Home " Projects " Expertise Network - BD
Updated:
February 27, 2023 12:58 pm

BD - Looking ahead and preventing thanks to AI & big data

Safety Delta Netherlands - Expertise Network

Find, connect, renew in safety

Our industry has relatively few incidents compared to other industries. If something goes wrong, the risk of harm is enormous. We want to get to zero incidents. We want to make our industry safer, both for the people doing the work (occupational safety) and people around it (environmental safety).

Multiple barriers are often broken in an incident. And, an incident rarely repeats itself in exactly the same way. Therefore, it is quite difficult to see things coming in advance. We want to further strengthen our vigilance. This requires more than just learning from incidents. We also want to prevent alertness from ebbing away after an incident.

To prevent incidents, we pick up weaknesses in our current risk management measures early on. We want to be able to properly assess changes in risks and the state of control measures in our plants at any point in time. We can then take further preventive measures where necessary.

We want to do this based on all available data: technical data and human feeling about how things are going. We are getting more and more data. Data analysis takes time. Our time is scarce. Besides, all the data together no one can oversee. People mainly look at data from their own field and extract information from it. Controlling safety requires broad information flows: from operations, maintenance, technology, engineering, contractors, HSE, etc. Across the pigeonholes, however, you lose information and you also lose information over time.

You want to develop smart tools based on new technology that can merge information from different systems and analyze it automatically.

Theory

We want to move from looking back and explaining (learning from incidents) to looking forward and preventing (learning from information).

Learning from incidents looks at where iet went wrong and what we need to do differently to keep doing well in the future. We solve an acute problem.

Learning from information is about pattern recognition. It answers the question of whether we keep doing the right things in the complex system every time. You can find the weak signals that certain management measures may not be working as thought.

Learning from information is also about making interventions based on context. It looks at what else is at play at the edges of operational activities that influence human behavior, performance and judgment, for example.

What do we hope to achieve?

We want a practical and usable tool that makes our work better, faster and easier. A tool that helps look ahead and prevent.

We include people in the field in this development. Our goal is user acceptance and adaptation within work processes.

The various project participants work together. This enables faster development and leads to less fragmentation through feedback between projects.

Netwerk activiteiten

The BD Expertise Network is a closed project group. To view documents you must log in.

Information for you on projects in which this expertise network is involved or has been involved

  • Research question: Improving (process) safety with big data, AI and machine learning

  • Project: Better Learning Information in Chemistry by Looking Ahead (BLIC Ahead)