-
Results:
The experiment has achieved the following objectives:
Provide a tool able to help operators to find the exact sheet they need for production line.
The tool can significatively decrease time usually needed to find out, handle and pick up the sheet that is needed from the production line.
The tool is expected to self-learn over time.
Shorten the engineering cycle => automatic feedback and decision making
Save time and reduce cost => reduce manual operations, automation process
Automatic recommendation for printing configuration and additive manufacturing feasibility => Covering all the lifecycle (ideation, request, design, simulation …)
Increased Overall Equipment Effectiveness
Reduced Carrying Cost of Inventory
Increased On Time Orders
improved accuracy of robot arm during milling operations
Improved Reliability of Maintenance and Service
Reduced Working Capital
Improved Customers Acceptance
Decreased Number of training data to adapt the model to new applications
Decreased Training time
An adaptive support system for training and guidance in assembly making use of an AI based adaptive algorithm based on both productivity and quality as well as operator capacity and needs
Score the quality of measurement
Detect failures and their root causes
Provide methods for the continual learning of AI model
Provide a specific user solution designed to help decision making of operators
Provide automatically feasible resource suggestions to the system designer
Reduce time used for system design and reconfiguration planning
Allow more alternative resources to be considered during design and planning, leading to potentially innovative solutions
Reduce humane errors in search and filtering
Automating the search and filtering of feasible resources and resource combinations for specific product requirements from large search spaces.
Automating the identification of required reconfiguration actions on current layout.
Deployment of an IIoT platform serving IILAB demonstrators
Increased real time visibility on a production system
Increased flexibility and efficiency of a production system
Increased robustness of operations of robotic mobile manipulators
Real-time Production Data Monitoring
Industrial Autonomous Systems Reliability
Enhance production and control strategy with AI,
Improve Worker’s safety using AI to identify unsafe working conditions
Improve predictive maintenance accuracy
Enable remote and highly interactive control strategy for working environment
Improved fault detection
Increased facility availability
Reduced maintenance service costs (for the customer)
Reduced analysis & diagnosis times (of the Maintenance Service provider)
Monitoring and prediction of water consumption
Automatic detection of water leakages
Optimization of the water pipes network maintenance and installation planning
Automation and integration of the manual Quality Control (QC)
Objectiveness and precision of the QC
Productive increase of QC supervisors
Online prediction of the maximum number of pieces to be produced by a specific machine according to the current status of the tooling and the readings of its control parameters.
Optimization of machine control parameters configuration to maximize the tooling life while the quality of the production is maintained.
Maximization of the tooling usage time of every machine, while maintaining the machining quality.
Optimization of machine control parameters configuration to maximize the tooling life while the quality of the production is maintained
Coordination of the tooling change in all the machines. This objective would face the tooling change considering the impact of machines among themselves, instead of just one by one.
-
Results:
Optimize the Handling
Reduce Technical Checks by Humans
Improve data quality
SWARM platform aims to enable building blocks to digitalize the life cycle, project management, collaborative conception and additive manufacturing prototyping.
It creates a unique and consistent source of data across the entire product lifecycle form the idea to production, including customer feedback, training and after-sale service.
Better life cycle and project management for plastronic product and automatic generation of the control process on the basis of specifications and other project documents.
The experiment provide a useful tool to help the enterprise to optimize the inventory and the use of resources, support components purchasing, support day to day production management on shop floor, support on-time delivery, improve quality.
Higher product quality, higher production flexibility, optimized costs
Optimization of maintenance intervals
Reduction of regular service intervals such as (scanner cleaning, ..) by replacing a service call according to schedule with a service call according to actual need
Early detection of sensor faults (Lidar, IMU, Camera, ...) to ensure a proactive fault clearance of the system
Increase of productivity and quality of work (increase competiveness)
Increase in workers employability
Increase in job satisfaction and reduction of work-related stress
Having insights on the accuracy of measurement in real-time is important not only for the end-user, as it can add a layer of subjectivity to the observations conclusions, but it also impacts the network management maintenance strategy.
The AI model reduces the time spent by the maintenance staff and optimize the routine of maintenance.
Estimate the weather conditions at the city of Purmerend to help operators on the decision making
Increase AI based projects at ARMAC BV
Increase the heat delivery on set point
Reduce heat loss
Letting the designers and reconfiguration planners to use their time for the actual design and planning tasks, instead of cumbersome search and filtering of feasible resources and resource combinations from various catalogues with somewhat incomparable information.
Reducing human errors in resource search and filtering (e.g. potential alternative resources overlooked by human, selecting resources with incompatible interfaces)
Increasing the amount of alternative resource solutions considered, leading potentially to more efficient production system configurations. This may mean, e.g. faster throughput time, better product quality, reduced investment cost or some other improvement, depending on the target KPIs.
Reducing the time used for system design and reconfiguration planning activity, and thus lowering the design costs.
Increased visibility on the execution of a production schedule
Increased flexibility and efficiency of a production system
Increased robustness of operations of robotic mobile manipulators
Reduction of maintenance costs and number of faults
Optimization of production quality (reduction of discards), costs (times, maintenance)
Optimization of safety and wellness of operators
Higher reliability and availability of stamping facilities by reducing downtime due to unforeseen breakdowns.
Reduced the analysis times of specialized personnel.
Reduce the costs of the Predictive Maintenance service for stamping companies, facilitating the access of SMEs to the advantages of failure prevention technologies
Improved ability to failure anticipation by being possible a much faster analysis,
Improve competitiveness of the stamping sector by reducing production costs due to unforeseen failures, as well as the Predictive Maintenance service provider
Optimize water consumption.
Generate Reports and Field Maps
Identify malfunction in the water consumption.
Prevent unintendedly water leakages.
Identify and Predict water consumption
Measure water quality
Early estimating the potential faultiness could help local manufacturing companies on planning, controlling and executing productive activities in an optimized and predictive manner.
Production planning optimization
Reduced costs
Optimized efficiency and effectiveness of the manufacturing systems
Tooling savings.
Operators savings. Less operators needed for machines management.
Increase of the production due to increase of the machine availability.
Increase of quality of pieces and tools life time.
The troubleshooting system implemented in the Experiment is an AI-driven advanced solution that improved an already existing system, totally deterministic. The new solutions instead is based on a probabilistic approach. Compared to the previous one, it is: