Programme

• Short introduction in AI in Big Data Analytics
• Introduction of phases for Data Science (Sandboxing, devOps, Scoring productive industry processes
• How to start smart? (based on an industry example)
• Remarks about safety and security

Michael Ehrmantraut, IBM Watson & Cloud Platform, Executive IT Architect,
Member of the IBM Technical Expert Council (TEC), IBM Analytics, Germany

• Data journeys at customers
• AI, visualization & preparation
• Learnings and critical notes

Mirko Schnitzler, Owner, InfoTopics, Germany

• The Digital Buffet: Enterprise IT between choice and essentials
• Nutrition fact 1: Convenience food: The Digital Platform
• Nutrition fact 2: Home Cooking: The API economy
• Nutrition fact 3: Recipe exchange: The Academy
• The menu: How Evonik tries to feed on a healthy diet

12:45

Lunch break, interactive networking and discussion @ "topic tables" in exhibition area

• Citrine is the world’s only AI-powered data platform for materials discovery
• One of the biggest challenges facing the materials industry is that materials lag behind product development. The demand for complex next generation products requires materials be developed much faster
• Data is key to speeding up materials development. Developing materials faster will require managing and using data more effectively, which includes consolidating data into a single consistent searchable format, as well as structuring, storing, and using materials data to harness the power of artificial intelligence.
• Citrine's technology combines a consolidated materials repository, the world’s largest materials dataset, and powerful artificial intelligence to accelerate materials discovery. Discoveries and IP always remain exclusive to the user

Douglas Ramsey, Vice President Business Development, Citrine Informatics, USA

• Benefit from Artificial Intelligence to accelerate processes
• Quality control and maintenance based on real-time decision
• Get a fast implementation of new technologies into your existing
• infrastructure
• Gain from early success in a solid environment

• Physical Analytics methods are often key ingredients in finding the optimal Smart Data solution given the starting data set or the data-driven question to answer
• Using theory-based model-driven approaches, a much deeper understanding of the underlying data problem and the interpretation of the results is possible
• A number of data-driven industry use cases that were solved with advanced Physical Analytics methods will be outlined and discussed

15:55

Coffee break, interactive networking and discussion @ "topic tables" in exhibition area

Some areas of R&D do have reasonable data e.g. clinical, and good quantities of it, But often in R&D data is not connected and is not accessible.

What can we do in the areas where we don’t have good data / poorly connected data?

Enrichment semantic, Automated meta data markup

Automated cleaning and quality assessments

Automated data capture and association with context

Human data e.g. speech

How do these technologies play together and how can we enable an ecosystem of systems to work on data and generate meaningful consequences?

The talk will explore these topics and provide case studies from work done with Pharma and biotechs to make the vision a reality.

• Big data in process industry vice or virtue
• Digitalization platforms and integrated data analytics solutions
• Process knowledge integration and hybrid modeling
• Case studies from different industry sectors (biopharma, refineries)

Michael Sokolov, COO, DataHow, Switzerland

• The Problem: Too much + too different Data = too late Insights
• How to capture large amount of data by tens of thousands of sensors distributed over an area as large as the city of San Francisco?
• How to handle around 500 million data records in well over 40 heterogeneous systems - within a time frame of 90 days?
• The Implementation: Off to the Races
• How to act and react in case of problems!

Wednesday, 25 Apr 2018

• The difference between the digital twin and a simulation environment
• The digital twin will be a companion to its physical twin over their entire lifecycle and then is reincarnated
• The concept of the digital chemist and the digital chemical enterprise

• The current state of quantum computing
• The journey ahead
• Applications in molecular screening and big data analytics

Kausar N. Samli, COO, 1QBit, Canada

• Trends and emerging business models in chemicals
• Technology enablers
• Recommended Business Process and IT platform
• Status new business model adoption and case studies

Stefan Guertzgen, Senior Director - Chemicals, SAP, Germany

• The evergrowing availability of data and the emerging hype around advanced machine learning algorithms can be tempting for the aspiring analyst to “skip the basics” and start attacking large datasets he does not really master with an arsenal of tools he does not really understand.
• As a result, the very discerning and well-ordered approach of applying statistical thinking to process and product improvement problems risks turning into pseudoscience.
• Big data is a great asset but in order to meaningfully benefit from it one has to first be able to deal with small(er) data sets, recognize one’s own process in that data, build statistical models that reflect one’s prior understanding and use these models to enhance the current state of understanding and ultimately attain better solutions.
• How is the data being collected? Will the data allow me to answer the questions I set out to investigate?
• Live demonstration of JMP

• Introduction to Self-Service Analytics
• Building an analytics-driven organization
• Broad overview of use-cases from TrendMiner across their customer base

• A standardized Big Data Analytics process to optimize a chemical plants
• Data Driven Modelling with historical process data
• Implementation of a Genetic Algorithm as Real Time Optimizer
• Increased throughput with simultaneous energy savings in an isocyanate plant

Thomas Froese, Managing Director, atlan-tec Systems, Germany
Jasper Rutten, Teamleader Process Automation, Huntsman, Netherlands

Solution concept: Connecting big data analytics and automation systems

Demo of an end-to-end solution of a coating process

Use cases in the process industry: From root cause analysis to process optimization and predictive maintenance

André Bosman, Technical Specialist for Industry 4.0 & Big Data Solutions, acs, Germany/Switzerland
Fredrik Stroem, Advanced Analytics Strategist, Dataiku, Germany