Coming to Grips with the Reality of Automation in Analytical Laboratories

By September 21, 2017

Did the promise of automation fall short?

Now that I’m a product manager, I’m fortunate to visit many analytical labs during the course of a year. One thing that astounds me about so many of these laboratories is the number of valuable automation systems that I see sitting under-used, or – worse yet – completely idle.

When I think about automation, I remember my days as a lab chemist. And as any laboratorian can relate to, one of my most important but time-consuming and tedious tasks was manually pipetting samples. You can imagine my excitement the day we received a programmable multi-channel pipette. That device was so easy to use and saved me so much time, I practically wept tears of joy.

If something as simple as a programmable multi-channel pipette can save so much time, what about the potential of these under-used automation systems that I see every day? Why are they collecting dust?

To get at the answer to this question, I found some research that I would like to share with you.

And the survey said…

Bioanalysis Journal recently published the results of a subscriber survey on the value of automation in today’s laboratory. What they found is quite interesting.

From what the survey tells us, the advantages of automation are clear:

  • Increased productivity
  • Consistency (I imagine that this is critically important especially for a workflow that involves several scientists from start to finish.)
  • Reduction of errors
  • Freeing scientists to concentrate on more rewarding and more valuable tasks

I can relate to these findings. Even with my brand new multi-channel pipette, I was tied to the bench longer than I should have been or wanted to be. This ate into my time that would have been better spent acquiring and interpreting results.

What are the hindrances to lab automation?

If, as the survey shows, the benefits of automation are so compelling, then why are so many automation systems collecting dust?

The answer seems to be cost along with integration/technical difficulties. The reality is that these systems are a large investment in both money as well as time. I visited with companies where, instead of scientists, they hire an engineer whose specific role is to keep the system up and running and to understand the software in order to write script.  These are very real hurdles that need to be overcome to justify purchasing automation tools.

What are the processes that benefit most from automation?

My lab work involved doing a lot of solutions and dispensing, something that the multi-channel pipette was ideal for. If I’d had a really complicated workflow to manage, one with many steps that may be hard to train new hires on, and one that was used as a release test where consistency is critical, then it’s easy to see that greater automation would have been worth it.

It is no surprise, then, to see that the top 3 processes that benefit from laboratory automation are sample preparation and extraction, data analysis, and separation.

The promise versus the reality of automation

During one of my recent business trips, I had a conversation with an analytical lab manager for a large biopharmaceutical company. I asked her what they expected from automation. She told me that she expected to save her time. She had been evaluating automation platforms, but she was slightly disappointed as she hadn’t found the walk-away solution she wanted. She noted that the automation instrumentation actually wasn’t the problem.

The real problem was that the consumables used around the workflows were not scaled for automation platforms – and because they weren’t, her staff continued to spend two to four hours each day preparing solutions, reagents, and washes for what was supposed to be an “automated” workflow.

So, basically, her “automated workflow” was broken and the broken link was the sample preparation phase. Sample prep!

This revelation points to the need for greater cooperation between instrument vendors and those performing the sample preparation analysis, a position that 89% of respondents would agree with.

Stay tuned for more to come on this topic!

In the coming series of posts on automation, we’ll explore the ins and outs of this topic in detail. Scientists from Waters will discuss their experiences with automating workflows as will several automation vendors. We’ll take a careful look at the issues that plague the adoption of automated workflows and what can be done to address them. Finally, we’ll share some ideas and suggestions for thinking about automating workflows in ways that just might make you the superhero your laboratory has been looking for to take it new levels of efficiency and profitability.

For our next blog post, we’ll take a look at one Waters scientist’s journey in applying automation and what she learned along the way.