Text mining, Natural Language Processing (NLP) or Artificial Intelligence (AI), they are still hardly used to improve (environmental and/or process) safety in the chemical industry. In general, chemical companies make little use of the potential offered by new data technology.

That is a missed opportunity! With the huge amount of available data volumes, there is an opportunity for safety prevention, such as the identification ofweak signals, with which you can prevent possible incidents and thus minimise suffering, damage, complaints, costs or downtime of installations.

SDN has placed the topics 'Innovation in Digitalisation' and 'Big Data' on the innovation agenda as a result of interviews with frontrunners from the chemical industry, ongoing research at companies and scientific literature. The application of (among other things) text mining, NLP and AI capabilities offers the chemical industry the opportunity to find hidden patterns and 'weak signals' in large quantities of (unstructured) data. These have been overlooked by humans until now.

SDN is looking for partners and funding to start (research) projects. The goal: to significantly reduce the number and severity of incidents in the chemical industry!
Examples of what these (research) projects might include are:
- The application of new types of (data-driven) technology,
and
- the development of new safety concepts and tools for analysing large quantities of data.

With innovations to 'mine' and analyse data, we want to improve methods for hazard identification on the one hand, and identify 'abnormalities' in installations and processes at an early stage on the other.

Developing these new concepts is a step-by-step process. The first step is to develop advanced technology for descriptive analysis; data mining to understand the past and answer the question "What happened?". The second step is to develop predictive analysis, statistical modelling and forecasting techniques to understand the future and answer the question: "What might happen? The third and final step is to develop prescriptive analysis, which uses optimisation and simulation algorithms to give advice on possible outcomes and answer the question: "What should we do?

This year (2021) and in 2022 the SDN will start with a first step in projects (descriptive analysis) in cooperation with VOTOB and Brightsite (Chemelot, Maastricht University, TNO and Brightlands Campus).

class="img-responsive
Innovations in digitisation and big data
Publication date:
6 December 2021