practice.
- Making optimisation behaviour consistent and reliable across realistic operating scenarios
- Diagnosing inconsistent outcomes between similar runs
- Clarifying trade-offs between competing constraints and objectives
- Making model outputs and trade-offs easier to explain and trust
My work typically sits in the gap between theoretically correct models and systems that actually need to work reliably in production.
Typical problems I work with
- The model works most of the time — but fails in specific cases
- It is not fully clear why results change from run to run
- Trade-offs between constraints are difficult to understand
- The system is hard to debug, validate, or communicate
Previous application: workforce scheduling
One application area has been workforce scheduling, where balancing coverage, compliance, fairness, and preferences requires transparent optimisation trade-offs.
This reflects the same type of modelling challenges found in energy systems, where physical constraints, uncertainty, and market dynamics must be handled consistently in operational decision-making.
My primary focus has been on energy-related optimisation systems, where I apply this modelling and diagnostic discipline to ensure robust and reliable behaviour in real-world operation.
Background & experience
My name is Kenneth Leerbeck, and I hold an MSc in Engineering from DTU, with a background in quantitative modelling, dynamic systems, and optimisation.
I work with constraint-based optimisation models across domains, including workforce scheduling and energy systems, where complex constraints and trade-offs need to be handled in practice.
Much of my work has focused on understanding why optimisation models behave inconsistently — and on building structured approaches to analyse and stabilise them.
The focus is not just on solving the model, but on ensuring that the system behaves reliably and predictably across realistic operating scenarios.
This typically involves bridging the gap between theoretically correct models and systems that need to perform consistently in real-world analytical and decision-support contexts.
Contact
If this resonates with what you're working on, feel free to reach out. I'm always open to discussing concrete optimisation challenges — whether it's a quick exchange of ideas or something more hands-on.
Email:
kenneth@leerbecksolutions.dk
LinkedIn:
View profile
Or use the form below: