In this project, the Saxion research group Ambient Intelligence (AmI) investigates for Scania how the data from the control computers of the production line can be used to prevent production disruptions.
This project is a follow-up to the TFF Uptime Improvement Scania project for reducing production stops due to disruptions in moving the truck assemblies down the production line. The aim was to predict failure rates by applying a top-down, model-driven approach and a bottom-up, data-driven approach. Despite the steps that have been taken in this respect, the project conclusion was that there is still insufficient detail data available to properly predict outages.
To take the next step towards predictive maintenance, two innovations are added to this project. As a first innovation, the operation of the Carriers who carry the truck assemblies through the production line will be monitored. This data will be linked to the top-down, model-driven analysis of the Carrier in order to find relationships between the mechanical analysis and the usage data. To this end, the Carrier is provided with a monitoring platform to which various sensors can be linked, so that extra detailed data about the operation and disruptions of the Carrier becomes available as an additional data source. For this, sensors are linked that measure the temperature and current/frequency of the motors. The challenge here is to collect this sensor data from the moving Carriers and reliably send it to the research server in an industrial environment. The second innovation is the research into which extra data can be made available from the control cabinets that control the Carriers. These control boxes contain the control electronics (PLCs). The challenge here is to obtain measured data from a closed system, the operational functioning of which cannot be disrupted in any way.
These innovations, which will be implemented and experimented with, will provide knowledge that will help us answer the question of how Scania can grow from corrective to predictive maintenance for the truck assembly production line.
Topic
Smart Industry, data-analyse, sensortechnologie, predictive maintenance, monitoring platform, data extractive.
Program objectives
The aim of the project is to determine how data and analyzes can provide insight into the underlying causes of failures, how specific process parts can be improved and how failures can be predicted.
In addition, answering the question: how can the synergy of data analysis and top-down driven model-based approach be applied for predictive maintenance?
Partners
Scania Production Zwolle B.V., Saxion lectoraat AmI(penvoerder), Hoogeschool Windesheim.
Duration
March 1, 2020 untill September 1, 2021.
More information
Financing
This project is financed by TechForFuture.