The GMσD project – strategy for Big Data in Design and Production!

The main objective of the GMσD project is the generation of the first, continuously updating, feature based process capability database. The database, offering objective and easy accessible data on the achievable manufacturing accuracy, will facilitate/ intensify communication and management of variation between Design and Manufacturing as well as across the whole supply chain.

Potentials for Horizon 2020
Factories of the future (FoF-08-2015)

Currently under evaluation
Submission date: 4th of February
The GMσD project:

Application status:

Overall project structure

The diversity of the GMσD project (adressing engineering design, production, metrology, software development etc.) requires special attention to the alignment of the large number of necessary partners, respectively the necessary expertise in the consortium. The objective is to allow for a synchronisation and alignment of in-depth analyses in highly specialised manufacturing areas and centralised research activities ensuring the aggregation of generated results. The consortium will therefore be organised based on seven chosen process areas, each of which will be formed by process experts from academia and industry focusing on a specific manufacturing technology and a specific market segment. The central coordination and support of these “knowledge hubs” will ensure a smooth deployment of all joint research and innovation activities which are consequently of strategic importance for the different partners.

The envisaged structure of the collaboration in the consortium is in accordance with the basic aims of a coherent and relevant industry-driven research and innovation strategy according to Public Private Partnership (PPP) character of the addressed topic under the Horizon2020 Factories of the Future topic. These are:
  • Solve problems jointly with industry
  • Strengthen European industrial leadership
  • Facilitate prioritisation of R&I in line with industry needs
  • Leverage research and innovation investments
This PPP character of the GMσD project will moreover be supported by a systematic effort for the implementation of generated project results, i. e. the Implementation and Population of generated PMσD databases as well as the aggregation in the GMσD database, in industrial practice.
Based on a systematic alignment of expertise and efforts in the chosen Knowledge-Hub structure, the project will allow for regular feedback and review of user requirements that ensure that the developed solution meets the needs of practitioners in Small-, Medium-sized and Large-enterprises from different industry sectors.

   Injection Moulding

The he term ‘injection moulding’ refers to a variety of different polymer moulding techniques that involve the production of parts by injecting molten plastic to a shaped cavity. It offers good dimensional control and is particularly suitable for high rate production.

All injection moulding processes are highly complex. The quality of the part produced can be influenced by a large variety of interdependent factors such as raw material used, ambient conditions, machine type, mould design, process parameters, mould cooling strategy, etc. As the achievable tolerance windows are dependent on the size of parts or their features, the management of geometric variation therefore often necessitates high levels of quality-related activities in an industrial setting, which can be very costly.

For these reasons, the availability of objective variation data for injection moulding in a GMσD database would be highly valuable and could be exploited in research, design, and manufacturing. Overall, the GMσD project will establish a common way of understanding variations in injection moulding, set a variation benchmark for each level of product complexity, and identify the types of good practice required to achieve consistent precision in mass production of any injection moulded material , size and shaped products.

   Composite Production

The use of composite materials in structural applications has significantly increased and seen outstanding advances over the last decades. The GMσD project will investigate Automated Fibre Placement (AFP) processes, which is a production process relevant to the use of composite materials within several key industry sectors of very high strategic and economic value for the EU, namely aerospace, automotive, and wind energy. A Robot/Gantry is used to lay multiple tapes of fibrous reinforcement in a predetermined sequence onto a tool surface. The application of a number of layers is then followed by intermediate vacuum consolidation stages before the full lay-up is transferred to an autoclave (heated pressure vessel) for cure.

Dimensional variation in composite manufacturing is still a very significant cost for manufacturers today. One study of defect types for example indicates that about a third of those are directly related to dimensional fidelity/variability issues. This is of particular importance for the aerospace industry, where composites are often used for critical structures parts in commercial aircraft such as spars, fuselage sections and wing skins. A non-conformance due to manufacturing variation in this type of application requires costly leads to rework, an in-company technical review, and a customer concession before being signed off. Improvements in dimensional variation in composite manufacturing processes could therefore provide a significant competitive advantage for the European aerospace, automotive and wind energy industries, and at the same time would facilitate the greater use of composites in more cost-sensitive application areas.

A systematic investigation of manufacturing variation and the database solutions conceived/provided in the GMσD project consequently support a more holistic view of tolerances for both composite and metallic elements in the assembled structures. This will benefit European manufacturers by reducing the necessity (and cost) of testing and will also allow them to build composite components in more complex structures and assemblies. As it is, moreover, very difficult in an academic context to acquire sufficient data sets to unambiguously identify the detailed relationships between cause and effect in the manufacturing processes that drive variability, the provided large data sets will enable academia to test models, validate approaches and understandings and contextualise
the composites dimensional variability issues against variability in other materials and process.


Machining is any of various processes in which a piece of raw material is cut into a desired final shape and size by a controlled material-removal process. The wide variety of processes, that have this common theme, controlled material removal, are today collectively known as ‘subtractive manufacturing’. To limit studies, the GMσD project will focus on computer numerical controlled (CNC) and portable machining processes used in contract manufacturing of large and complex steel structures, i. e. equipment and solutions for offshore, energy processing and machine-building industry.

As in other manufacturing operations, geometric variation in machining results from a variety of factors, e. g. process type, operator skills, material, process parameters, designed features, etc., and is largely size dependent. However, in contrast, the achieved accuracy can directly be related to the cost of manufacturing operations. For example, relative production time is a function of surface finish produced, so that suboptimal specifications can easily increase part costs by an order of magnitude. A variation database for CNC and portable machining processes will be systematically developed in the GMσD project. This will provide a better understanding of design- and production-related costs which in turn can be used to find the optimal production strategy for a given part. The aim is to define the most optimal machining process based on defined functional requirements: cost, production time, material, environmental conditions, etc. and thereby provide a competitive advantage to European machining companies.

   Additive Manufacturing

Early adoption of Additive Manufacturing techniques focused on prototyping applications, generally involving polymer materials. In contrast, today’s technologies support productionquality manufacture using of a wide variety of metals. The relatively new and rapidly evolving Additive Manufacture is consequently still enjoying high growth with many new potential applications yet to be exploited. To help explore the potential of Additive Manufacturing and to enable a wide adoption within the European Manufacturing Industry, the GMσD project will investigate Selective Laser Melting (SLM) techniques, which have become a relevant production method for manufacturing ready-to-use parts made from metals such as stainless steel, nickel, titanium, and aluminium alloys.

The fundamental process steps SLM techniques of layer based coating and melting of metal powder are subject to a variety of technology-specific effects and noise factors that can be detrimental to the geometric accuracy of the resultant part. Currently, the potential for application of SLM is limited by a lack of in-depth understanding and knowledge in key areas such as: the interdependencies between the factors that dictate geometric variation; the design strategies that are appropriate when employing SLM; or the potential for combining additive manufacturing techniques with post-processing steps.

The GMσD project will help to capitalise on the full potential of Additive Manufacturing technologies such as SLM by addressing these key knowledge gaps, leading to a better understanding of the benefits and limitations of these processes with respect to the manufacture of specific geometric features as well as promising combinations with mature manufacturing processes. This, combined with the simplification of research efforts focusing on the predictability and the performance of corresponding processes, will offer the chance for a wider adoption of Additive Manufacture in industrial practice.

   Laser Welding

Within the diverse field of material joining processes, the GMσD project has elected to concentrate on Laser beam welding. Whilst it is widely applicable, laser beam welding has become particular important for the automotive industry. The possibility of long welding seams joining thin-walled and high-strength sheet metal offers enormous potential for the production of light weight car bodies.

However, although the technology has existed for over 20 years,
its application is currently still limited due to enormous infrastructure and equipment costs. Moreover, laser beam welding processes are largely influenced by process uncertainties and thus, high and (until today) unknown geometric variations of the welded parts. The current state of the art does not yet support a well-founded, reliable and cost-efficient tolerance specification for laser beam welded parts.

The GMσD project aims to deliver a significant improvement in the quality performance and reproducibility of laser beam welding by supporting the choice of adequate joining scenarios, which lead to smooth surfaces of the deep and narrow welding seams required for corrosion resistance and paintability. The project will investigate the relationship between the focus and amplitude of the oscillating laser beam and the resultant geometric variation in laser based technologies, taking into account influencing factors such as the existing interdependencies of varying joining features, gaps, etc. Overall, the GMσD project will support a deeper understanding of laser based technologies and facilitate the adoption of innovative
laser welding process in the European Manufacturing Industry.


Cast iron is one of the oldest, and still one of the most important, manufacturing processes used for large volumes of structural components, such as engine blocks and axle boxes for heavy vehicles and bearing housing in many stationary and moving applications. Sand casting requires precise positioning of the cores that will form the cavities in the castings. Each core varies dimensionally and also in positioning within and between moulds. The positioning of larger cores is often robotized, while smaller are normally placed by hand. This causes geometrical variation within and between engine blocks during manufacturing that influence both the geometrical tolerances for engine design and the final material properties in the thinner internal walls of the component.

Higher environmental demands call for stronger and lighter components that increase sensitivity to geometrical variations; thinner walls and tougher operating conditions. However, the principle investigation is relevant in the overall perspective for most iron casting and similar processes for other materials such as nickel-based alloys and titanium, which also requires positioning of cores for casting of thin-walled parts. The scope of GMσD is limited to the geometrical variation of iron castings. GMσD would reinforce the necessary knowledge build-up in both academia and industry on interdisciplinary communication regarding variations in order to facilitate the co-operation between design and manufacturing departments on component design, casting system design and cast process development to reach guidelines for tolerance determination.

   Sheet Metal Forming

Sheet metal forming is a common means of producing surfaces and structures in many industries within the transportation sector, and is therefore an important forming process for the GMσD project to consider. The process lends itself well to high volume production, which makes it ideally suited to mainstream automotive body shell applications in particular. In fact, the majority of today’s car body shells consist of 95% or more sheet metal components, with castings occasionally being used for certain applications.

Stamped automotive body panels may undergo a number of operations in order to achieve the finished component and the overall forming process, despite having been employed in the industry for several decades, is still subject to variation which can be difficult to predict. Furthermore, the growing complexity of vehicle styling (and hence the form of exterior panels), coupled with the development of new high strength, lightweight materials (e.g. aluminium and new grades of high strength steel), has added more variables to an already complex process.

Understanding and controlling variation in car body shells is critical to the quality of the finished vehicle, with the premium market in particular demanding exceptionally high levels of fit and finish. With the average car body shell consisting of over 250 individual panels, there is significant scope for the variation of these panels to result in unacceptable tolerance stack ups over the length of the vehicle.

The GMσD project will investigate variation in the sheet metal forming process, considering the effects of multiple forming operations, and those of different material grades, giving designers and engineers within the European Manufacturing Industry a better understanding of process capability and the influence their design can have on the variation of the end product.

Process areas

According to their relevance for different sectors of Europe’s manufacturing industry (see figure above), four widely adopted and mature manufacturing technologies (Injection moulding, Machining, Casting and Sheet Metal Forming) and three (novel) enabling manufacturing technologies (Composite production, laser based material joining and Additive manufacturing) have been chosen for investigation. Bringing together corresponding and complementary expertise, the project will allow for a comprehensive overview of manufacturing variation and will enable its aggregation, thus the creation of a global manufacturing variation database (GMσD).