Work packages

WP1 “Project management” – Leader PA

Objectives:

• Ensuring the effective management of the project within time and budget constraints.
• Coordinating project partners to achieve synergy in collaboration.
• Active communication with the EU.

The Project management activities coordinated by PA will require the support and synergy of the WISE consortium in
order to assess and consolidate the project progress from the scientific, technological and financial perspectives.

Description:

This work package covers all aspects of project management and control that will ensure successful achievement of the objectives within time and budget. Contractual, legal and administrative issues are also handled within this work package.

Task 1.1 – Progress management and DMP (PA)
In this task, the administrative progress of the project and of the individual work packages will be assessed. The task coordinator will be responsible to manage the consortium for all legal, contractual, ethical and administrative matters, coordinating the knowledge management too. The task coordinator will also make sure that progress and annual reports will be completed and delivered on time.

Task 1.2 – Financial administration (PA)
The project manager will be responsible for managing the financial issues of the project with a financial management report sent by each member at the end of each project year to track partner and project expenses.

Task 1.3 – Gender equity and workforce competence, inclusiveness, and upskill (IRIS)
IRIS will be responsible to manage data and activities carried out by the project consortium, ensuring a strong commitment to comply with the gender fluidity and inclusiveness in the targeted manufacturing value chains in addition to special policies to up/re-skill the industrial workforce. A gender mentor will be appointed by SC (as part of the STC) to review deliverables addressing the societal impact.

WP2 “Product design by hybrid-technologies” – Leader IRIS

Objectives:

• Formulating new design principles, rules and prototypes driven by the adopted technologies.
• Integrating in MedTech, Aerospace, and Power generation components new smart functionalities
• Mapping the product requirements to the novel technologies capabilities.
• Definition of design-to-lifevalue policies for products and round robin part families.

Description:

The process of designing and implementing the WISE solution starts with the investigation of product information, covering design, manufacturing and end of life phases. Specifications such as geometry, components, working environments and materials will be analysed in order to redesign them with respect to the available technologies. The activities of WP2 are structured in the following tasks:

Task 2.1 – Product family requirements definition (SUPSI)
T2.1 collects the information about the end-user requirements and synthesizes the production scenarios that will be considered as references for the design of the parts and will provide baseline data for the Knowledge Base (T3.1) and quality specs for the AI-powered functionalities (WP3). Particularly, the design process will translate the user requirements in product features of the final part, thus involving all players in the design process (from the user to all manufacturers). Examples products are:
MedTech: spinal cage prosthesis functionalized by biocompatible nanocoating surfaces enabling the surface diffusion of organic molecules in preferential directions and areas by means of low-power health-safe radiation; dimensions up to 25mm x 25mm x 50mm; coating resolution:<10-8 mm; surface finishing: 0.1 – 1 micron; production volume (worldwide, n. pcs/year):> 2’500; min expected cost saving: 29%. Core component material: Ti-6Al-4V; Coating material: biocompatible oxides (e.g., Ca10(PO4)6(OH)2, Ca3(PO4)2) and biocompatible polymer (e.g., PEEK).
Aerospace: smart CCMs turbine blade implementing self-healing mechanisms exploiting severe working environments. Dimensions up to 200mm x 200mm x 200mm; dimensional accuracy: < 0.05mm; production volume (worldwide, n. pcs/year):> 300; min expected cost saving: 35%. Composite material: Ni-based core + CCAs functionally graded layer CCO coating.
Power generation: smart sensors enabling autonomous repairing mechanisms of hydroelectricity pipelines when structural damages occur, assessing defects’ behaviour over time to plan maintenance actions. Dimensions: 100mm – 1’000mm; coating resolution: < 10-8 mm; surface finishing: 0.1 – 1 micron; production volume (worldwide, n. pcs/ year):> 400; min expected cost saving: 30%. Product with complex structures: a core of metallic alloy (e.g., Stainless Steel, Ni-based alloy) + external polymeric coating + ultra-thin ceramic.

Task 2.2 – Product family design by technologies and design for macro-to-nano manufacturing (IRIS)
The user-based requirements and scenarios analysed in T2.1 will lead to the definition of a radical new concept of products where design specifications will be fused with technological information. Particularly, the description of the family of parts will drive the identification of materials and components to be managed through DED, femto laser (ablation, 2PP), nano-laser, and DALP. The task will include an extensive development of the products in accordance with the physical and chemical properties of the different components that will be assembled, e.g.:

• Definition of the best DED/femto/nano setup enabling high-precision and multi-process manufacturing (WP4).
• Definition of the best set of system enclosure materials printable through WISE technologies.
• Definition of the working requirement for ultra-thin layers depositions through DALP (WP4, WP5).
• Definition of the best setup for part high-precision positioning control for DALP and femto-based technologies.
• Definition of the best macro-to-nano-building sequence for the entire product family (WP7, WP8).
  

Task 2.3 – Round Robin Part Family Definition (SUPSI)
A comprehensive Round Robin Family of products will be defined, with specific geometric, material and functional properties, capturing the aspects associated to product complexity and manufacturability challenges. This will include a comprehensive analysis electrochemical, thermal and mechanical properties of parts in order to identify the critical areas for the process execution. The geometric features will be associated to multiple manufacturing strategies, coherently with the available process technologies incorporated in the WISE machine, and become the backbone for any product design and the development of the AI-Aided Product Engineering Platform (WP3). A set of user-based and process-based KPIs will be determined (quality, performance vs dimensions, throughput, cost). as well as a number of formal benchmarks.

WP3 “AI-Aided Engineering Platform” – Leader SUPSI

Objectives:

• Establishing a centralized data and knowledge bank, the Knowledge Base.
• Design and test multi-scale models, based on multi-physics and data-driven modelling, for DED, femto-based (ablation and 2PP), nano-based, and DALP processes associated to Round Robin Parts.
• Ensure the efficient organization and dispatching of parallel batches and technologies.
• Train and test ML support models, like strategy recommenders and surrogate models.

Description:

WP3 will address the realization of a software platform that allows to chain the optimized sequence of manufacturing processes, by selecting the best set of technologies and defining the process parameters and precedence constraints. A key feature of the Platform is the enrichment of the design strategy with part lifetime simulation models. The activities of WP3 are split basing on their reference module in the AEP:

Task 3.1 – Knowledge Base design and implementation (MOR)
The Knowledge Base (KB) will be established by setting up its two main components:

• The relational database that manages the whole design, process and quality control measurements. The database will
  be optimized also for efficient extraction of big data sets to be used by the ML sub-modules. A first version will be
  issued to test the architecture and completeness of described data, in synthetic design tasks, while an advanced version
  will be thoroughly tested to ensure efficiency and scalability towards high data volumes.
• The process and product ontologies. These will be defined starting from templates for manufacturing like STEP-NC,
  and enriched bottom-up with specific information, collected in WP2, to ensure an exhaustive but parsimonious set of
  entities and relations. Moreover, a specific ontology will be dedicated to human-machine interaction. The process will
  be sped-up by support tools like Protegé, and the ontology set will be released in OWL and RDF standards. A Prolog
  interface, accessible from the other AI modules, will provide access to ontology contents.
  

The KB will be implemented on a dedicated workstation, optimized for data flow, with a double data storage system: a local version (2-5 TB), for fast saving of recent jobs, and a remote backup, slower but with much larger, expandible storage capacity.

Task 3.2 – AI-Aided Lifevalue Optimization (SUPSI)
This task will take care of setting up the library of multi-physics and data-driven models. In the first phase, simulation tools will be assessed to find the best available options (e.g. COMSOL, Ansys, Flow3D) in terms of accuracy and computation time. Secondly, model setup pipelines will be established, to easily pass new parameters and data tables and launch simulation runs. Finally, each model will be fully tested on Round Robins. In parallel, a process characterization campaign will be set up to train the data-driven models and instantiate the parameters of M-P models. For the datadriven models, specific architectures will be chosen basing on the type of process and outcome of interest. Uncertainty,
and thus reliability, for each of the models will be characterized by sample distributions of simulation/prediction runs, and validated against Round Robin parts. The Virtual Self-Awareness will firstly be instantiated with nominal parameters, while a data integration interface from
the KB will convert the prior to the posterior Virtual Self-Awareness, instantiated with measured parameters. The validation of the prior version will be performed in a Monte-Carlo setting, while the posterior version will be validated by building likely lifecycle scenarios based on the Round Robins. All the aforementioned models will be set up and tested on a high-end workstation, equipped with a GPU card for fast running of M-P and ML models. ML models training will be performed on a cluster solution.

Task 3.3 – AI-Aided Manufacturing (MCH)
Task 3.3 involves the development of the following main sub-modules:

• Product configurator: an existing CAD-CAM platform will be equipped with an additional interface to support the user in realizing the special functionalities by applying the correct process chain. An initial set of product templates will be designed basing on Round Robins, to set up the data-driven model for template generation; different models will be compared (e.g. KNN, Decision Trees) with a focus on heterogeneous data and learning from few-examples.
• Multi-process planner: a task scheduler will be implemented, integrating ontology rules coded in the KB, to ensure correct technology sequence and product quality. Different options for optimization will be tested: time-optimal, energyoptimal, quality-optimal.
• Strategy recommender: this parameter-to-lifevalue ML model (possibly a robust regressor for heterogenous data, like XGBoost) will be trained by generating thousands of part programs, with a wide combination of process and strategy parameters, covering typical usage in the use cases. This run of simulations will form the basis of the recommender, while a second phase will implement continuous integration of user feedbacks by Reinforcement Learning. Easy and comprehensive reporting of AI decision paths will be also implemented.
  These features will be developed on top of a SoA CAM platform, that will be set up to generate post-processed part programs for all the involved technologies.

WP4 “Design of WISE modules” – Leader PA

Objectives:

• Design and engineering of an innovative DED system capable to perform high-build rates in conjunction with highprecision depositions, as well as high-quality graded structures.
• Development of a high-efficiency, high-beam-quality laser femto system for ablation, surface texturing and 2PP, designed to deliver photons packets exactly where required. The same scanning solution will be shared by the nanosecond laser.
• Design and engineering a novel DALP head, capable of depositing thin and precise reactants layers faster than current state-of-the-art solutions.

Description:

WP4 will address the selection, design and optimization of laser sources, the design and optimization of the combiner, optical chain and FAST unit, the design and optimization of the station 1 head for laser-based processes and of the station 2 for DALP, the engineering and realization of the WISE modules. T4.1-4.4 are devoted to the technologies’ development, whereas T4.5 is dedicated to the integration of the developed
solutions into the WISE machine. All the tasks include tests to qualify their outcomes.

Task 4.1 – Selection, design and optimization of laser sources (PA)
WISE photonics setup relies on four complementary laser sources (i.e., IR CW laser, Blue CW laser, femto-laser, and nano-laser). For the DED process a hybrid IR/blue laser solution will be used, obtained with the multiplexing of two different sources. The first one is a commercially available 6kW IR fiber laser optimized to ensure high efficiency and beam quality, with M2<1.1. The second one is a 600W CW blue DL, conceived on three 200W multi-emitter modules based on the latest generation of GaN single-emitters, each capable of emitting approximately 10W. A beam expander will be designed and integrated into the blue laser source to allow a dynamic variation of the laser spot, while the IR beam
diameter will be adjusted in the combiner (T4.2). High-quality and high-precision femto-based (e.g., ablation, 2PP) and nano-based (e.g., polishing, surface treatments) processes will be ensured by the implementation into the WISE solution of a high-power IR 300W 300 fs laser source and a 500W 100 ns pulsed source. The IR CW beam together with the fs and ns sources beams will be directed towards the combiner, to couple all the beam in the optical chain to deliver the radiation to the station 1 head.

Task 4.2 – Design and optimization of the combiner, optical chain and FAST unit (ALITE)
This task deals with the design, development, and optimization of the optical elements necessary for the delivery, shaping and scanning of the laser beams for proper laser processing of the part. The selection of the optical elements and their spatial configuration in the combiner will be optimized in order to couple all the IR beam laser (CW, fs, and ns) into the same optical chain without power losses and beams degradation. As first concept, it will be constituted of 2 QBH connector, for the fiber emitted lasers, and an input aperture for the femtosecond laser, which has a beam emitted in free-space. A motorized beam expander/reducer will consent to adapt the beam dimension for the suitable resolution for the DED process. Special AR/R coating mirrors consent to direct all the beam to the output aperture. Two different sets of mirrors (with AR coating @1030-1070 nm and @450 nm respectively) will deliver the beam to the laser head on parallel optical paths. A final motorized mirror can switch its position in order to direct the beam towards the FAST Unit (fs and ns beams) or to reflect it on a dichroic mirror to obtain the CW IR and blue beams combination. As done with the combiner’s optics, the optical chain’s elements will be selected after proper simulations with optic modelling programs with consequent validation tests in order to choose lens/mirrors that guarantee no-power losses and avoid beam quality reduction, selecting elements with suitable parameters (focal, R/AR coatings, material damage threshold). For each element, a water-cooled mount will be evaluated to prevent damages due to overheating. The motorized elements must be reliable, accurate, and repeatable. The Fast and Advanced Scanning and Tailoring (FAST) unit will be designed to achieve the target scanning speeds (20 m/s) and beam quality. The resulting configuration will be tested on all the relevant WISE technologies (custom beam shaping for fs and ns processes), targeting their specific KPIs. In addition, a calibration and control apparatus will be engineered and integrated in the FAST unit by exploiting the light reflected from the mirrors in the “off” laser state, which will be acquired by a beam profiling camera in order to control both the shape and quality of the outgoing beam and enable to optimize the spot size, shape, and focus depth for the operating conditions. The use of ultra-short pulses in a high repetition rate regime – as in high-speed patterning – will require the optimization of the trade-off between speed and accuracy.

Task 4.3 – Design and optimization of the station 1 head for laser-based processes (PA)
This task deals with the design of the module equipment necessary for the laser multi-processes. The FAST module, designed in T4.2, will be positioned in an optimized parallel configuration with the DED head, integrated after the hybrid blue/IR source dichroic multiplexing mirror. Moreover, a photopolymer dispenser will be selected and positioned on the station 1 head, in order to have an accurate deposition of resin on the surfaces that required polymeric coatings. The DED head is based on the SUPSI DED patent. The starting multiple-nozzle design head will be optimized to avoid head overheating and clogging of the nozzles, thus increasing process performance for long printing time. For this reason,
a chromium plating will be validated for the nozzles and the main body of the deposition head will be manufactured in high-conductive metal alloys (e.g., Cu-based alloy). The DED head will be connected with the multi-hopper powder feeding system, enabling the manufacturing of functionally graded structures and composite materials, and will exploit the flexibility of the designed hybrid IR/blue laser source configuration to manufacture both high absorption and HRM metal components. The employment of a dynamic laser spot will be validated to ensure high deposition rate (>500 cm³/h) with a 5 mm laser spot in conjunction with precise micro-scale depositions for chemical and structural surface functionalization when a 0.2 mm laser spot is involved.

Task 4.4 – Design and optimization of the station 2 for DALP (ATLANT)
The current task addresses the DALP module design and engineering, in order to achieve the best mechatronic solution in terms of robustness and reliability. The main phases of the task are:

• Design of the printer head solution (mechanical design, actuation and materials), in order to fulfil the demanding constraints of nozzle distance (roughly 100um) and planarity (roughly 5um) to the target surface.
• Engineering of the nozzles and piping systems, to increase the deposition resolution as well as its reliability and robustness. To achieve that, a heating system capable to manage the temperature of the gases in different positions will be realized.
• Integration of different sensors aiming to measure and control different key variables of the process, such as the nozzle position, temperature (100 to 300°C) and gas flows as well as the related hardware enabling their control.

Task 4.5 – Engineering and realization of the WISE modules (MCH)
This task is dedicated to the integration of the modules developed in T4.1, T4.2, T4.3 and T4.4. The modules will be tested with respect to their assembly effectiveness as well as their process capabilities in terms of parameters recipes defined by the AEP. To do that, a specific test bench for each module and related auxiliaries will be developed. This activity will not only allow the machine designer to perform an extensive test campaign simulating actual working conditions, but also provide a platform for data collection over processes and realized part features, that will be used to train the AEP.

WP5 “Design of machine architecture and efficient mechatronics” – Leader MCH

Objectives:

WP5 regards the design of the machine architecture, including the super-stiff basement. WISE is conceived to operate in highly dynamic production contexts, where the products families can change in volume and technological requirements, therefore, this activity is structured to generate an evolving mechatronic solution based on the modifiable use of modules, according to lifecycle degradation status and process needs.

Description:

Based on the reference family of parts identified in WP2 and the processes and modules design addressed in WP3-WP4, WP5 will manage the entire mechatronic configuration process; this will be supported by RTD partners with regards to the methodologies development, while industrial partners will provide functional specifications and verify the industrial viability. The WP is organized in the following tasks:

Task 5.1 – Design of WISE super-stiff structure (PA)
The WISE basement structure conception will embrace both mechatronic, process and operation-related aspects to be concurrently considered during the configuration. The stiff basement structure will be designed to host precise (<1um) processes, also thanks to the implementation of temperature stability control. Moreover, the housing of the modules will be realized in a compact and lightweight head solution, integrating active or passive damping systems. The design process will be driven by working and reachability space, accuracy and reliability, energy efficiency and throughput over time.

Task 5.2 – Customization and engineering of auxiliary systems (SUPSI)
This task deals with the customization and engineering of auxiliary systems needed to perform high-efficient, highprecision, and high-quality multi-process manufacturing, such as: power supply, powder feeding system (DED), precursor gases supply (DALP), inert gas supply (fs and ns), suction and filtering unit, and cooling systems devices (for optics). The high-precision manufacturing of the laser-based hybrid processes from macro-to-nano scale and vice versa will be ensured not only by the super-stiff structure designed and built in T5.1 but also by the proper engineering, functioning, and positioning of the auxiliaries in the WISE structure. Electric power fluctuations, residual magnetic fields, overheating of the optical chain, pulsating gas flows, as well as a discontinuous precursor, would lower the production efficiency of the multi-process solution and affect the final quality of functionalised parts. Moreover, SUPSI’s powder supply system will be upgraded in order to:

• handle powders of different grain sizes, both as metal alloys and as simple chemical elements;
• ensure continuous and dense powder flow at the laser spot;
• ensure proper chemical powder feeding compositions and continuous material mixing during FGM fabrication (through
  an improved control logic on powder metering and gas flow for CCAs and CCOs).

Task 5.3 – Mechatronic machine integration and model-based representation (MCH)
This task will deal with the mechatronic integration of basement, FAST unit, DED/2PP head, DALP head and auxiliary systems. The integration activities will be followed by a number of commissioning tests, including: the verification of mobility and capability to cover reachability space; the ability to perform full acceleration and deceleration at maximum speed, the capability to deposit material in accordance to all the available technologies, the heat and fluxes management. Concurrently to the mechatronic integration, a comprehensive machine model that takes into account all inputs from WP3 and WP4 will be realized. This task will be followed by full integration tasks where also the control and the
monitoring systems will be validated (WP9).

WP6 “All-around nano-scale vision and monitoring infrastructure” – Leader MCH

Objectives:

• Develop smart sensing system architecture capable of monitoring, at a nano-scale level, the entire part processing and, at a macro-scale level, the machine.
• Develop an advanced vision system composed of high speed cameras and interferometers operating in the nanoscale range measurement system.
• Design a SW and HW infrastructure for data fusion and elaboration.

Description:

The sensing system in WISE solution will enable the following monitoring activities:

• 2PP and femto-based ablation processes quality, by monitoring the laser beam.
• Melt pool size/shape analysis, gas and powder flow, temperature imaging of the part for DED processing.
• Surface quality evaluation, in support to DALP and 2PP processes.
• Process chain efficiency, integrating data about process quality, generated heat and gases, power consumption.

The task activities are divided as follow:

Task 6.1 – Design of all-around monitoring system HW architecture (PA)
The architecture of the sensing system will be selected and designed by considering the machine functionalities, capabilities and the possible configurations, complementing the sensors already nested in the modules. Depending on their position within the machine, sensors might be demanded to target properties such as lightweight and compactness, robustness to chemical, thermal and computational thresholds. The positioning of critical defect identification sensors, like ellipsometer or optical profilometer will be performed in strategic positions and protected from the environment, in dedicated areas. The real time monitoring of key process parameters will be implemented as well: a high-speed camera will monitor the laser beam and, together with a dedicated pyrometer, support the management of the emitted power. Moreover, a beam profiler will be installed to calibrate and monitor the FAST unit.

Task 6.2 – Design of all-around monitoring system SW architecture (MCH)
To effectively perform process control in the sensing system, several tasks like vision and topology measurement, monitoring, diagnosis, generation of control actions and learning will be addressed by the vision system. This requires identifying a set of features within the sensor capturing data that are indicative of the degradation processes and utilizing them with an appropriate condition monitoring method to deduce the equipment health. Statistical process control (SPC) approaches will be implemented and applied for achieving robust performance as well as for implementation and training of the involved data-driven models (e.g. LSTM for time series data). The vision system will locally evaluate the quality of the laser beam and the material ejection directly in the optical chain as well as the in-progress part realization and assembly accuracy by focusing on the manipulation station. This will be realized by performing high-speed image processing as well as position analysis, implemented on a dedicated GPU solution. The monitoring system will be deployed on a dedicated workstation, connected to the KB (T3.1) for data logging.

Task 6.3 – High speed data fusion from smart sensors (MCH)
This task will develop the algorithms for fast data collection, alignment and pre-processing. Since the evaluation of a measured quality values requires the analysis of many different aspects, complex algorithms are required to extract the needed information. They can be implemented directly in hardware, when the smart sensors include an FPGA device, or by shifting the processing on the dedicated workstation, in a speed/flexibility/memory trade-off. Another important topic addressed by this task is the data reduction. The acquisition of process information will be performed periodically, and the sensors will collect and generate a large amount of data at every sampling time. For economic and technological reasons, the extraction of the meaningful part of data from the complete acquired set, at each sampling period, will be performed as much as possible directly at the sensors level, in order to minimize the amount of transmitted data.

WP7 “Adaptive automation and CNC” – Leader MCH

Objectives:

• Identification and modelling of the dynamics of the production environment to be controlled.
• Functional specification and development of high speed automation architecture and its adaptation logics.
• Design of the CNC architecture and adaptive control algorithms based on the process models.
• Development of an open middleware infrastructure.

Description:

The design and development of the WISE control architecture is organized in tasks for logic control, CNC and middleware system. In addition, the range of technologies available in WISE requires the development of special adaptive control algorithms and a control architecture to be designed as modular, distributed and reconfigurable.

Task 7.1 – Automation Architecture Design (MCH)
The automation development process considers the machine configuration solution and functional specifications developed in WP5, as also the characteristics of the WISE solution, with particular regard to the various processes integrated in the machine (WP3 and WP4) and the intelligence distributed in the various sensing modules (WP6). The activities in the current task consist of three major phases. The first one is the Preliminary Design and Verification, which involves the control system specification, the control system HW and SW architecture design, the control system functional design and the high-level logic verification. The second phase is the Control Development and Implementation, which consists of the design of algorithms and the control code implementation. The last phase is the Control System Verification and Validation, which involves the control code verification and validation and the delivering of the executable control code.

Task 7.2 – CNC architecture design (SUPSI)
The design of the WISE CNC architecture will determine the capability of the machine to realize the assigned operations with the desired accuracy and productivity. The fundamental modules will be:

• Data Set Builder: alignment and recording of process data in order to create a data.
• Trajectory generator: workspace trajectory generation for each specific “end-effectors” trajectory.
• Kinematics Transformation module: converting Cartesian position information to mechanical axes/scanner positions, solving also kinematic redundancies.
• Process control: synchronization of motion actions with peripheral elements (laser, scanner, precursor gases).

Task 7.3 – Closed-loop CAx chain (MOR)
This task will deal with the adaptation of existing part program regeneration routines, already implemented by SUPSI for DED and fs ablation, to the additional technologies, and their orchestration in a multi-technology context. A communication interface from the check-point measurement sensors (those dedicate to 3D scan and surface quality) to the part program regeneration module will be defined. Starting from the consolidated part program generation engine for DED deposition + pulsed ablation, the part program generation and post processing for each technology will be programmed and the reasoning system will be extended, taking into account specific process parameters (e.g., gases
flow, temperature for DALP; laser power and frequency for femto-ablation and 2PP). The reasoner will be fine-tuned taking into account process and product tolerances and scales, to avoid “scan-correct” infinite loops or repairing of negligible deviations. Moreover, sensor-machine re-calibration routines will be established, to periodically check the spatial coherence between measurements and corrective tool paths.

Task 7.4 – Adaptive CNC algorithms (SUPSI)
Geometrical and technological adaptiveness in WISE will be realized, at local scale, by compensating for machine anomalous behaviour or a change in the optimization strategy, which might vary the current operating policies. Adaptive algorithms will be developed to manage the trajectory generation, the motion parameters and the process-related parameters, considering the possible sources of variability. In order to do that, the “Control Modeller” will process (offline) data gathered form the Data Set Builder along with the electro-mechanical, geometrical and quality performance of the manufactured parts (optimal requirements) to generate technology-specific, data-driven process models by machine learning techniques. This off-line elaboration is completed with an on-line suite called “Control Online Interpreter” where the data collected form the various sensors are identified and interpret in order to generate corrective actions based on different KPIs.

Task 7.5 – Middleware and communication infrastructure (MCH)
This task will be responsible for selecting and implementing the communication protocol and approach to guarantee interoperability in a heterogeneous environment. Specifically, the objective is to realize a platform-independent architecture through which various kinds of systems and devices will be able to communicate by exchanging messages of different size and with different constraints (e.g., real-time response and quality of service). Beyond the physical and syntactical interoperability, particular focus will be dedicated to the semantic interoperability. For this reason, serviceoriented architectures (SOA), and particularly the ones applicable in the industrial domain like the new OPC-UA will be considered and investigated. Such architectures require the definition of services and the implementation of a robust mechanism through which the individual participants specify what service sets they support. The detailed needs for reliability, scalability, performance, security, interoperability, services and extensibility will be also strongly considered and will drive the final choice for middleware architecture.

WP8 “WISE integration, monitoring and optimization” – Leader PA

Objectives:

• Integrating the WISE machine in all its mechatronic modules and SW infrastructure.
• Efficiently managing the machine set-up and ramp-up.
• Defining of multi-level optimization and adaptation strategies.
• In line evaluation of the effective material property.

Description:

The integration of the WISE solution starts from the results obtained from WP3 to WP8 enables the realization of the lab scale and industrial prototypes. The following tasks will address the steps characterizing the island integration and commissioning.

Task 8.1 – WISE machine integration (MCH)
This task refers to the SW and HW integration in the WISE solution. The comprehensive integration will involve the integration of the mechatronic solution developed in WP4 and WP5 with the CNC, automation and middleware systems and the sensing system developed in WP6 and WP7. The physical mechatronic integration as well as the communication and SW integration will be tested under a number of testing cycles. As a second integration stage, the machine will execute standard cycles verifying all its functionalities. In the end, the machine will manufacture the round robin family of parts and the envisaged thresholds for the KPIs will be tracked. These phases will conclude the beta level of integration which will be followed by analysis related to the machine and software robustness to unforeseen events (by fault injection) and adaptation capabilities (by generating fake anomalies).

Task 8.2 -Multi-level Optimization (SUPSI)
This task will implement a context recognition feature, designed to automatically reproduce an abstraction of the production context and compare it with the nominal one. This entails to continuously operate on a contextual model consisting of inference patterns associated to different values read from the sensors (e.g. device vibrations, temperature, energy consumption) as well as timelines related to the different variables of interest (e.g. precursors consumption over time or equipment degradation). Based on the recognized context, WISE will evaluate and optimize the appropriate capabilities, functionalities and settings. The optimization goals will trade-off on multiple objectives, such as precision, efficiency, productivity and energy efficiency. As a result, the optimization component will embed a reasoner, leveraging the optimization strategies and defining the most suitable optimization pattern to be implemented in a specific time horizon and to be adapted over time coherently with the production needs. WISE capabilities will be improved in time through the integration of new rules and policies also tracking the module lifecycle as well as their adaptation to new unforeseen events. This will require the incorporation of learning mechanisms contributing to improve its decision making process.

WP9 “WISE demonstration” – Leader PA

Objectives:

The focus of WP9 is to demonstrate the industrial application of WISE solutions and validate the scientific andtechnological benefits with regards to the identified targeted industrial applications, certifying:

• The feasibility of the WISE, femto-based (ablation, surface texturing, 2PP), nano-based, DALP, DED process strategies.
• Demonstrate the functionalities and operativity of the FAST unit.
• The WISE capability to produce a large product variety accomplishing productivity and high quality targets.

Description:

The demonstration process will be structured as a progress of three major phases: the virtual prototype, the lab scale prototype and the full industrial solution.

Task 9.1 – Virtual demo (SUPSI)
The conceptual models developed in WP3-8 will be translated in the current task in a virtual prototype with the objective to evaluate functionalities, performance and benefits determined in cooperation with end-users. The virtual demonstrator of WISE will comprehend all the machine design aspects. The solution will be analysed, tested and validated against a number of indicators:

• Functional indicators under nominal behaviour such as reliability, efficiency, maintainability and productivity.
• Dynamic indicators under non-nominal behaviour embracing the robustness towards unforeseen events.
• Efficiency indicators both in nominal and non-nominal conditions, examined under the energy efficiency perspective, the economic as well as the business sustainability (opportunities for new productions).
• Lifecycle indicators in nominal and non-nominal conditions.

The WISE conception will be also referred to existing market solutions about the identified indicators. The virtual demo will be developed and validated primarily with project technology providers, then with end-users in T10.3.

Task 9.2 – Lab scale pilot demo (MCH)
The virtual demonstrator will be physically realized and tested in the preliminary form of lab scale pilot. The size and performance of the machine will be targeted to the realization of a subset of micro-assembled products and within a set of specific components. The machine will be physically connected to all the auxiliary systems and the IT infrastructure. Successively, it will undergo the complete set-up phase by evaluating its functional behaviour across all modules as well as its control, monitoring and communication capabilities. Then the machine will face its ramp-up by activating the production of small number of product examples realized with standard testing cycles. The ramp-up will concern the functional, process, production, reconfigurability and reliability aspects outlined in T9.1. The final step will start the complete production process with the machine manufacturing the entire family of parts identified as round robin test parts. During this phase, the entire set of experimental evaluations realized in T9.1 will be considered in order to assess the benefits and performance of the physical solution and indicators will be assessed with reference to the machine productivity and efficiency.

Task 9.3 – Final products-materials properties assessment (IRIS)
A complete characterization of the final selected materials will be carried out to validate the whole process, also by comparison with the standard technologies, and their compliance with the European Green Deal objectives. Sets of samples produced both by WISE machine and standard manufacturing technology will be submitted to electromechanical, physical and chemical characterization. The behaviour of the materials and device obtained with the two different technologies will be compared to demonstrate the real WISE machine capability. In instances where conventional manufacturing techniques are insufficient or unavailable, the examination will be based on equivalent
components that exhibit the most similar functionality. A partial material characterization will be carried out also on the final components in order to demonstrate the real homogeneity of the technology also with respect to complex shapes and increased component dimensions.

Task 9.4 – Full scale industrial pilot (PA)
The full-size machine, established at PA, will be capable of processing the whole set of products associated to all the identified industrial sectors, thus requiring the entire set of technologies and materials. Considering the scope and dimension of the machine, PA will allocate a special area of their shop floor to the machine components building, their integration and testing. Similarly to the lab scale demo, the machine will undergo preliminary rounds of testing production by performing standard production cycles both in nominal and fault injected conditions and only after it will be tested against the round robin parts. Then the machine will start manufacturing small batches of components whose
composition functionalities will go beyond the predetermined ones. As a result of this steady production phase, the machine will reveal the accuracy and soundness of the entire set of performance indicators as well as its benefits compared to the identified industrial benchmarks. This particular evaluation phase will be deeply supported by the end users, which will provide actual information about the cost and quality of the products today produced with different production chains. The successful accomplishment of the objectives will start the activities associated to the translation of the prototype into an industrial solution in the PA machine catalogue.

WP10 “WISE dissemination and exploitation” – Leader IRIS

Objectives:

• Ensuring proper scientific dissemination through open science policies.
• Developing plans for exploitation in the EU of laser based micro-assembly and manufacturing applications.
• Industrial dissemination and establishment of business interest groups as possible channels for the implementation of exploitation plans.
• Facilitating the activities for standardizing machine and process according to optoelectronics sector standards.

Description:

A strategic plan for exploitation of the results of the project in the manufacturing industry within 3 to 5 years after the conclusion of the project will be coordinated by IRIS in cooperation with related EU Technology Platforms in the NMP, FoF and People initiatives such as AM Platform, EUROP- (Robotics), Medtech, AGE, Photonics21, European Automotive Cluster Network (EACN). Project results will be disseminated among researchers via technical committees/working groups, conferences, articles in journals and in special issue academic journals and, among industry via seminars, workshops trade exhibitions and demonstrators.

Task 10.1 – Scientific dissemination (FORTH)
Dissemination to the scientific community is about wider promotion and assessment of the developed technologies, techniques, and methodologies. Dissemination activities consist of:
1) Clustering with related European and IMS-International projects, discussions will be promoted on the application of the WISE deliverables (e.g., methods, tools, and automation technologies) to other areas (e.g., standardization and best practices).
2) Creating an Interest Group in the field of smart-component manufacturing, involving International Parties.
3) Publications and presentations at major international events. Moreover, a workshop or special sessions within international conferences, involving senior consortium partners as well as delegates from other related projects (especially in FoF and Health), are planned across the whole project.
4) Definition of a training programme, for different education level in order to transfer the necessary knowledge to future workforce employed either to industrially exploit the system of to prosecute R&D activities.

Task 10.2 – Communication and Industrial promotion (PA)
Promotion within the industrial sector will include the organization of a number of events specifically tailored to the needs of different audiences across the spectrum of additive and subtractive processes as well as atomic layer deposition, two-photon polymerization and texturing and functionalization of surfaces. These events will include handson technology demonstrations, to showcase and convince of the strategic advantages obtained from the embedding of smart functionalities into components. The establishment of Business Interest Groups will be facilitated and supported as instruments for the implementation of the project exploitation plans.

Task 10.3 – Standardization and best practice procedure (IRIS)
The WISE machine is meant to become the industrial best in laser-based multi-process manufacturing of smart products and functionalised complex structures. This will require the definition of a standardization strategy that will also take into account energy management practices and design-for-recycling rules where WISE consortium – represented by IRIS will follow the procedure addressing strict requirements to ensure the machine can operate by realizing industrial, medtech-aerospace-power generation, environmental and automotive labelled products.

Task 10.4 – Exploitation Program Definition (IRIS)
The goal of this task is to develop the exploitation strategy of the results produced by the WISE project. IRIS will additionally engage industrial stakeholders interested to replicate WISE results in their applications.. A preliminary exploitation plan will be developed, mainly by industrial partners during the first phase of the project, while the definitive exploitation plan will be consolidated in the second half of the project. This final plan will be based on the best scenarios for end-user adoption of the technology in European countries within five years after the completion of the project, and will incorporate a structured business plan, which will state the strategy to support the follow-up, industrialization, commercialization and usage of the WISE solution. Moreover IRIS will coordinate ER owners to implement the strategy
for the management of IPR as described in sec. 2.2.4