Awarded Proposals of Type-B/second Open Call

KITT4SME invited SMEs and midcaps from the AI-developer community with a proposition for the manufacturing industry to join the project and gain access to technical know-how and the chance to create ‘success stories’ for their solutions through the project’s pilot experiments. 53 proposals were submitted to the Type-B Open Call, and 7 of them were selected for funding with a total of €1.318.622 awarded to the winners. A brief description of each awarded proposal is summarized below including a graphical abstract or a project logo.


More than ever it is now fundamental for SMEs to invest in modular, resilient and sustainable solutions. Automatic quality control makes no difference, and with the AIMS 4QC project S.r.l. will provide workers with an easily reconfigurable solution that will empower operators to teach machines what are “good” components. The solution will provide real-time inspections fully integrated in production environment, thus enabling companies to reach the 100% quality controls for packaging production. The industrial app will be integrated within the Kitt4SME ecosystem, thus enabling and building up new solutions and making the most of the quality control data generated. The innovative system will be installed at the ROBOPLAST S.r.l. plastic packaging plant, providing a fully operative use case for real world performances evaluation.


The project involves the automation of the quality check for Bipolar Needle Electrodes through the deployment of a computer vision system (CV), supported by artificial intelligence (AI) that focuses on detecting needle tip defects. The purpose is to replace the human quality check with the CV-AI based quality system with cobot for manipulation. The object to be analyzed is positioned by a cobot in front of one or more cameras. An AI system acquires and processes the image to carry out the required checks in order to distinguish conformity or discrepancies and decide the destination of the object. Good AI training allows the system to recognize acceptable or defective pieces even if their type does not belong to the set used for training. The human expertise will be useful to train the neural network lying at the basis of AI in order to fuse artificial intelligence and human problem-solving expertise into a single digital brain with extraordinary shop floor coordination capabilities.


FASPAS addresses the challenge of scaling up the number of work orders in a discrete manufacturing environment by leveraging automatic generation of plans and schedules. The aim is to optimize manufacturing operation as well as workforce in mass production of heterogeneous custom products. When dynamic aspects such as physiological state and fatigue of employee are taken into consideration the problem becomes even more challenging. Main benefits include cost reduction, reliable commitments to customers, improved production throughput as well as higher job satisfaction level in the workplace. In order to integrate all aspects of the problem, FASPAS semantically annotates data from FaMS, an AI-based KITT4SME component, that analyzes physiological data indicating fatigue acquired by wearable devices (such as blood pressure, heart rate, galvanic skin response) together with static characteristics (skills, age, work experience and similar). The semantically annotated datasets will be used for semantics-based reasoning in order to perform optimal work order planning and scheduling taking into account all relevant factors.


Many SMEs that manufacture IoT devices in small quantities face a special challenge in terms of quality assurance. On the one hand, manufacturing small numbers of devices is too expensive to move to a fully automatic production system, but on the other hand, manual or semi-automated processes are more exposed to human error.

Our JitJip solution will enable this type of small-scale manufacturers to decrease the risk of human-made errors or defects during manufacturing and assembly by:

  • Creating and implementing quality assurance processes, including:
    • Ensuring that the correct modules are included in the device (no missing components, no extra component added)
    • Verifying that the device meets environmental and waterproofness requirements
    • Verifying the proper installation of antennas
  • Tracking defects
  • Producing regular quality assurance reports and recommendations for further improvements.


Quality of welding is essential in many industrial sectors. Today’s “AS-IS” process starts with manual parameters setting, proceeds with the actual welding activity and ends with a quality assessment and certification performed by human experts, often leading to rework in case of unacceptable anomalies.

The POWERDECK™ Action aims at exploiting the KITTS4SME functionalities for quality control, field data acquisition and analysis to build a data processing, augmentation, and knowledge transfer/classification pipeline for welding quality assessment and certification through an innovative distributed storage and ledger technology.

The project has been proposed by MYWAI S.r.l., one of Europe’s fastest growing Edge AI start-ups, together with SDG Studio, a specialized Industry 4.0 system integrator, to address the quality improvement needs of San Giorgio SEIGEN S.p.A., a world-leading provider of welding services in the power generation equipment manufacturing sector.


The QAI4GFRPs project combines the best of the digitalization platforms Edelog and KITT4SME. With Edelog, SMEs can easily model all their production processes, integrate them with IoT data, and capture all relevant production data. With the AIQuS module from KITT4SME, the data from Edelog can be analyzed and corresponding parameters for optimal production can be provided. To realize and validate the project in a real environment, the software company behind Edelog, Conclurer GmbH, and C.F. Maier Polimer-Technikai work together. C.F. Maier Polimer-Technikai specializes in the production of glass fiber-reinforced plastic parts. The production is mostly done by hand, and for optimal results, the expertise of the foremen and operators on one hand and optimal production conditions on the other are essential. The aim of the project is to use KITT4SME’s AI to improve the quality of production.


The vision of this proposal is to enable SME-affordable (inexpensive, less interaction with domain experts and easy to deploy) process quality monitoring services, which consider product quality in the context of environmental friendliness and energy consumption, assuring so called eco-aware process quality, one of the key concept for realizing sustainable manufacturing.

Main goal is to develop, deploy and validate an AI-based software-hardware system for enabling a comprehensive monitoring, analysis and improvement of the eco-aware process quality in the manufacturing end user SME, with a very clear exploitation intension to offer that kind of the products/services for Europa-wide manufacturing SMEs.

The focus will be on two main factors/waste of environmental footprint: Energy and Emission, measured in a non-intrusive way (without a direct intervention with machines), using external sensors (energy, air quality). This heterogenous data will be processed using advanced AI methods to develop comprehensive models of the process behaviour regarding eco-aware process quality.

Outcome is an innovative KITT4SME-enabled (integrating KITT4SME AI for Quality Systems solution) system for monitoring and assuring eco-aware process quality in manufacturing SMEs, integrated in KITT4SME Platform and offered through RAMP Marketplace.


pArtIcle-QC is focused on the HMI human operator, can get feedback on detected defects with picture of the detected defect. Software part of solutions are algorithms and cloud UI platform where user can see production statistics, defects, defects rate and verify defects by reviewing and labeling them. By doing so user can improve the AI algorithm itself since labeled and reviewed data is used to retrain the AI models. Important thing to mention is that this solution can work in offline mode on the edge while having Internet connection to the cloud just adds value and enables solution to get better over time.