EXPERIMENT OVERVIEW
ELECTRA addresses a common situation in small and medium-sized enterprises (SMEs) in food manufacturing: heavy reliance on manual supervision and legacy systems with limited real-time data access. Key tasks such as counting packaged products, registering production batches, checking stock levels, and monitoring machine status are typically performed manually, leading to fragmented information, delayed reporting, and higher operational costs. Inventory checks often require physical verification in storage (e.g., repeatedly opening fridges to confirm availability), creating energy management issues, and increasing the risk of inefficient stock handling and even expired stock.
Within WASABI, ELECTRA will develop, deploy, and demonstrate the use of a task-oriented Digital Intelligent Assistant (DIA) that combines real-time Closed-Circuit Television (CCTV) video streams from packaging machinery with data from energy meters installed on key machinery operating within the food production facility. The aim is to generate actionable insights through an Artificial Intelligence (AI) conversational interface, automatically monitoring machinery activity, counting and logging packaged items, detecting visual inconsistencies that may indicate defects or packaging errors. The DIA correlates this with batch data and energy‑consumption patterns, providing important information to workers via a natural‑language conversational interface (e.g., questions on current stock, energy use, or products nearing expiry). In this way, the experiment directly supports inventory tracking, packaging-line supervision, and quality assurance, enhancing efficiency, traceability, and sustainability across HELIOS’s manufacturing processes.
How the solution works
ELECTRA adopts a retrofit, low-barrier approach that builds directly on data sources that are often already available in production facilities (CCTV cameras and low-cost energy meters). Instead of introducing new proprietary hardware, ELECTRA leverages these ubiquitous data sources and transforms them into intelligent, actionable insights through a DIA which combines computer vision, energy monitoring, and conversational AI to support operators in day-to-day manufacturing activities, including inventory management, product counting, packaging quality assurance, machinery anomaly detection, and order fulfilment based on available stock.
The architecture of ELECTRA links the food manufacturing machinery, the DIA, and the WASABI integration layer through a unified and modular data flow. CCTV cameras stream video via RTSP, while energy meters provide real-time power and operational data from packaging-line equipment. These data sources are processed by the DIA’s analytics modules, where computer vision algorithms extract production metrics and correlate them with energy performance indicators.
The DIA will be integrated with different WASABI components (e.g. OVOS, WISE, and PREVENTION), enabling natural, task-oriented interaction to facilitate workers in the food production environment. ELECTRA’s architecture is deliberately modular, replicable and scalable. Built on open-source frameworks such as OVOS, Docker, and open LLMs such as Llama and standard data exchange formats (JSON, MQTT, REST APIs), it can be adapted to other production environments with minimal integration effort. In support of replicability, the developed assistant will be published through the WASABI White- Label Shop (WWLS) instance, together with documentation, and demonstration materials, allowing other SMEs to reuse, extend, or commercialise the solution.
What will be demonstrated
The experiment is designed to demonstrate the DIA in realistic conditions and show both operational value at HELIOS and replicability for other food-manufacturing SMEs.
First, the ELECTRA solution’s technical components will be developed and integrated in the DIA, leveraging consistent data flow from energy meters and cameras, to support workers through natural language interface for at least four food-manufacturing tasks, such as packaged items counting, monitoring machinery activity and logging production batches.
Once the development of the ELECTRA solution is finalized, it will be deployed on the operational packaging line at HELIOS, tested in real time, and validated under real production conditions so that usability, reliability, and impact on production efficiency and sustainability can be evaluated. The solution’s outcomes and impact will be monitored and assessed, aligned with the experiment’s key performance indicators.
Finally, the developed DIA will be published via a PrestaShop-based WWLS marketplace instance so it can be offered as a reusable skill for other SMEs. The marketplace instance will include documentation and demonstration material to enable straightforward replication and adoption, including at least one open technical webinar and a short video demonstration. The shop and its core modules will also be evaluated using the evaluation forms provided by WASABI.
Who will be involved
ELECTRA is implemented by a three-entity consortium with clearly defined roles:
HELIOS: Coordination and pilot owner
HELIOS is the lead SME and project coordinator, providing a real production environment, with already deployed monitoring hardware, as the pilot site for ELECTRA. HELIOS leads overall coordination, and provides business/process requirements for the experiment, leads deployment and ensures operator training on the packaging line, and contributes to evaluation through operator feedback and iterative improvements.
Plegma Labs: Technical development and exploitation
Plegma leads the technical work on the DIA and WASABI integration and its key activities include the design of the system architecture and interfaces that will integrate WASABI components, analytics, and data integration through CCTV/energy meters, including WWLS integration. Plegma will also implement the DIA’s core functionalities and integrate the DIA with HELIOS data sources via open data exchange formats and REST APIs, and provide technical support during deployment, validation, and improvement activities in HELIOS and lead exploitation planning. Finally Plegma will deploy/configure the WWLS marketplace instance, distribute the ELECTRA DIA via WWLS, set up the core functional module for the seller profile, and upload the related OVOS skill to the shop.
DIGIAGRIFOOD: Ethics, legal compliance and dissemination
DIGIAGRIFOOD EDIH will lead communication activities, referencing this experiment as a regional showcase of trustworthy, human-centric AI adoption. Also, they will contribute their capacity in digital transformation and sustainability for agri-food SMEs, ensuring that the experiment aligns with ethical AI principles, the AI Act, and GDPR compliance. They will consult on responsible AI compliance, support data governance, oversee ethics compliance.
Why the experiment is relevant
ELECTRA is directly relevant as it targets the core operational challenges of food-manufacturing SMEs. By turning real-time video and energy-meter data into actionable insights accessible through natural-language interaction, ELECTRA aims to enhance human and AI collaboration in manufacturing, as the developed DIA will reduce manual counting and logging effort, provide continuous digital visibility of inventory levels, assist workers towards evidence-based decision making, improve packaging-line quality assurance, and increase traceability through automated batch logging. Sustainability and resilience are supported by integrating energy monitoring with production information (including reducing unnecessary cold-storage checks), while responsible deployment is ensured through full compliance with EU principles for trustworthy AI and relevant legislation, including the AI Act and GDPR.
Currently, SMEs largely depend on manual supervision and legacy systems with limited real-time data access, despite significant efforts towards improving efficiency and sustainability in the food manufacturing domain. This makes it difficult to optimize production processes and energy usage, since multiple tasks such as product counting, registering batches, and machine monitoring are often manual, resulting in fragmented data, delayed reporting, and higher operational costs. Existing systems for packaging control and inventory management mostly rely on complex high-cost infrastructure, such as dedicated computer vision solutions, barcode/RFID tracking systems, and advanced warehouse management platforms. Such solutions require substantial capital investment and specialized expertise, placing them beyond the practical reach of most food manufacturing SMEs.
Challenge 1: Limited real-time data access in food manufacturing.
SMEs often face inventory management inefficiencies, product batch handling issues, and higher production line costs, as the integration of data is often delayed and/or fragmented.
Challenge 2: Manual activities can lead to inefficiencies and wasted resources.
Existing manual processes adopted by SMEs, such as counting and verifying packaged products, slow production and potentially introduce errors, resulting in overproduction and energy management issues (e.g., opening fridges often to check inventory), and even expired stock.
Challenge 3: Low adoption of intuitive and human-centered digital tools.
Food manufacturing facilities either do not integrate digital tools to aid workers or adopt tools with complex dashboards that discourage engagement and require specialized training.
Challenge 4: Lack of openness and interoperability across digital solutions.
SMEs often adopt proprietary or isolated digital tools, which limits interoperability and knowledge transfer. This fragmentation slows digital transformation and increases scaling costs.
Objective 1:
Develop, and demonstrate a DIA for inventory management, production line efficiency, and packaged product quality assurance in food manufacturing facilities.
A task-oriented, conversational assistant will be developed, integrated with the Closed-Circuit Television (CCTV) system and energy-metering infrastructure at HELIOS Bakery. The solution will automatically count packaged items, monitor machinery activity, log production batches, and provide real-time natural-language insights to workers, and it will be deployed on the operational packaging line at HELIOS with real-time testing and iterative feedback from operators to validate usability, reliability, and sustainability. Through this approach, the assistant will enhance production efficiency, improve packaging-line quality assurance, and reduce manual workload and human errors in stock handling and process supervision.
Objective 2:
Promote sustainability, resilience, and human-AI collaboration in food manufacturing and ensure transparency and interoperability with open, modular, and replicable technologies.
Integration of energy meters for packaging-line machinery will enable the monitoring and analysis of power consumption, and the assistant will correlate energy data with production throughput, helping operators identify periods of unnecessary consumption or anomalies and supporting optimization of energy usage and insights for equipment efficiency. The solution will also integrate open components from the WASABI ecosystem (such as OVOS, RASA, and open LLM models (e.g., Llama) for task-oriented dialogue, rely on open data exchange formats, and be deployed with Docker, while complying with GDPR and AI Act principles.
Objective 3:
Distribute the developed DIA via a WASABI marketplace instance through deployment of WWLS instance hosting the developed DIA as a reusable skill for other SMEs in the food manufacturing sector. The marketplace instance will include documentation and demonstration material, enabling straightforward replication and adoption across the food manufacturing domain.
ELECTRA is carried out in the food manufacturing sector, focusing on the packaging quality assurance and inventory management of a food manufacturing SME. The experiment targets day‑to‑day shop‑floor needs such as inventory tracking, packaging‑line supervision, and quality assurance, by introducing a DIA that combines real‑time CCTV feeds and energy‑meter data from packaging machinery to generate actionable insights through a conversational AI interface.
The experiment will be carried out at HELIOS’s facilities in Spata (Attica), Greece, in real-life food manufacturing conditions. The ELECTRA solution will be deployed on the operational packaging line at HELIOS, with real‑time testing and iterative feedback from operators to validate usability in a real food manufacturing environment. HELIOS provides the pilot site and access to a real production environment with already deployed hardware (IoT sensors, energy meters, CCTV) that will be used as a basis for the DIA integration and validation.
The target users are employees in the facility who interact with packaging and inventory processes, i.e., workers/operators and supervisors who will use the assistant through natural, task‑oriented dialogue (through intuitive voice or text commands) to ask for information (e.g., inventory levels, production needs, energy consumption) and receive real‑time answers without need for technical expertise or manual data entry. Beyond the pilot, the solution is intended for other SMEs in the food manufacturing sector via distribution through a WASABI marketplace instance.
ELECTRA does not introduce any additional safety, regulatory, or technical constraints beyond those already present in the pilot environment. The solution operates purely at software level and builds upon already installed and operational infrastructure, including energy meters and CCTV systems at HELIOS, without requiring new physical installations, electrical modifications, or changes to certified equipment. Therefore, it does not pose any additional safety risks nor does it require new certifications, as it neither replaces nor alters existing certified components but functions as a data-driven analytical and conversational assistance layer. Access to the necessary operational data is already ensured through the established collaboration between HELIOS, as consortium leader, and PLEGMA, with existing data governance and technical pathways in place. Furthermore, integration is limited to the systems already described in the experiment documentation (energy meters and cameras). As such, all relevant constraints regarding safety standards, certification, data access, and system interoperability have already been addressed within the current framework. In terms of data flow, the DIA will integrate with HELIOS data sources and the WASABI ecosystem using open and standardized data exchange formats, i.e., data from cameras and energy meters will be transmitted via open protocols such as RTSP, HTTP, and MQTT, while the energy data will be using formats, such as JSON and XML, exposed through through RESTful APIs.
EXPECTED IMPACT
EXPECTED IMPACT
The ELECTRA experiment is expected to deliver measurable impact at HELIOS by introducing a DIA that acts as an interactive co-worker on the production floor, and supports inventory tracking, packaging-line supervision, and quality assurance through natural, task-oriented dialogue. The assistant will combine real-time video analytics and real-time energy-metering data to automatically count packaged items, monitor machinery activity, and log production batches, while providing contextual, voice or text-based feedback to supervisors. Expected operational benefits include reduced manual counting/logging effort and improved real-time visibility and evidence-based decisions, resulting in reduced over-production, which in-turn leads to cost-savings; improved packaging-line Quality Assurance, and higher traceability via automated batch logging, with improved worker support through natural language interaction. ELECTRA also targets sustainability and environmental benefits through improved energy awareness and more efficient control of energy-intensive processes. A practical and measurable impact will come from improved control of energy-intensive cold-storage facilities. By providing continuous digital visibility of inventory levels, the DIA will eliminate the need for workers to repeatedly open fridge doors simply to verify stock, reducing thermal losses and unnecessary energy consumption.
Beyond the pilot site, ELECTRA is expected to generate technological, economic, and business value for all consortium members while contributing to the wider digital transformation of Europe’s manufacturing SMEs. For DIGIAGRIFOOD EDIH, the experiment reinforces its role as a regional enabler of responsible AI adoption in the agri-food and manufacturing sectors, strengthening advisory and training activities through a concrete demonstration of trustworthy, human-centric AI integration. Outreach and dissemination activities aim to reach over 300 SMEs and stakeholders and will include at least one open technical webinar, and a short video demonstration published on the WWLS. For Plegma Labs, the experiment advances its applied industrial AI activities by extending capabilities toward building conversational, task-oriented AI assistants and opens new commercial opportunities through distribution via the WASABI White Label Shop (WWLS) and an AI-as-a-Service (AIaaS) direction. Success will be demonstrated through real deployment and validation in HELIOS’s operational environment (the solution deployed on the operational packaging line, real-time testing, iterative feedback from operators, and an evaluation including adoption of the DIA and its deployment through the WWLS).
Some of the experiment’s measurable KPIs that will be monitored are the following, with stated baselines and post-experiment targets:
- KPI 2 Food-manufacturing tasks handled by DIA: Current data (pre-experiment): 0. Expected outcome (post-experiment): ≥ 4.
- KPI 5 Number of meters integrated in the DIA: Current (pre-experiment): Existing meters but not connected to DIA. Expected outcome (post-experiment): > 3.
- KPI 8 Number of open components integrated to DIA: Current (pre-experiment): 0. Expected outcome (post-experiment): > 3.
- KPI 9 Number of IoT devices integrated: Current (pre-experiment): Existing devices not connected to DIA. Expected outcome (post-experiment): > 7.
- KPI 10 Positive feedback: Current data (pre-experiment): 0. Expected outcome (post-experiment): ≥ 5.
- KPI 11 Total outreach of SMEs/stakeholders: Current data (pre-experiment): Expected outcome (post-experiment): ≥ 300.
- KPI 12 Newsletters: Current data (pre-experiment): 0. Expected outcome (post-experiment): 2.
- KPI 13 Posts on social media: Current data (pre-experiment): Expected outcome (post-experiment): ≥ 4.