An Industrial Engineering Approach to Laboratory Automation for High Throughput Screening

EXTENDED ABSTRACT

Across the industry, there are a variety of approaches to laboratory automation for High Throughput Screening (HTS). At Sphinx Pharmaceuticals, the principles of industrial engineering have been applied to systematically identify and develop those automated solutions that provide the greatest value to the scientists engaged in Lead Generation.

The Handbook of Industrial Engineering defines Industrial Engineering as the field “concerned with the design, improvement and installation of integrated systems of people, materials, equipment and energy.” This description is certainly applicable to the discipline that has become known as High Throughput Screening. Many approaches to laboratory automation for HTS have focused on a single aspect, be it robotics, miniaturization, or a particular detection format. A much more successful approach recognizes the necessary integration of not only hardware and software, but of people, materials and process as well.

At Sphinx, this approach has been organized using operations analysis and targeted process improvements. Operations analysis has involved stratifying assays according to their complexity, throughput, assay type and unit operations.

With the proliferation of new detection technologies available, the number of potential assay permutations continues to increase. When one considers the variables of biochemical vs. cell based, number of reagent additions, type and number of incubations, plate format and signal detection method, the number of possible assay configurations exceeds 10,000. However, as shown in figure 2, as few as twelve different unit operations can describe every assay run at Sphinx over the last two years.

Once unit operations have been identified, the suitability of equipment available to perform those operations can be assessed. Conducting an equipment standardization exercise has the following benefits.

If the workstation approach to automation is utilized, then unit operations can be decoupled. Capacity can be assessed independently for each unit operation. Process improvements can be targeted at specific unit operations, with minimal or no impact on other unit operations.

OPEN IN VIEWER Figure 1.

SUMMARY

Automate the process, not the assay. Of course, this requires knowledge of the process. Assay specific improvements will have short term impact. Process improvements will continue to provide benefit as long as that portion of the process continues.

Identifying unit operations is the key. Once assay processes are broken down into unit operations, each unit operation can be standardized and optimized. Capacity can be assessed by unit operations and assays can be assembled.

Standardize until it hurts…but be flexible. The attitude should be “Why can't we standardize this method?” There will always be assays that require exceptions, but those exceptions should be justified, and not just for the sake of doing things differently. In fact, standardizing wherever possible will make it easier to be flexible on those issues that really matter.

OPEN IN VIEWER Figure 2.

Look for the bottlenecks. Using the concepts of cycle time and throughput will allow the determination of the bottlenecks in the process. Targeting process improvements at these bottlenecks will provide true increases in capacity and capability.

Remember the bottom line. Ideally, laboratory automation will provide more consistent and reproducible results than manual operations. However, the principal driver behind automation will always be cost. If the total life cycle cost of an automation project does not provide a cost advantage, it should probably not be pursued.

ACKNOWLEDGEMENTS

Members of the HTS Automation Group at Sphinx Pharmaceuticals