Case Studies: Moving Toward Industry 4.0

Session Chairs:

• Kushal Sinha, Abbvie Inc.
• David Couling, Corteva Agriscience

Session Description:

Computational power, interconnectivity, and analytical capabilities are ever-increasing, and companies have begun to capitalize on these themes, resulting in increased automation, deeper system understanding and faster decision making.  While these trends are still emerging in the chemical industry, their impact on the development of chemical processes could be profound.  This session focuses on several case studies within the chemical industry where these principles have been applied.

*All session and speaker information is subject to change pending finalization

Schedule:

Abstracts:

Utilizing in-situ Analytical Measurements for Batch Process Analysis and Control

Bryon Herbert, Corteva Agriscience

Process analytical technology (PAT) is an established field within the chemical and pharmaceutical industries to directly measure high value and high throughput manufacturing processes.  The goal of any in-situ analytical measurement system is to generate compositional information for both process control and process understanding with timely, actionable information.  Any installation is justified by contributing toward advancing manufacturing concepts such as quality by design, real-time release, just in time manufacturing, lean manufacturing and even right to operate.

Successful implementation of PAT methodologies starts at small scale with feasibility studies and process development, carried forward into the pilot plant and implemented at manufacturing scale.  Knowledge is gained through each step to ensure the consistency of the reaction characteristics and to identify factors imparted upon scaleup.  The goal is gaining understanding of the compositional changes and sample makeup within the chemical process as each successive reaction builds in value.

Building upon advances in data science and informatics, PAT measurements can introduce a direct feedback for control systems.  Coupling the inputs and sensory outputs of the plant data systems with the real-time analysis can lead to identifying new schemes for monitoring key manufacturing steps or determining attributes to avert costly upsets.  Bringing all of the elements together can help realize the goal of being preemptive versus reactive.

Batch Data to Multivariate Analysis in 30s: Eliminating Data Silos in Process Development

Timothy Gardner, Riffyn, Inc.

Today's breakthroughs lie deep in the midst of complex, multivariate data sets that span scientific disciplines and time. Experimental anomalies and fundamental discoveries often go unnoticed because they are buried in uninterpretable spreadsheets, inaccessible databases, or excessive experimental noise. The cloud-based Riffyn software structures and links experimental designs and measurement data for analysis within seconds after it is collected.  This talk will discuss examples how bioprocessing groups use these capabilities to integrate data, identify unexpected correlations, uncover root causes of error, improve process quality, and deliver right-first-time technology scale-up and technology transfer.

Leveraging Natural Language for Accelerating Workflows in Process Development

Praful Krishna, Coseer Inc.

Artificial intelligence (AI) and natural language processing (NLP) together have immense potential for automating complex workflows in process development. This talk will present a tactical approach to such automation, the pitfalls usually faced and how to avoid them, and the key success factors to ensure high ROI from such a project. The talk will take real world examples from pharma, energy and other sectors to illustrate power of AI in interpreting historical data, natural language search, structured knowledge management and other such use cases.