With all its twists and turns 2020 is eventually going to end with a good few achievements in the EFPF projects. The last 12months period focused on pilot activities i.e. the design, development and implementation of innovative applications that make use of EFPF platform and services to address specific needs of EFPF pilot partners
Based on pilot requirements gathered early in the project (described in D2.3 - Requirements of Embedded Pilot Scenarios, accessible on the project website), a number of innovative solutions were designed to be offered through the EFPF platform. An overview of some of these was provided in the November blog. Here we highlight some of the other pilot solutions being developed in the EFPF project.
1. Data Analytics – Machine Profile: Manufacturing companies require effective data analytic solutions that can help them optimize their production as well as supply chain
related processes. The Data Analytic solutions developed in EFPF enable manufacturing companies to perform real-time monitoring of the machine profiles and identify any problems based on machine vibration profile methodology. The analytic solution is offered through a dashboard that is integrated in the EFPF Portal. The extensible nature of this solution allowed the integration of various data analytics methodologies/algorithms for tonnage forecasting and a deep learning extension for price forecasting for various materials. The implementation of the solution is illustrated here.
2. Matchmaking: Finding new partners and collaborators is a tricky business in the manufacturing domain. Partners are usually
found through previous contacts and word of mouth recommendations. However, problems are faced when finding partners for new products and services e.g., where to search for new suppliers, how to compare
between suppliers and how to contact potential collaborators? To address these questions, the EFPF platform offers Matchmaking and Agile Network Creation solution that uses federated search techniques to help companies find new partners/suppliers and setup collaborative teams or networks for new products and services. The integration of this solution in the EFPF platform is illustrated here.
3. Blockchain-based Delivery Application: EFPF users and manufacturing companies in general face problems related to the track and trace of assets in the supply chain. Generally, the companies lack transparency and visibility of material and assets in the
different phases of a circular supply chain. This problem relates to the lack of a common way or platform where information about assets can be added and tracked.
The EFPF blockchain solution for supply chain addresses this issues by providing a web-based Dapp (Delivery Application) that allows them to track and trace (to create an asset and add details and metadata, monitor and update its status, search corresponding documents and history, track transportation etc. in a legitimate way) different assets in a supply
chain. At the back-end based, a blockchain with smart contracts support is used to automate the processes and ensure immutable transactions, permissions handling and identity management. A mobile Dapp is also developed to add and receive real time data information related to the transportation and delivery processes. The implementation of this solution and a use-case scenario is illustrated here.
4. Workplace Environment Monitoring: This EFPF solution allows manufacturing companies to monitor their workplace environments and make sure that they comply with relevant health and safety obligations and regulations.
This solution makes use of IoT devices and EFPF solutions, such as the TSMatch Gateway that enables communication and discovery with available IoT devices and provides an interface for the end-user and/or a specific service (e.g., analytics tools). The overall solution improves production efficiency by monitoring and maintaining the optimal environmental conditions required for the production. The implementation of this solution is illustrated here.
5. Data Analytics – Anomaly Detection: Many manufacturing companies require data analytic solutions that can help them identify unexpected behaviour in the operations of their machines. Undetected anomalies can cause critical incidents such as a technical glitch or lead to significant problems such as machine breakdown.
Timely or advance identification of anomalies can allow manufacturing companies to save costs and time in scenarios such as equipment malfunctions or quality degradation. Real-time detection of anomalies also allows companies to tune their KPIs and better evaluate their operational performance. The anomaly detection solution in the EFPF platform allows manufacturing companies to identify anomalies in their machine behaviors both using historic dataset or real-time data stream. The solution is offered through a simple GUI-based workflow – without needing extensive data science expertise. The implementation of this solution is illustrated here.
Stay tuned to the EFPF website if you are interested in learning more about the EFPF pilot and other activities.