Skip to main content

UNIFIED PREDICTIVE MAINTENANCE SYSTEM

Periodic Reporting for period 1 - UPTIME (UNIFIED PREDICTIVE MAINTENANCE SYSTEM)

Reporting period: 2017-09-01 to 2019-02-28

UPTIME aims to design a unified predictive maintenance framework and an associated unified information system in order to enable the predictive maintenance strategy implementation in manufacturing industries. As products become more complex due to evolution of technology, high quality and reliability have become issues of high significance. To reach the required levels of availability, maintainability, quality and safety of production machinery, while considering the system as a whole, and throughout the entire production lifecycle, manufacturing companies are increasingly considering turning to predictive maintenance, by utilising the capabilities of condition monitoring. The UPTIME predictive maintenance system will incorporate information from heterogeneous data sources, e.g. sensors, to more accurately estimate the process performances. Therefore, UPTIME will extend and unify the new digital, e-maintenance services and tools in order to exploit the full potential of predictive maintenance management, sensor-generated big data processing, e-maintenance support, and proactive computing. UPTIME will deploy and validate the solution in the three business cases and diffuse it in the manufacturing community, by creating and managing a community of industrial end-users to maximise impact of UPTIME solution in the industry and accelerating its deployment.
"Main technical activities in the first 9 months of the project have concentrated on the development of the UPTIME predictive maintenance methodology and conceptual architecture. In order to define technical concept of UPTIME, first of all, analysis on the existing models, techniques and platforms implementing predictive maintenance functionalities as well as an assessment of the technological maturity of UPTIME components has been performed. The result of the analysis has been documented as a catalogue containing information regarding the state-of-play of predictive maintenance models and techniques. As a next step, the project was focused on designing the conceptual architecture and specification of UPTIME system. The conceptual architecture forms the basis for the development of the unified information system of UPTIME. During this activity, seven phases of UPTIME architecture have been defined. Each phase will be built according to the five existing baselines components (UPTIME_SENSE from USG - Universal Sensor Gateway, UPTIME_DETECT and UPTIME_PREDICT from preInO Process Engine, UPTIME_DECIDE from PANDDA - ProActive seNsing enterprise Decision configurator DAshboard, UPTIME_FMECA from DRIFT - ProActive seNsing enterprise Decision configurator DAshboard, UPTIME_VISUALIZE from SeaBAR - Search Based Application Repository). Additionally, one new component for addressing the analysis on the legacy and historical data has been identified and to be developed within the project to cover the UPTIME_ANALYZE phase. A functional (high level) and a technical view of the conceptual architecture has been developed. Furthermore, stakeholder requirements, technical and system requirements as well as their specifications have been identified and analysed for the three business cases. Moreover, a 1st version of the UPTIME Predictive Maintenance Management Model has been defined with a detailed description the first 3 facets, namely Maintenance Business Processes, Functional Map, Business Data Model. The model introduces the main high-level concepts developed in the technical architecture and data model, which have been used as guidelines to support definition, specification, design, and implementation of the UPTIME Platform. These outcomes mentioned above have marked the successful achievement of the 1st Milestone ""Availability of the UPTIME System Conceptualisation”.

Three Business cases are involved in UPTIME. The use case of FFT deals with the maintenance of manufacturing equipment within the aviation sector. The use case of WHIRLPOL deals with a complex automatic production line to produce drums for dryer. The use case of MAILLIS deals with cold rolling mill for the production of steel strapping. During the period M1-M9, business and technical requirements and system conceptualisation of these three business cases have been defined. These outcomes have marked the successful achievement of the 2nd Milestone “Availability of the UPTIME Business Cases Conceptualisation”.

Six new component prototypes have been developed within M1 to M12. Requirements towards technical framework of each component from each business case and UPTIME Architecture have been defined. The first prototype of the data acquisition and manipulation component, SENSE, has been released. The first prototype of the diagnosis and prognosis components, DETECT & _PREDICT, has been released. Moreover, the new component ANALYZE has been developed to leverage manufacturer’s legacy data and operational data related to maintenance. The first prototype of the maintenance decision making and actions planning component, DECIDE, has been released. The first prototype of the data aggregation and visualisation component, VISUALIZE, has been released. The first prototype of the data-driven FMECA component, DECIDE, has been released. These first releases of the components prototypes have marked the successful achievement of the 3rd Milestone “Availability of the first release of the UPTIME Framework Components”.

Accordingly, by M15, the first prototype of the UPTIME Platform, which has demonstrated main functionalities of the system, the integration points as well as the interaction among the components, has been released. This has marked the successful achievement of the 4th Milestone “Availability of the first release of the UPTIME Integrated Platform”.

Moreover, development, deployment, integration and evaluation of the first prototypes/deployments to the three business cases have been carried out from M1 to M18. The first iteration of the UPTIME specific architecture design for FFT business case, focusing on 1st phase implementation of the SENSE, DETECT, PREDICT and VISUALIZE components, especially on stream data acquisition and processing as well as detection and visualising significant asset behaviour, has been released. The first iteration of the UPTIME specific architecture design for WHIRLPOOL business case, focusing on 1st phase implementation of the ANALYZE, FMECA & VISUALIZE components, especially on historical & operational data and FMECA analysis, has been released. The 1st iteration of the UPTIME specific architecture design for MAILLIS business case, focusing on 1st phase implementation of the SENSE, DECIDE, FMECA, & VISUALIZE components, has been released. These outcomes have marked the successful achievement of the 5th Milestone “Readiness of first piloting phase of the UPTIME Demonstrators”.
"
The developments in the UPTIME project will directly impact European manufacturing industry, foremost in the sectors represented by the use cases. UPTIME will contribute to reducing failure rates, downtime due to repair, unplanned plant/production system outages and extending component life. By investigating and demonstrating the applicability of the UPTIME predictive maintenance system in the three use cases in different manufacturing sectors, UPTIME will contribute to a more widespread adoption of predictive maintenance and demonstrate more accurate, secure and trustworthy techniques at component, machine and system level.