Periodic Reporting for period 1 - CLARUS (Optimizing Production and Logistic Resources in the Time-critical Bio Production Industries in Europe)
Reporting period: 2022-09-01 to 2024-02-29
The CLARUS project aims to connect the Sustainable Paradigm in the food industry and AI-based applications, with the goal of developing a platform with high communications and processing capabilities, as well as the use of standardized open protocols and data models that will allow resource consumption assessment and traceability for food industry processes.
Currently, two pilots have been selected for validating the CLARUS solution. The first pilot focuses on the production of frozen food, where energy and water consumption can be reduced using AI and data technologies. The other pilot focuses on the meat by-product production where the aim is to reduce the energy and maintain the quality of the products by optimizing the logistics of the by-products arrival.
CLARUS ambitions include not only contributing to resource and logistic optimization methods through the two pilot solutons, but also making a more general contribution through the creation of a Green Deal Index (GDI).
To demonstrate the impact of the green deal concept, the CLARUS project will provide three Tangibles Expected Outcomes which represent the Key Exploitable Results of the project:
• CLARUS Green Deal Index: methods, tools and data used to calculate the Green Deal Index (GDI)
• CLARUS Data Space: FAIR data models and Industrial Data Platforms tools that are developed and deployed for edge data management, cloud data management, and data harmonization, transformation and sharing
• AI Toolkit: AI algorithms and the trained models that are developed, tested, and validated in the project.
Currently, two pilots have been selected for validating the CLARUS solution. The first pilot focuses on the production of frozen food, where energy and water consumption can be reduced using AI and data technologies. The other pilot focuses on the meat by-product production where the aim is to reduce the energy and maintain the quality of the products by optimizing the logistics of the by-products arrival.
CLARUS ambitions include not only contributing to resource and logistic optimization methods through the two pilot solutons, but also making a more general contribution through the creation of a Green Deal Index (GDI).
To demonstrate the impact of the green deal concept, the CLARUS project will provide three Tangibles Expected Outcomes which represent the Key Exploitable Results of the project:
• CLARUS Green Deal Index: methods, tools and data used to calculate the Green Deal Index (GDI)
• CLARUS Data Space: FAIR data models and Industrial Data Platforms tools that are developed and deployed for edge data management, cloud data management, and data harmonization, transformation and sharing
• AI Toolkit: AI algorithms and the trained models that are developed, tested, and validated in the project.
The work performed and the main achievements within the first reporting period are the following (grouped along the CLARUS project objectives):
CLARUS Green Deal Index:
The performed activities included the identification and definition of a quantitative environmental sustainability methodology (and related KPIs) for the environmental sustainability assessment of food manufacturing systems, to be integrated into a unique Green Deal Performance Assessment (GDPA) methodology. The GDPA methodology has been defined as quantitative metrics that deliver a final index: the Green Deal Index (GDI). Before the development of the GDPA Methodology, the requirements definition activity, which includes the identification, classification, and consolidation of all the requirements needed for developing CLARUS solution was conducted.
The metrics and methodologies selected for evaluating environmental sustainability performances have been described. All the indicators that can be potentially calculated starting from these metrics have been reported for the pilot use case partners as well as the link between the different indicators and the metrics of GDPA methodology able to calculate them. Starting from these indicators, and according to the data available in the pilot’s processes, the first sub-set of indicators has been reported.
CLARUS Ecosystem:
The carried-out activities provided the context for the exploitation and sustainability business modelling through a detailed analysis of the existing market and the business opportunities of CLARUS developments. The conducted market analysis helped to identify and understand competitors’ strengths and weaknesses concerning CLARUS offerings. The performed Opportunity Analysis included an analysis of external factors, identifying legal, political, economic, technological, and societal trends relevant to the developed CLARUS services. Data were collected from market providers mainly through desk research, SWOT and PESTLE analysis. The performed activities allowed the project to structure its value proposition aligned with the market needs. Its conclusions will be factored into the technical solution and tested in the pilot use cases.
The main achievements are:
• Overview of the actual status of the three CLARUS TEOs and an update on the general CLARUS exploitation strategy
• Increased knowledge of CLARUS business environment
• Detailed information on competition in terms of offerings, organisational insights strengths & weaknesses as well as learnings and recommendations for CLARUS offerings
• Input for CLARUS competitive position and strategic planning.
• New Methodology for the Exploitation Roadmap
• Identified main exploitation and business aspects for the CLARUS TEOs.
CLARUS Data Space:
A data-driven industrial platform will be the final goal of the technical effort in CLARUS project building a Data Space for manufacturing, able to improve business opportunities for value-added services relating to the industrial data of the participants involved in the project. The CLARUS Data Space is based on IDSA reference models and architecture (IDS RAM), constituting the standard de facto in European data-sharing landscape. The performed activities started with workshops to define a common methodology, followed by an intensive requirements collection phase by integrating CLARUS tech partners and end-users and finalised with the implementation and deployment of the CLARUS Minimum Viable Data Space (MVDS).
The main achievements are:
• The CLARUS Minimum Viable Data Space has been released.
• The IDS components have been configured and deployed in the cloud infrastructure.
• The IDS Identity Provider has been deployed integrated and tested to start the Cloud-Edge communication.
CLARUS AI Toolkit:
This toolkit encompasses AI algorithms, trained models, and an MLOps workflow definition. It aligns with the approach proposed in CLARUS Data Space and CLARUS Green Deal Index. For the CLARUS MLOps workflow first the building blocks associated with the MLOps strategy have been identified and state-of-the-art MLOps tools have been analysed. Based on collected requirements a set of edge services have been implemented to cover the entire lifecycle of AI models in the edge to cloud continuum. Furthermore, AI algorithms were developed for two use cases, the Primary Food processing and the By-product processing.
The main achievements are:
• Implementation of MLOps Workflow and associated toolchain
• Development of the AI Models associated with the industrial Use Cases
CLARUS Green Deal Index:
The performed activities included the identification and definition of a quantitative environmental sustainability methodology (and related KPIs) for the environmental sustainability assessment of food manufacturing systems, to be integrated into a unique Green Deal Performance Assessment (GDPA) methodology. The GDPA methodology has been defined as quantitative metrics that deliver a final index: the Green Deal Index (GDI). Before the development of the GDPA Methodology, the requirements definition activity, which includes the identification, classification, and consolidation of all the requirements needed for developing CLARUS solution was conducted.
The metrics and methodologies selected for evaluating environmental sustainability performances have been described. All the indicators that can be potentially calculated starting from these metrics have been reported for the pilot use case partners as well as the link between the different indicators and the metrics of GDPA methodology able to calculate them. Starting from these indicators, and according to the data available in the pilot’s processes, the first sub-set of indicators has been reported.
CLARUS Ecosystem:
The carried-out activities provided the context for the exploitation and sustainability business modelling through a detailed analysis of the existing market and the business opportunities of CLARUS developments. The conducted market analysis helped to identify and understand competitors’ strengths and weaknesses concerning CLARUS offerings. The performed Opportunity Analysis included an analysis of external factors, identifying legal, political, economic, technological, and societal trends relevant to the developed CLARUS services. Data were collected from market providers mainly through desk research, SWOT and PESTLE analysis. The performed activities allowed the project to structure its value proposition aligned with the market needs. Its conclusions will be factored into the technical solution and tested in the pilot use cases.
The main achievements are:
• Overview of the actual status of the three CLARUS TEOs and an update on the general CLARUS exploitation strategy
• Increased knowledge of CLARUS business environment
• Detailed information on competition in terms of offerings, organisational insights strengths & weaknesses as well as learnings and recommendations for CLARUS offerings
• Input for CLARUS competitive position and strategic planning.
• New Methodology for the Exploitation Roadmap
• Identified main exploitation and business aspects for the CLARUS TEOs.
CLARUS Data Space:
A data-driven industrial platform will be the final goal of the technical effort in CLARUS project building a Data Space for manufacturing, able to improve business opportunities for value-added services relating to the industrial data of the participants involved in the project. The CLARUS Data Space is based on IDSA reference models and architecture (IDS RAM), constituting the standard de facto in European data-sharing landscape. The performed activities started with workshops to define a common methodology, followed by an intensive requirements collection phase by integrating CLARUS tech partners and end-users and finalised with the implementation and deployment of the CLARUS Minimum Viable Data Space (MVDS).
The main achievements are:
• The CLARUS Minimum Viable Data Space has been released.
• The IDS components have been configured and deployed in the cloud infrastructure.
• The IDS Identity Provider has been deployed integrated and tested to start the Cloud-Edge communication.
CLARUS AI Toolkit:
This toolkit encompasses AI algorithms, trained models, and an MLOps workflow definition. It aligns with the approach proposed in CLARUS Data Space and CLARUS Green Deal Index. For the CLARUS MLOps workflow first the building blocks associated with the MLOps strategy have been identified and state-of-the-art MLOps tools have been analysed. Based on collected requirements a set of edge services have been implemented to cover the entire lifecycle of AI models in the edge to cloud continuum. Furthermore, AI algorithms were developed for two use cases, the Primary Food processing and the By-product processing.
The main achievements are:
• Implementation of MLOps Workflow and associated toolchain
• Development of the AI Models associated with the industrial Use Cases