Lab Glassware Washing: Cobot Cells Between Test Batches

Can a cobot cell take over glassware washing between batches?

Only in part, and the realistic part is narrower than most brochures suggest. Collaborative robots can move glassware between a staging rack and a washer, and research teams have shown arms placing glass into dishwashers. What I could not find is a documented production cell that washes, loads and unloads glassware end to end without staff.

For a commercial or contract lab running repeated test batches, the useful goal is to cut the repetitive handling that eats staff hours. Technicians still set up, inspect, and decide what goes back on the shelf. The arm takes the lifting and placing that repeats the same motion again and again between runs.

This guide covers how to map the wash-and-rack cycle, where a cobot cell fits without blocking staff, and which checks to run before the first live batch.

How do you map the wash-and-rack cycle first?

Write the cycle as a sequence of handoffs: soiled glassware collected from benches, sorted by type, pre-rinsed if your procedure calls for it, racked, washed, dried, unracked, inspected, and returned to stock. Time each step across several full batches with a stopwatch, and note who touches each item and where items sit waiting.

Those handoffs decide what the robot is asked to do. A fixed pick-and-place at one rack is a different job from a mobile run that crosses a corridor. Most of the mapping effort goes into finding the waiting time, the bottleneck at the washer door, and the items that still need a human eye.

What do the robot arm demos actually show?

An arXiv paper on the WARA Robotics Mobile Manipulation Challenge describes teams building mobile systems that move glassware carts and load an industrial dishwasher. Vision and grasping placed clear glass bottles and beakers onto dishwasher pins, and unloading stayed manual. The challenge was designed with AstraZeneca, according to the search summary of that paper, and one entry from Orebro University used a Franka Emika Panda arm for the loading task.

Vendors make broader claims. As of November 2026, Dobot's research lab page describes cobot arms for loading and unloading, centrifugation, and chemical processing, plus a mobile unit that pairs a cobot with an AGV. A RB-THERON+ mobile manipulator is listed for washing and drying laboratory materials. These are vendor claims that I could not confirm through independent testing. An ABB YuMi mobile dual-arm demo at the Texas Medical Center covered centrifuge loading, pipetting, and tube handling, and did not cover glassware washing.

Read these as evidence that the motions are possible, not as proof that a given cell will hold up in your building with your racks.

Where does the cell fit without blocking staff?

Put the cell on the route the work already takes, not in the path of the people who walk it. Keep the washer door, the staging rack, and the return shelf in one short arc, so the arm moves between them without crossing a bench aisle. Give technicians an access point to the washer and a physical boundary around the arm's working envelope.

Plan the people handoffs too. If a technician must pull a rack mid-cycle to check a flask, decide in advance whether the arm pauses and where hands are allowed to enter. Test the stop method with the cell powered and staff walking their normal route, not in an empty room on a quiet morning. Layout choices are far easier to change before anchors go into the floor.

When is a cobot cell the wrong tool?

Low volume is the first warning sign. If a lab washes only a few racks a week, the handling time may not justify a cell, and a well-run manual bench with a standard washer may serve better. The vendor volume threshold discussed below is one reason to measure your volume before you ask for a quote.

Glassware variety is the second. A cell tuned to one flask size needs new grippers and rack inserts when the catalog changes, and fragile, odd-shaped or heavily etched pieces may still need a technician's hands. A third is the washer itself. If racks still need manual loading and unloading at the washer door, the arm may save only a fraction of the walking and lifting.

What do the washer figures prove, and what do they not?

Vendor material offers useful numbers, with caveats. Labconco's comparison, a vendor study the search summary dates to 2013, reported that an automated washer saved about 211 liters of water and about 24 minutes of wash time versus hand washing, and that the time figure excluded drying. In the same material, a Labconco product manager said that for labs washing more than 25 flasks a day, a washer pays for itself within about two to two-and-a-half years. That payback refers to the washer, not to a robot arm added around it. Water use was 20 gallons by hand versus 13.6 gallons by machine for 30 items.

Miele reports that 60 items can be reprocessed by machine in 24 minutes, about 59 percent time saved, and says manual handling raises breakage risk. One lab user's blog post describes hand-washing a single batch as taking at least 30 minutes, which is one anecdote and not a measured study. All of these are vendor or single-source figures, so use them as hypotheses for your own timing study, not as a payback you can bank. I found no independent or peer-reviewed labor-savings study for this use, and no documented dollar savings from a site, so I have left dollar figures out.

What should you check before the first live batch?

Run these checks in order, and do not start a live batch until each one has an owner and a recorded result.

• Dry-run the full cycle with empty racks, timing each handoff against your map.
• Load-test with clean, non-critical glassware, confirming each piece seats in its rack and clears the gripper.
• Confirm the washer's cycle-end signal reaches the cell controls, and that a failed cycle halts the arm.
• Verify the stop method with staff walking their normal route while the cell runs.
• Start a breakage log from day one, recording item type, rack position, and cause.
• Record baseline staff hours for one comparable batch, so the first comparison is honest.
• Name a single owner for the go or no-go decision.

How does Service Robot Co. approach a lab glassware cell?

Service Robot Co. is an OEM-neutral integrator for US businesses. We choose the robot for the task across manufacturers, then lease it to you through financing, deploy and integrate it, train your staff, and service it under RoboCare, our service plan, backed by regional service engineers in every state we service. On-site dispatch and remote triage are part of that service.

The engagement starts with a site assessment, where the cycle map and layout checks above are done on your floor. Our offer is a 3-year lease with one fixed lease payment quoted after that assessment. We would expect a lab to start with one leased unit on one washer, with go-live support, before deciding on a larger cell. A given washer and rack set is judged during the site assessment, not in a brochure.