On May 6, 2023, BLPredict team researcher Elodie Toufaili gave a talk in Greece atIDEAS 2023, the 27th International Symposium on Database Engineering Applications. She presented an innovation designed to make equipment diagnostics more reliable for maintenance engineers!
Optimizing the performance of industrial processes
In predictive maintenance, the challenge is to be able to visualize the condition of a piece of equipment in real time, and prevent potential breakdowns before they become too critical! To achieve this, Elodie has identified criteria for selecting the specific parameters to be analyzed from the mass of pre-existing data, which is growing in a continuous flow.
How did it work? It represented multivariate time series data using co-occurrence matrices, so as to visualize the state of a piece of equipment and prevent possible breakdowns before they become too critical.
The result? The development of a visual analysis tool for equipment operating status that is easy to understand, and at the same time provides them with information on the general condition of their equipment, to :
- Easier monitoring and faster detection of anomalies
- The assurance of qualified interventions
- Better-maintained equipment
- Optimized industrial performance
Application to an airport baggage sorting system
Let’s take the example of the baggage sorting system at Lyon Saint Exupéry airport (2020 data, Figure 1). The conveyors are equipped with sensors that measure, among other things, motor temperature, conveyor belt speed and baggage weight. The visualization below shows a correlation between the increase in: 1) conveyor temperature, 2) and the baggage jam problem.
In this way, Elodie’s advances are directly in line with the prospect of enriching BLPredict’s features, in order to support both industrial companies and public organizations towards sustainable management of their equipment and infrastructures.
Full article
You can read the full article online, entitled “Information visualization for industrial process monitoring”, carried out in collaboration with the Laboratoire d’InfoRmatique en Image et Systèmes d’information de l’agglomération lyonnaise.
