Con esta herramienta te facilitamos un acceso a todas las ofertas y demandas de tecnología europeas y a búsquedas de socios para participar en propuestas europeas de I+D publicadas en la red Enterprise Europe Network, pudiendo filtrar los resultados para facilitar las búsquedas más acordes con tus necesidades.

¿Quieres recibir estos listados de oportunidades de colaboración en tu correo de forma periódica y personalizada? Date de alta en nuestro Boletín

Para optimizar los resultados de la búsqueda, se recomienda utilizar términos en inglés.

Búsqueda de socios industriales para un proyecto de investigación y optimización de procesos de fabricación basados en datos

Resumen

Tipo:
Búsqueda de socios
Referencia:
RDUK20160106001
Publicado:
11/01/2016
Caducidad:
11/01/2017
Resumen:
Un equipo de investigación de una universidad británica busca un socio industrial interesado en participar en un concurso de proyectos de investigación financiado por el programa "Diseño por la Ciencia" del Consejo de Investigación de Ingeniería y Ciencias Físicas (EPSRC), sobre optimización de procesos de fabricación basados en datos. El socio buscado, una pyme o gran empresa, se encargará de llevar a cabo el trabajo de evaluación en un entorno de fabricación. El objetivo es establecer acuerdos de investigación.

Details

Tittle:
Industrial partners sought for data-driven, manufacturing process optimisation, research project.
Summary:
A research team from a UK University seeks an industrial partner to join a research project bidding for funding through the UK Engineering and Physical Sciences Research Council (EPSRC) "Design by Science" programme, relating to data-driven optimisation of manufacturing processes. The partner´s role would be to deliver evaluation work in a manufacturing environment. The partner may be an SME or a large industrial company. Research cooperation agreements are offered.
Description:
Process control data obtained from sensors (optical, chemical, vibrational, etc) contain uncertainties that can be caused by natural variations of parameters such as, characteristics of raw materials; wear of instruments; measurement errors; production environment and sensor noise. These factors affect manufacturing processes making them unstable and inefficient in terms of production outputs, value and/or quality. Handling of these negative factors is costly and laborious.

Typical manufacturing processes involve multiple stages. Each stage uses measurements of process parameters that are required in order to define the settings of machinery equipment. The technological data can be distorted by noise, and so the settings can be defined beyond the optimal mode of equipment. The noise present in data is one type of uncertainty that affects results of modelling and optimisation.

For optimisation and control of a manufacturing process, technological equipment is typically modelled by using an analytical approach whose output is dependent on measurable parameters of the technological process. Typically these models are learnt from real process in order to provide accurate estimation and prediction. The lack of knowledge required for reliable modelling of process causes another type of the uncertainty.

Both types of uncertainties present in data and in technological models will affect the manufacturing process, decreasing the product value and yield. Therefore the cost-efficiency of production will be significantly improved if the manufacturing process can be optimised and modelled in a new conceptual way of minimising impact of uncertainties.

A UK university ICT research group has identified a new approach to optimisation and decision making recognising the uncertainty present in process control data.

In cases of real scale data read from sensors and equipment (also known as in-line measurement data), the above methodological advantages are achieved using computational (Graphics Processing Unit and Field-Programmable Gate Array) accelerators.

An example of manufacturing use is the optimisation of semiconductor manufacturing in part of designing of experiments and finding patterns which lead to value maximisation.

This approach has been shown to outperform existing approaches in cases of predicting health care outcomes as well as predicting failure events that can happen in technological process. This method allows technologists to interpret decision models within the quantitative, probabilistic framework and estimate possible risks most accurately. The new approach allows risks to be interpreted as uncertainty intervals that are reliably inferred from available data, without theoretical assumptions that are often unrealistic.

The optimisation methodology is expected to benefit users in a broad spectrum of manufacturing applications related to Automotive, Chemistry, Bioengineering, Energy, Electronics, Food, Materials, and Pharmaceutical industries.

The research group is forming a consortium to bid for funding through the UK EPSRC "Design by Science" competition in order to further this work with the development of practical, manufacturing process control applications. It therefore seeks industrial partners (SMEs and/or larger companies) to carry out evaluation assessment in real manufacturing environments.

It is likely that the consortium formed for this competition will have the opportunity to bid for further, possibly European funded programmes, in the future. Hence, although the call deadline for "expression of interest" is 21st January 2016, the deadline for expressions of interest will be 30th June 2016. The project itself will have a duration of 3-4 years and the funding intervention rate for EPSRC competitions is 80% of the full economic cost.

Research cooperation agreements are offered.
Advantages and Innovations:
· The main advantage is the provision of full probabilistic information that is crucially important for estimating and minimising risks, specifically in safety-critical systems and high-value manufacturing.

· The approach overcomes the limitations of Physics-based modelling which often requires unrealistic theoretical assumptions.

· The approach enables identification of the most influential factors in multi-dimensional technological data.

· The approach offers the capability of discovering new patterns (insights) represented by new combinations of settings and parameters which would not have otherwise been seen.

· The desired solutions are represented by influence diagrams (i.e. fault trees) interpretable by Technologists/Managers.
Stage of Development:
Proposal under development
Technical Specification or Expertise Sought:
Manufacturing operations and environment suitable for the evaluation of this approach and provision of relevant feedback to enable development of practical process control applications.

Partner sought

Partner Sought:
The UK university is looking for SME or larger industrial partners. The partner´s role would be to provide evaluation of proposed technology aimed at optimising equipment settings, sensor readings and outputs. A research cooperation agreement would be offered.
Type of Partnership Considered:
RDR

Client

Type and Size of Client:
University
Already Engaged in Trans-National Cooperation:
Si
Languages Spoken:
English

Dissemination

Restrict dissemination to specific countries:
Belgium
Denmark
France
Germany
Netherlands
Norway
Poland
Spain
Sweden
United Kingdom

Programme-call

Coordinator Required:
No
Deadline for Call:
30/06/2016
Project Duration:
156