Optimization of robotic liquid handling as a capacitated vehicle routing problem

Area of Science:

• Automation and Robotics in Chemistry
• Computational Chemistry and Cheminformatics
• Laboratory Automation

Background:

• Automated liquid handling is crucial for high-throughput experimentation.
• Current methods often lack optimal efficiency, leading to longer execution times.
• Reducing liquid handling time is key to accelerating scientific discovery.

Purpose of the Study:

• To develop and evaluate an optimization strategy for liquid handling operations in automated chemical laboratories.
• To reduce the overall execution time of complex liquid handling tasks.
• To demonstrate the applicability of logistics optimization techniques to laboratory automation.

Main Methods:

• Formulating liquid handling tasks as a capacitated vehicle routing problem (CVRP).
• Applying heuristic solvers from logistics and transportation planning to optimize task sequences.
• Testing the strategy on an 8-channel pipette with individually controllable tips and various labware formats.
• Validating the approach in a real-world high-throughput materials discovery campaign.

Main Results:

• Achieved up to a 37% reduction in liquid handling execution time for random tasks compared to baseline sorting.
• Demonstrated robust optimization performance across different labware formats (well-plates, vial holders).
• In a materials discovery campaign, 3 minutes of optimization reduced execution time by 61 minutes versus the best sorting strategy.

Conclusions:

• The proposed CVRP-based optimization strategy significantly reduces liquid handling execution time in automated labs.
• This approach offers substantial improvements in throughput and efficiency without requiring hardware modifications.
• The method provides a practical and scalable solution for accelerating combinatorial experimentation in diverse scientific fields.