Periodic Reporting for period 2 - PreCoM (Predictive Cognitive Maintenance Decision Support System)
Berichtszeitraum: 2019-05-01 bis 2021-02-28
Cheaper and more powerful sensors, algorithms for specific functions, intelligence, together with big data analytics, offer an unprecedented opportunity to track machine-tool performance and health condition. The evolution of current maintenance practices from condition-based to smart predictive maintenance (Smart PdM) would be an important success factor for manufacturers to reach longer production time, fewer stoppages, higher productivity rates, product quality and reduction of defective products. However, it is estimated that manufacturers still, in general, spend only 15% of their total maintenance costs on predictive (vs reactive or preventative) maintenance.
The PreCoM project aimed to deploy and test a cognitive predictive maintenance (Smart PdM) decision support system (DSS) able to identify and localize damage (in which machines and components), assess damage severity, predict damage evolution, assess remaining asset life, reduce the probability of false alarms, provide more accurate damage detection, issue notices to conduct preventive maintenance actions and ultimately increase in-service efficiency of machines.
The focus of the project was to develop and apply an innovative system for Predictive Cognitive Condition-Based Maintenance (CBM) as a new model for sustainable factories and Maintenance 4.0. The PreCoM system can also be counted as a Smart CBM, i.e. predictive, intelligent, digitalized and automated CBM using one or more Condition Monitoring (CM) parameters (including the time to failure) in order to describe and follow up the condition of a machine (and its significant components), assess its current condition and predict its condition development in the near future in a cost-effective way. The PreCoM system gathers information from all maintenance related areas, such as machines, production, quality, energy, materials, and economy. Its performance is digitalized and automated using a combination of probabilistic and deterministic approaches.
The overall objectives of the PreCoM system are to maintain machine quality (to maintain the machine technical specifications), reduce probability of failures and downtime, ease conducting maintenance actions through facilitating and integrating the application of Augmented Reality (AR), follow up maintenance impact on company business, secure production planning, enhance production continuity and company profitability and competitiveness.
The PreCoM consortium included 3 end-user factories, 3 machine-tool suppliers, 1 leading component supplier, 4 innovative SMEs, 3 research organizations and 3 academic institutions. Together, we developed, demonstrated and validated the PreCoM system in a broad spectrum of real-life industrial scenarios (low volume, high volume and continuous manufacturing). We evaluated the direct impact of the platform on maintainability, availability, work safety, material and energy losses and costs, in order to document the results in detailed business cases and for widespread industry dissemination and exploitation. The evaluation of the PreCoM demonstration in the three use case companies highlighted an impressive, positive impact on the key performance indicators (KPIs) and expectations.
The PreCoM project aimed to deploy and test a cognitive predictive maintenance (Smart PdM) decision support system (DSS) able to identify and localize damage (in which machines and components), assess damage severity, predict damage evolution, assess remaining asset life, reduce the probability of false alarms, provide more accurate damage detection, issue notices to conduct preventive maintenance actions and ultimately increase in-service efficiency of machines.
The focus of the project was to develop and apply an innovative system for Predictive Cognitive Condition-Based Maintenance (CBM) as a new model for sustainable factories and Maintenance 4.0. The PreCoM system can also be counted as a Smart CBM, i.e. predictive, intelligent, digitalized and automated CBM using one or more Condition Monitoring (CM) parameters (including the time to failure) in order to describe and follow up the condition of a machine (and its significant components), assess its current condition and predict its condition development in the near future in a cost-effective way. The PreCoM system gathers information from all maintenance related areas, such as machines, production, quality, energy, materials, and economy. Its performance is digitalized and automated using a combination of probabilistic and deterministic approaches.
The overall objectives of the PreCoM system are to maintain machine quality (to maintain the machine technical specifications), reduce probability of failures and downtime, ease conducting maintenance actions through facilitating and integrating the application of Augmented Reality (AR), follow up maintenance impact on company business, secure production planning, enhance production continuity and company profitability and competitiveness.
The PreCoM consortium included 3 end-user factories, 3 machine-tool suppliers, 1 leading component supplier, 4 innovative SMEs, 3 research organizations and 3 academic institutions. Together, we developed, demonstrated and validated the PreCoM system in a broad spectrum of real-life industrial scenarios (low volume, high volume and continuous manufacturing). We evaluated the direct impact of the platform on maintainability, availability, work safety, material and energy losses and costs, in order to document the results in detailed business cases and for widespread industry dissemination and exploitation. The evaluation of the PreCoM demonstration in the three use case companies highlighted an impressive, positive impact on the key performance indicators (KPIs) and expectations.
During the PreCoM project, the consortium first carried out a progressive identification of critical machines and significant components to be monitored by PreCoM system, definition and determination of the PreCoM system concept and design. The work enabled the consortium to design concretely the PreCoM system and its functionalities, which resulted with 24 (hardware and software) modules connected to a centralised cloud and the overarching ‘PreCoM Brain’ (i.e. the steering rules to secure data flow, interoperability and right recommendations to the end user companies). Second, the consortium developed and adapted the selected modules and technologies for making them interoperable. Third, lab tests of individual modules/technologies and the overall PreCoM were conducted in order to verify and validate them. Finally, the PreCoM system was applied in a 14-month demonstration and validation in three industrial use cases, with continuous monitoring and evaluation of its impact.
The evaluation is summarized below as an average of the impact achieved in the three use cases, based on the components monitored by PreCoM:
- Reduced downtime by about 88%;
- Maintainability (MTTR) improved by about 24%;
- Reduced supervision time in the training for new technicians by about 76%;
- Increased machine overall equipment effectiveness (OEE) by about 5%;
- Applying predictive maintenance by company personnel has been increased from previous level of 0% to about 80%;
- Energy is rationalized by reduction equivalent to about 16%;
- Material loss is reduced by about 16%;
- Work safety was confirmed, as no accidents occurred neither before PreCoM introduction or during PreCoM demonstration;
- Reduced maintenance hours by about 16%;
- The saving in maintenance hours (at minimum) is about 92 hours per year;
- Time to repay the investment (at maximum) is about 1.3 years;
- The expected reduction of maintenance costs in 10 years through reducing maintenance hours is about 190,000 EUR.
Furthermore, we performed a wide dissemination of the PreCoM project and its results. We estimated to have reached over 102,000 people among media and general public (e.g. newspapers, magazines), 36,000 people in the industry ecosystem (through fairs, exhibitions and other events/occasions) and 10,000 people in the scientific community (through scientific papers, presentations at conferences and other events).
We had carried out a comprehensive work for defining exploitable results. In the end, nine IPR results were identified (belonging to different partners), of which three are key results (immediately exploitable). For these latter ones, a complete business plan has been developed.
The PreCoM project has been also part of the Cluster FoF-09 – named ForeSee (http://foresee-cluster.eu) for coordinating EU-funded PdM projects on several activities: working together on standardization issues; developing a road map on PdM; organising a series of dissemination events on PdM.
The evaluation is summarized below as an average of the impact achieved in the three use cases, based on the components monitored by PreCoM:
- Reduced downtime by about 88%;
- Maintainability (MTTR) improved by about 24%;
- Reduced supervision time in the training for new technicians by about 76%;
- Increased machine overall equipment effectiveness (OEE) by about 5%;
- Applying predictive maintenance by company personnel has been increased from previous level of 0% to about 80%;
- Energy is rationalized by reduction equivalent to about 16%;
- Material loss is reduced by about 16%;
- Work safety was confirmed, as no accidents occurred neither before PreCoM introduction or during PreCoM demonstration;
- Reduced maintenance hours by about 16%;
- The saving in maintenance hours (at minimum) is about 92 hours per year;
- Time to repay the investment (at maximum) is about 1.3 years;
- The expected reduction of maintenance costs in 10 years through reducing maintenance hours is about 190,000 EUR.
Furthermore, we performed a wide dissemination of the PreCoM project and its results. We estimated to have reached over 102,000 people among media and general public (e.g. newspapers, magazines), 36,000 people in the industry ecosystem (through fairs, exhibitions and other events/occasions) and 10,000 people in the scientific community (through scientific papers, presentations at conferences and other events).
We had carried out a comprehensive work for defining exploitable results. In the end, nine IPR results were identified (belonging to different partners), of which three are key results (immediately exploitable). For these latter ones, a complete business plan has been developed.
The PreCoM project has been also part of the Cluster FoF-09 – named ForeSee (http://foresee-cluster.eu) for coordinating EU-funded PdM projects on several activities: working together on standardization issues; developing a road map on PdM; organising a series of dissemination events on PdM.
The PreCoM project finalised a solution that goes beyond the state of the art of solutions for predictive maintenance (Maintenance 4.0) currently available in the market. We demonstrated and validated the PreCoM system in operational environment, with important results of economic impact for manufacturing companies (resulting e.g. for avoiding unnecessary stoppages, improving production and maintenance planning). Furthermore, machine tool and component builders benefited from the system for better monitoring and identification of issues, data describing deterioration and breaks, which can lead (in the medium-/long-term) to improvements in the next generation of machines and components. In addition, the system has modules for detecting unhealthy sensors (i.e. to secure data quality) and for digitalizing and automating manual data (in order to make PreCoM fully digitalized and automated). The results from the planned demonstration will be further used for future refinements of the PreCoM system and its modules, continuing to address real industry needs.