Embedding design structures in engineering information

Lead Research Organisation: University of Leeds
Department Name: Mechanical Engineering

Abstract

Engineers use design structures, such as Bills of Materials (BoMs), to tailor product definitions, including shape, for particular activities. For example, an engineering BoM defines the as-designed product whereas a manufacturing BoM defines the as-built state of the same product and a service BoM includes information on how the product has been maintained. All of these BoMs relate to the same designed product. However in practice, because of restrictions arising from current computer aided design technologies and associated business systems, different BoMs are usually related to separate digital definitions of the same product. This creates significant data management problems that add cost, time and rework into product development processes. If resolved, substantial business benefits, through improved efficiency and effectiveness of product development processes, could be achieved.

Key challenges for engineers lie in (1) understanding how the range of BoMs and other design structures of a given product relate to each other and the product itself, and (2) in ensuring they have the best design structure(s) for specific tasks. For example, a BoM is a hierarchy of part-whole relationships that are useful when a product breakdown structure is needed whereas engineering design tasks typically need design structures that capture how the part being designed relates to the parts to which it must interface. In this second type of [lattice] structure, assembly mating relationships are needed. These and other kinds of connection relationship are fundamentally different to the part-whole relationships of a BoM.

Embedding allows one instance of a mathematical construct to be superimposed on another [http://en.wikipedia.org/wiki/Embedding]. It has been documented since the 1930s in the mathematics literature. Descriptions of concrete applications are less common but do occur in, e.g., in the shape computation literature. Methods to enable the robust implementation of embedding for use in real-world applications remains an open research issue. This project will explore the feasibility of computational tools that can be used to embed design structures in engineering information. If successful we will demonstrate ways in which engineers can associate multiple design structures with a given design as and when such structures are needed. In doing so, key areas of work will be in
1) The development of user-focussed case studies where the added value of embedding can be better understood and
2) The creation of software prototypes for use in demonstrating this potential.

The project team brings together researchers with track records in engineering design and associated information systems, organisational psychology, mathematics and computing. We will work with industrial and other end user partners to define case studies and use them to support demonstrations of how embedding might be implemented and used to enhance real-world engineering design, manufacturing and through life support processes.

Planned Impact

The proposed research has high potential impact for a number of problem domains and solution providers. For engineering design, the research could improve key decision making activities by changing the ways in which engineers think about and work with engineering information. In addition, there is potential for wider impact in other engineering sectors. For example, in the construction industry, embedding could enhance the efficiency and effectiveness with which Building Information Models (BIM) are created and used through the whole life of a building. Embedding could also add value in other problem domains where there is a need to create relationships across scales or cast new lenses on large data sets, and the research will impact the maths and computing communities by providing requirements for general purpose implementations of embedding. Our challenge lies in realising this impact for such a wide range of audiences and communities.

With respect to the RCUK typology of research impacts, the research will create both academic, and economic and societal impact.

Academic impact: The research will use innovative approaches to implement embedding and in doing so create new cross-disciplinary knowledge on how embedding might support engineering tasks. Scientific advancement will be in the form of new implementation mechanisms that will make embedding accessible to users and improved understanding of requirements for mathematics and computing. The work will improve the health of the academic discipline of engineering design by providing theoretical foundations for future generations of digital design system and improving our ability to describe the concept of embedding and outline its benefits to engineering design. Embedding has the potential to solve the general problem of relating data across scales in many academic communities. Examples include the construction sector (with the introduction of BIM), materials design, medical bio-informatics and big data communities. The project will also develop two highly skilled PDRAs and enhance their CVs through publications and engagement with industry and other academic disciplines.

Economic and societal impact: If it became possible to easily embed design structures in engineering information, as and when needed, then the efficiency and effectiveness of product development processes would be dramatically improved. This, in turn, could improve the innovation capacity of the organisations that operate these processes so contributing to wealth creation and economic prosperity. One of the real challenges in maximising environmental sustainability and protection through the life of large complex products lies in the poor quality of the product data that is available to engineers and other professionals once a product is in use. This is especially so when there is a need for information on how the product has been used after it was commissioned. For example, in the nuclear industry, information related to radiation exposure could be embedded into the definition of a product and then used at the end of its life when decisions related to disposal are made. The project will also contribute to the development of engineering students (undergraduate and postgraduate): at Leeds and the OU through student projects linked to the research (e.g., students might create case studies from their own design work and explore the potential benefits of embedding) and in other institutions through digital learning assets published on Jorum (or equivalent) and downloadable software prototypes whose use will be integrated with projects associated with the digital assets.
 
Description We uncovered a latent industry need to be able to associate multiple BoMs with one of more descriptions of a given design and demonstrated that lattice theory can be used to underpin such a technology. The following summarises our achievements with respect to the project objectives.
1) To specify three case study scenarios in collaboration with the industrial partners and other end users.
Two robot case studies are reported in papers. A third, related to an industry partner's product, was developed. The detail of this is confidential but we replicated the design structures and their representations in a synthetic case study used the definition of a technology demonstrator (see later, Objectives 3 & 4). Project staff supervised six UG student projects where students developed case studies from their own industry experience including aircraft maintenance and the automotive and chemical sectors. Together our case studies covered BoMs to support manufacture, assembly, maintenance, supply and service, along with other kinds of design structure such as functional decompositions and descriptions.
2) To review approaches that could be used to embed design structures in product information from these case studies and implement at least two in software prototypes.
The review of approaches is reported in our Design 2016 paper and is the subject of a journal paper that is currently under review. A series of four software prototypes (StrEmbed-1 to StrEmbed-4), based on lattice theory, were built. These allowed us to export BoMs from CAD systems in a neutral format (a STEP AP 214 file) and use them to generate hypercube lattices. In the first three prototypes, the lattices were visualised using a publicly available tool, LatDraw. In the fourth, a bespoke visualisation method was developed to overcome size limitations in LatDraw. In this way we demonstrated the technical feasibility of embedding BoMs into lattice structures and provide an early interface to restructure BoMs and that hypercube lattices can be used to underpin a new generation of design tools that allow multiple BoMs to be configured and superimposed on a given design description.

3) To create technology demonstrators, built on these prototypes, which can be used to illustrate the potential value and impact of being able to embed design structures in product information.
The case studies were used to demonstrate the functionality of the StrEmbed prototypes and we defined a technology demonstrator, driven from a synthetic engineering process challenge, using data from both robot case studies where we replicated the design structures and representations from the confidential industry case study. However, size limitations in the StrEmbed prototypes (the lattices we generated were vast, even for designs with a relatively small number of parts, and the problem is exponential in that, for a design with n component parts, the resulting lattice has 2n nodes and many more links between nodes) meant that we were unable to implement fully a technology demonstrator. For this reason, and to demonstrate the potential value of being able to embed design structures in product information, we used the definition of the technology demonstrator to spell out a development plan for a software tool that could demonstrate this potential in a journal paper that is currently under review.
4) To evaluate the technology demonstrators with end users and use learning to inform requirements and usage scenarios for the embedding of design structures in product information.
The industry need that informed the definition of the technology demonstrator was established through confidential discussions with three engineers at a partner company: a design specialist, a PLM technical leader whose role was to support the management of BoMs, and a product structure/geometry controller whose focus was on geometry-based BoMs. They identified BoMs as being critical elements of technical communication and highlighted limitations in current IT support for the configuration of BoMs. This led to funding for a separate industry funded project on complex decision making (with Imperial College London) which was reported in a paper that was presented at ICED 17. The industry requirements for embedding of design structures in product information are elaborated in a journal paper that is currently under review.

5) To identify future opportunities for engineering users and within wider academic communities including potential future solution providers such as maths & computing and potential users of embedding such as those who wish to relate data across scales.
We contributed to the development of shape grammar implementations through participation in a DCC'16 workshop that led to a paper proposing a novel lattice-based implementation method for shape grammar interpreters. The project also enabled us to begin building relationships with colleagues in other disciplines (for example, in Theoretical Physics Molecular and Cellular Biology) where we considered the application of embedding to cross-scale modelling of molecular structures. In addition, we are now working with a colleague in Computer Science who is supervising a Masters project exploring new opportunities that the use of lattice representations creates for applications of machine learning in design and manufacturing. Further, we are in early discussions with colleagues in Civil Engineering on the potential value of embedding in the realisation of BIMs (Building Information Models), especially in supporting the handover and life-cycle support of smart buildings.
Exploitation Route We are in discussions with colleagues in other disciplines including omic, medical physics and construction/architectural engineering and developing follow-on research proposals.
Sectors Aerospace, Defence and Marine,Digital/Communication/Information Technologies (including Software),Manufacturing, including Industrial Biotechology

URL https://sites.engineering.leeds.ac.uk/embedding/
 
Description The findings have informed the development of an interface for the Witness simulation package that derives a supply chain from a BoM structure and an associated make-buy scenario. This was funded through an impact acceleration account project with Lanner (suppliers of Witness) and RR, and the interface was applied to a construction case study by colleagues at AMRC, Sheffield.
First Year Of Impact 2019
Sector Aerospace, Defence and Marine,Construction,Digital/Communication/Information Technologies (including Software),Education
Impact Types Economic

 
Description Assuring the quality of design descriptions through the use of design configuration spaces
Amount £1,185,152 (GBP)
Funding ID EP/S016406/1 
Organisation Engineering and Physical Sciences Research Council (EPSRC) 
Sector Public
Country United Kingdom
Start 12/2018 
End 12/2021
 
Description Complex Decision Making Pilot: Unravelling Complex Decisions in the RR Product Development System
Amount £41,119 (GBP)
Organisation Rolls Royce Group Plc 
Sector Private
Country United Kingdom
Start 05/2016 
End 11/2016
 
Description Designing functionally resilient multi-robot systems for smart cities and national infrastructure
Amount £9,708 (GBP)
Funding ID ENCORE Network Plus - URMS 143938 
Organisation University of Sheffield 
Sector Academic/University
Country United Kingdom
Start 11/2018 
End 02/2019
 
Description Desktop/hybrid manufacturing process
Amount £10,000 (GBP)
Funding ID EPSRC Impact Acceleration Account - IAA1718-05-0618 
Organisation Queen's University Belfast 
Sector Academic/University
Country United Kingdom
Start 06/2018 
End 07/2019
 
Description Visualising the impact of early design decisions on engineering supply chains (Impact Acceleration Account)
Amount £33,211 (GBP)
Organisation University of Leeds 
Sector Academic/University
Country United Kingdom
Start 01/2019 
End 08/2019
 
Title Embedding design structures in engineering information. 
Description  
Type Of Material Database/Collection of data 
Year Produced 2017 
Provided To Others? Yes  
 
Description AMRC Modular Housing case study 
Organisation University of Sheffield
Department Advanced Manufacturing Research Centre (AMRC)
Country United Kingdom 
Sector Academic/University 
PI Contribution We provided an interface to the Witness simulation package which takes as input an indented parts list and make-buy scenario and produces supply chain risk profiles. This interface had been developed in the aerospace sector.
Collaborator Contribution AMRC applied the interface to a case study they developed in the modular housing/construction sector
Impact We are in the process of writing a co-authored paper for the 2020 WinterSIM conference.
Start Year 2019
 
Title StrEmbed-1 
Description First software prototype - generates a lattice from a STEP file that can be visualised using LatDraw. 
Type Of Technology Software 
Year Produced 2016 
Open Source License? Yes  
Impact The software has been used to demonstrate the idea of lattices and to undeprin subsequent prototypes. 
URL https://doi.org/10.5281/zenodo.48066
 
Title StrEmbed-2 
Description Second software prototype - generates a lattice (up to 2^7) from a STEP file and embeds it into a full lattice. Uses LatDraw to visualise the lattice and has a command line interface to define new BoMs. 
Type Of Technology Software 
Year Produced 2016 
Open Source License? Yes  
Impact The software has been used to demonstrate the idea of lattices and to undepin subsequent prototypes. 
URL https://doi.org/10.5281/zenodo.238763
 
Title StrEmbed-3 
Description Third software prototype - generates a lattice from a STEP file and supports visualisation of larger lattices. A command line interface allows new BoMs to be embedded into a common lattice. With respect to StrEmbed-2, StrEmbed-3 is significantly faster and can work with larger lattices - 2^15 and above. 
Type Of Technology Software 
Year Produced 2017 
Open Source License? Yes  
Impact The software has been used to demonstrate the idea of lattices and by undergraduate project students on their own case studies 
URL https://doi.org/10.5281/zenodo.232162
 
Title StrEmbed-4 
Description Fourth & final software prototype - generates a lattice from a STEP file and supports visualisation of larger lattices. A graphical interface allows new BoMs to be embedded into a common lattice. 
Type Of Technology Software 
Year Produced 2017 
Open Source License? Yes  
Impact The software has been used to demonstrate the idea of lattices and by undergraduate project students on their own case studies 
URL https://zenodo.org/record/889272
 
Description Contribution to online Masters module 
Form Of Engagement Activity Engagement focused website, blog or social media channel
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Postgraduate students
Results and Impact Learning from the project informed the development and content of an online Masters module that has been piloted with ~12 undergraduate, PhD and Masters students
Year(s) Of Engagement Activity 2017,2018,2019
 
Description Contribution to shape grammar implementation workshop at Design Computation and Cognition conference (2016) 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Other audiences
Results and Impact Presentation of a paper (EXPLORING LATTICE STRUCTURES IN SHAPE GRAMMAR IMPLEMENTATIONS Hau Hing Chau, Alison McKay, Amar K Behera, Alan de Pennington) and discussion at an international conference workshop. This work has the potential to significantly improves the way in which shape grammars are currently implemented and has led to a journal paper that is currently under review.
Year(s) Of Engagement Activity 2016
 
Description Embedding project web site (part of a web site for a wider activity) 
Form Of Engagement Activity Engagement focused website, blog or social media channel
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Public/other audiences
Results and Impact A project web site as part of a wider Configuration Spaces web site
Year(s) Of Engagement Activity 2019
URL https://configuration-spaces.leeds.ac.uk/embedded-structures/
 
Description International project workshop 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Postgraduate students
Results and Impact International workshop on Embedding. Delivered to five masters/PhD students from Leeds and OU by Prof George Stiny, MIT.
Year(s) Of Engagement Activity 2016
 
Description Poster presentation at ICED 2019, Delft, NL 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Professional Practitioners
Results and Impact Poster presentation at ICED 2019. Poster, "Visualising the impact of early design decisions on engineering supply chains" was co-authored with industry partners on the IAA project
Year(s) Of Engagement Activity 2019
 
Description Poster presentation at INCOSE Systems Engineering conference in Leeds, UK 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Industry/Business
Results and Impact Poster presented at INCOSE UK ANNUAL SYSTEMS ENGINEERING CONFERENCE 2019, 19-20 NOV 19, 2019; ROYAL ARMOURIES MUSEUM, LEEDS, UK.
Title: "Organisational Design: The systematic derivation and evaluation of extended enterprise processes from system architectures"
co-authored with industry authors, won Best Poster prize
Year(s) Of Engagement Activity 2019
 
Description Poster presentation: Lattice representations and design descriptions 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Other audiences
Results and Impact Poster presentation to introduce the idea of using lattices to support the reconfiguration on bills of materials. The presentation sparked a collaboration with Prof Saeema Ahmed-Kristensen at Imperial College that resulted in further funding from Rolls-Royce.
Year(s) Of Engagement Activity 2016
 
Description Video on systems design 
Form Of Engagement Activity Engagement focused website, blog or social media channel
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Undergraduate students
Results and Impact A video that couples ideas developed through the Embedding project with activities in a RAEng VP project on engineering supply chain innovation.
Year(s) Of Engagement Activity 2018,2019
URL https://mymedia.leeds.ac.uk/Mediasite/Showcase/default/Presentation/21ed76fe59294898a9d9e92d0225dc21...