Community Research and Development Information Service - CORDIS

H2020

IMPROVE Report Summary

Project ID: 678867

Periodic Reporting for period 1 - IMPROVE (Innovative Modeling Approaches for Production Systems to raise validatable efficiency)

Reporting period: 2015-09-01 to 2017-02-28

Summary of the context and overall objectives of the project

Within the manufacturing industry, the complexity of production plants is steadily rising due to increasing product variances, product complexity, and pressures for production efficiency. Production systems must therefore now evolve rapidly and operate optimally, creating challenges for larger industries and serious problems for SMEs without the needed expertise or sufficient resources to adapt new technical possibilities.
By adopting new innovative technologies to manufacturing enterprises, they are able to create more jobs. By now already 30 million people are employed by manufacturing companies. Additionally, the innovative technologies integrated in the whole life cycle of factories will provide opportunities for novel business strategies. The proposed IMPROVE results have a number of positive impacts regarding workplace attractiveness. First and foremost, the amount of monotonous tasks throughout the whole factory lifecycle is greatly reduced. This is enabled by reduced implementation effort, as a result of the modelling and model learning approach, and novel HMI concepts, which support decision making and process adaption, as well as a reduction of recurring manufacturing tasks as a result of predictive maintenance. Furthermore, services and novel interaction methods to support handling of increasingly complex production control require great know-how especially in the design process, and therefore provide ample incentive for engineering graduates and doctorate holders to design and work on cutting edge manufacturing systems.
The main objective of IMPROVE is to create a virtual Factory of the Future, which provides services for user support, especially on optimization and monitoring. Anomalous behaviour is detected automatically by comparing sensor observation with an automatically generated model, learned out of observations. This will ensure and establish a cheap and accurate model creation instead of manual modelling, which is expensive and time consuming.
These learned models will also be used to predict the condition of vital components and estimate their remaining lifetime to reduce the risk of unexpected downtimes. Optimizations will be performed and verified through simulations before they are transferred to the real plant. Root Causes Analysis will be performed on the plants alarm data to derive the root cause of a fault and therefore speed up the repair process.
These results of the aforementioned skills will be available through a Decision Support System (DSS), which assists the operator during the operation of the plant. The interaction will be done by a new developed HMI (Human Machine Interface) providing the huge amount of data in a reliable manner.
The basis for IMPROVE are industrial use-cases, which are transferable to various industrial sectors.

Work performed from the beginning of the project to the end of the period covered by the report and main results achieved so far

The project work is divided into work packages, which complement each other and together allow to build a comprehensive solution for the SMEs. It began with a requirement analysis, where each project partner indicated how can they contribute to help reach the project goals and how can they cooperate with the other partners. In cooperation with the industrial partners, a set of concrete use cases was identified, which guide the research throughout the project.

The first task was to define and implement an appropriate data acquisition system, since good quality data is a must for a useful analysis. A data acquisition and management system, with a special focus on data security, has been presented in two deliverables, and the implementation is undergoing.

Meanwhile, the industrial partners provided first datasets for research. There are three main research tracks connected to the use cases in this project: condition monitoring, diagnosis and optimisation of technical systems. Condition monitoring is focused on tracking the readings from a specific part in the machine to try and detect when the part is reaching the end of its durability. This allows a preemptive action, that is e.g. replacing the part, before it breaks and causes production standstil or damage to the other equipment. Diagnosis comes into play when an unexpected failure occurs in the system. Its goal is to advise the machine operator what is the root cause of the failure and even suggest a repair procedure. Optimisation on the other hand is aimed at computationally selecting the best parameters for a chosen production aspect. Here, it is performed in a simulated environment, where different sets of parameters are evaluated and the best set can be chosen.

In this project, all three research tracks are addressed in a model-based and data-driven fashion. Therefore, a digital twin of the machine is created as a model learned automatically from the data. We used several different model formalism for the different use cases, and produced the first models based on the available data.
Using these models, we were able to develop analysis algorithms and provide the industrial partners with first results. The work on the algorithms will continue throughout the whole length of the project.

Last but not least, after the analysis is performed, the results must be presented to the machine operators in a user-friendly and understandable format. Therefore, the human-machine interaction is an important aspect and we devote a separate work package to it.

All the work in this project will be concluded with an implementation in project demonstrators and prototypes, developed in the second half of the project. So far, all the the deliverables were submitted on time.

Progress beyond the state of the art and expected potential impact (including the socio-economic impact and the wider societal implications of the project so far)

This project has a potential to yield many valuable and measurable results for the European SMEs. Condition monitoring, diagnosis and optimisation solutions are able to give the competetive edge to the enterprises; they help reduce the costs and production downtime as well as optimise the operational parameteres to maximise the output while minimising the costs.

Condition monitoring solutions will allow the enterprises to monitor the critical parts of their machinery. Replacing or refurbishing a part generates much lower costs and downtime than allowing the part to break, causing a much longer downtime, production loss and possibly damaging other parts of the equipment.

Diagnosis solutions will reduce the time needed by an operator to identify the root cause of a machine breakdown, which reduces the downtime and the production loss. Furthermore, they can support the operator by reducing the number of alarms and making it easier for them to have a clear overview of the plant situation - which helps prevent dangerous situations.

Simulation and optimisation solutions allow the system designers and operators to test alternative plant configurations virtually and find the most beneficial setup without actually tampering with the machinery. Observing the behaviour of a virtual twin of the plant allows assesing the parameter sets and evaluating its performance.

The analysis algorithms developed in the project are innovative and developed in a way that enables SMEs to integrate them seamlessly, without the need to overhaul their existing systems. The main KPIs that can be improved are: machine availability, avoided costs, faults detected prior to failure, reduction of plant downtime, mean time between failure, mean time to repair, overall equipment effectiveness, productivity etc.

Related information

Follow us on: RSS Facebook Twitter YouTube Managed by the EU Publications Office Top