This experiment aims to create the Digital Product Brain of lines processing aluminum sheet metal.
Motivation of the Experiment
The experiment targets the development and implementation of a Digital Brain solution for one of GIGANT’s manufacturing systems, intended for laser-cutting and forming of aluminium products. This manufacturing line (laser machines, hydraulic presses, conveyors, handling systems, and robots) is a complex, automated manufacturing system where different technologies, different vendors, and different control systems co-exist and cooperate.
The current limitations related to the actual GIGANT production line, that reflect the current status of most European manufacturing SMEs, are:
- The inability to perform real-time “what-if” analysis based on the actual plant data, that strongly impair the ability to promptly tackle production needs and constraints.
- The inability to use the digital twin in the bidding/negotiation phase, that hampers the possibility to demonstrate all plant configurations, leading to under-exploited machine customization potential.
- The inability to provide simulation services and analytics empowered by the knowledge developed all along the plant’s life-cycle reduces plant productivity and efficiency, reducing customer satisfaction.
GIGANT Manufacturing System
Purpose of the Experiment
The implemented solution aims to simplify the workflow for the design of a new plant, allowing to developing of a tool able to integrate design and simulation of a new layout. The layouts will be composed using reusable modules with a simple drag and drop interface, generating a 3D model of the new plant integrating discrete event simulation engine used to estimate productive performances. The simulation model will be in a second phase, interfaced with the real plant through CPSizer technology developed by NXT control. The tool will read configuration parameters from the real plant and will be used to optimize production scheduling with an increased level of credibility.
At the end of the experiment, the end-user will have a functioning tool to design new production layouts for the fabrication of goods starting from aluminium sheets.
The product will allow exporting the simulation model of the designed plant and deploying it into the DIGITbrain platform to be connected with the real plant through FIWARE to read parameters from an existing facility in Verona.
Simulation will be executed on the cloud platform to optimize the production scheduling of the existing plant and increase production performances.
Currently, the process of designing a new layout requires approximately 14 days and the cooperation of different professional figures from different vendors.
The tool developed in the experiment will reduce the time to design a new layout to 1 day, and the process will require only a single person. Moreover, the productivity of the plant will be estimated with a greater level of accuracy, also thanks to the refinements of the simulation models derived from data acquired on the field from past deployed plants.
When the plant is installed in the facility, the same simulation model will be interfaced to the real plant and will be used to improve production planning
Expected Economic Impact
The usage of the new tool will reduce the time and costs of the bidding phase with potential new customers and increase the success rate of the negotiations by 10%. Moreover, the technical reference of the CPS data model in the digital realm will be a key enabler for data standardization and continuity all along the plant’s life-cycle, with a quantified impact of 50% faster ramp-up, due to effective virtual commissioning.
Together with the possibility to implement decision rules and control logic in a simulated environment that mimics faithfully the real world, reliable modelling will improve resource optimization and efficiency, reducing the time to acquire such results, with the quantified impact of “time to production” down by 20% (including plant modelling, iteration with the customer, ramp-up, and production first trials).