Automated planning, monitoring, inspection and control with AI

RISE in Västerås and Automation Region invite you to a workshop where we discuss the possibilities of AI in your business. 

It is becoming cheaper and easier to collect and use process data, images and video to automate the planning, monitoring, inspection and control of different sections of production. Software is getting better and easier to use to take further steps towards flawless automated, event-driven production.

Today, large amounts of measurement signals and process data are stored in the process industry. Video is used to control and monitor processes. Video and photographs are used to check products and as sources of information for assessing quantitative measures such as internal process variables and product quality. Better algorithms, better computer hardware and more data have made it possible to use artificial intelligence (AI) to monitor, automate and simplify various operations in industry. Images and video can be analysed through machine vision to replace or assist personnel in carrying out inspections, monitoring and control. In some cases, it may be better to automate control because it takes a long time to train staff to become proficient at producing consistent quality. Measurement signals and process data can also be used in this context.

The challenge is to find ways to use these new tools in the process industry

Production and maintenance planning can be carried out automatically and updated as new information/data becomes available. Optimisation can be used to find the best parameters or settings. This can save time and often find solutions that are better than a human can achieve. Planning and optimisation can also be carried out and updated at the second level using, for example, model predictive control.

Machine learning and machine vision can be used in many contexts:

• To predict process parameters and signals, and/or classify or detect deviations 

• To solve various problems in the process industry, both where monitoring, inspection and control are currently performed manually, and where it becomes too expensive or difficult to assign personnel 

• Online for active control of a process if applicable 

• To present information in a clearer way during production and when monitoring or troubleshooting 

• Different types of visualisation can make it easier to get an overview and interpret large amounts of data, images and video in a short period of time, which can normally be time-consuming to go through. In troubleshooting, for example, it can be valuable to be able to quickly find the cause of an error.