WIMRC - Star Recruit in Automotive Engineering
Lead Research Organisation:
University of Warwick
Department Name: Sch of Engineering
Abstract
SummaryAs part of the creation of Innovative Manufacturing Research Centres (IMRC) the EPSRC Innovative Manufacturing Programme earmarked support for two Star Recruits to take leading positions in Warwick IMRC. The University of Warwick has undertaken an extensive international search which has led to the selection of the first of these. He is an exceptional individual from a North American university who is ready to immigrate to the UK to join Warwick IMRC with a new permanent academic position created specifically for the Star Recruit. The vision for Warwick IMRC for the period October 2006 to September 2011 is to be a centre of research excellence, demonstrating an ability to innovate, influence and perform in manufacturing research internationally at the highest academic level . Its ambition is to use its ability to apply innovative cross-disciplinary research in manufacturing technology and operations, materials, business processes and related management activities to enhance the competitiveness and effectiveness of companies or organisations within specific priority areas of intelligent and eco-friendly vehicles and lean healthcare.The Star Recruit is a recognised international leader in his research field of digital manufacturing (automotive and healthcare). His research expertise is closely aligned with the two strategic priorities of Warwick IMRC.
Organisations
Publications
Wang H
(2009)
A graph-based data structure for assembly dimensional variation control at a preliminary phase of product design
in International Journal of Computer Integrated Manufacturing
Wang H
(2012)
Representation, Generation, and Analysis of Mechanical Assembly Sequences With k-ary Operations
in Journal of Computing and Information Science in Engineering
Wang H
(2009)
Variation propagation modeling and analysis at preliminary design phase of multi-station assembly systems
in Assembly Automation
Tolio T
(2010)
SPECIES-Co-evolution of products, processes and production systems
in CIRP Annals
Shukla N
(2015)
Pathway variation analysis (PVA): Modelling and simulations
in Operations Research for Health Care
Shukla N
(2014)
Improved workflow modelling using role activity diagram-based modelling with application to a radiology service case study.
in Computer methods and programs in biomedicine
Shukla N
(2013)
Genetic-algorithms-based algorithm portfolio for inventory routing problem with stochastic demand
in International Journal of Production Research
Shukla N
(2014)
Modelling variations in hospital service delivery based on real time locating information
in Applied Mathematical Modelling
Shrouti C
(2013)
Root Cause Analysis of Product Service Failure Using Computer Experimentation Technique
in Procedia CIRP
Prakash P
(2013)
Multi-step Functional Process Adjustments to Reduce No-fault-found Product Failures in Service Caused by In-tolerance Faults
in Procedia CIRP
| Description | My research on Digital Lifecycle Management integrates heterogeneous data and processes leading to novel data mining & emerging process mining approaches. These are applied to integrate product, processes, and complex services with system design to create a novel closed-loop lifecycle modelling and synthesis framework, self-resilient production and service systems that are robust to changes and 6-sigma faults. This breaks new ground by establishing a research field on the interface between product design, system design, manufacturing and intangible services. The methodologies have significant impact on a range of areas: automotive, aerospace, consumer goods and healthcare services. Broadly, my research aims to inform Factory-of-the-Future & Healthcare-of-the-Future. This has led to 21 journal papers, implementation of developed methodologies, one patent & 2 patents pending, and one spin-out company which is commercializing the developed methodology. Below I have provided some details. (1) Informing Factory-of-the-Future. Achieving current and future interconnected industrial systems will require tremendous advances in the development of fundamental methods and simulation approaches that can effectively integrate design, manufacturing and service engineering models with heterogeneous measurement data captured during manufacturing and service phases. Developing such methods demands an interdisciplinary focus that can integrate design models (CAD), system engineering models (product/process architecture; performance optimization), manufacturing models (CAM), control engineering models, statistical models (for data with large sample size and small dimensionality), and computer engineering and data mining models (for data with relatively small sample size and large dimensionality) with focus on: (1) modelling and analysis of inter-loops between lifecycle phases: No-Fault-Found (NFF), diagnosis by integrating information available from manufacturing and service phases of lifecycle; (2) modelling and analysis of intra-loops between lifecycle phases: remote service diagnostics/prognostics; and (3) intra-loop: service system modelling and improvement. (2) Informing Hospital-of-the-Future and Future Health Service Delivery Creating and sustaining a robust health care service delivery system for high quality care entails developing fundamental and innovative methodologies that can model and analyze health care services from a systems perspective using interdisciplinary approaches. Our health care research integrates systems engineering approaches with heterogeneous data and health care modeling to inform patient care, whole systems quality improvement and productivity, i.e., integration of data mining and emerging process mining approaches. This allows modeling and analysis of unneeded variation in healthcare systems. For example, our Pathway Variation Analysis (PVA) methodology goes beyond current lean health care tools by addressing unneeded variations in the service system by comprehensively examining the crucial interactions between patient characteristics with clinical decision-making, hospital operational parameters (e.g. bed capacity, availability of decision-makers, hours of operation etc.) along with performance targets overseeing the service environment to assess the effects of these factors on patient care and service delivery organizations. Some examples of our research include modelling service delivery organizations for the management of chronic diseases, for e.g. stroke and diabetes in Emergency Department; and maximize patient services through effective utilization of capital intensive equipment such as MRI and CT. My research supported by the EPSRC STAR award has led to 21 journal papers, implementation of developed methodologies by manufacturing & healthcare industries, 1 patent, 2 patents pending, & a spin-out company which is commercializing the developed methodology. For more information: digiPLM.org |
| Exploitation Route | Led to development of simulation engine - stream-of-variation published in several papers as well as led to several research project - FP7-FoF: Remote Laser Welding System Navigator (PI: Ceglarek; 14 international partners). Example of papers: -Huang et al. 2007a/b-Top2 & Top3 most cited papers in ASME Trans., JMSE among all 681 papers published in 2007-2012; -Wang & Ceglarek 2009-Top-2 most cited paper in Assembly Automation J. among all 414 papers published in 2007-2012. Paper received 2010 Highly Commended Award from publisher -Huang et al. 2009-Top-9 most cited paper among all 391 papers published in IIE Trans. (2008-12) -Phoomboplab & Ceglarek 2008-Top2 most cited paper among all 557 papers published in ASME Trans. (2008-12) -Kong et al.,2008-Top3 most cited paper among all 557 papers published in ASME Trans. (2008-12) (2) Hospital-of-the-Future: My health care research integrates systems engineering approaches with heterogeneous data and health care modelling to inform whole systems quality improvement and productivity, i.e., integration of data mining and emerging process mining approaches. This allows modeling and analysis of unneeded variation in healthcare systems. For example, my Pathway Variation Analysis (PVA) methodology goes beyond current lean health care tools by addressing unneeded variations in the service system by comprehensively examining the crucial interactions between patient characteristics with clinical decision-making, hospital operational parameters along with performance targets overseeing the service environment for service delivery improvement. RESULTS: PVA: (i) has led University Hospitals Coventry & Warwickshire, the second largest acute trust in UK to meet DoH contractual stroke care quality threshold for the first time since threshold mandated in 2008; (ii) was granted one patent, US Patent 20,120,226,508, another pending; (iii) is slated for adoption as best practice for quality improvement in acute stroke for all hospitals with stroke programs in NHS Midlands & East; and (IV) creation of Warwick Analytical Software Ltd spinout to commercialize the PVA. |
| Sectors | Aerospace, Defence and Marine,Healthcare,Manufacturing, including Industrial Biotechology |
| URL | http://www2.warwick.ac.uk/fac/sci/wmg/people/profile/?wmgid=462 |
| Description | The research results developed within the project led to: (I) Ceglarek's EPSRC-sponsored research in product lifecycle engineering was augmented by $300k funding from GE-Healthcare matched to a Dorothy Hodgkin Postgraduate Award in 2008 to explore its transfer into the stroke treatment environment at UHCW. The results are enabling the hospital to achieve Department of Health contractual indicators every quarter, with further benefits in increased rates of thrombolysis (clot busting); numbers of patients scanned within 4-hrs and 24-hrs, increases in timely swallow screens, speech and language assessment and other indicators essential for post-stroke recovery. Through this process the hospital has been able to increase earnings from best practice tariff uplift due to improved services; cut operational costs and avoid financial penalties for not achieving stroke treatment targets. The developed methodology is slated for uptake as innovations and Best Practices for all hospitals with stroke programs across Midlands and East. The process has been patented by Warwick/GE-Healthcare (Ceglarek et.al, Systems and Methods for Health Service Data Analysis, US Patent 20,120,226,508; published on September 6, 2012). |
| First Year Of Impact | 2012 |
| Sector | Healthcare |
| Impact Types | Societal,Economic,Policy & public services |
| Title | Fixture Optimizer for Compliant Assembly |
| Description | The tool offers the possibility to evaluate and optimise product performances for given joint layout and to optimise clamp layout for given joint requirements (i.e., max gap per stitch). Tool's integration capabilities are: (i) optimised product design loop to generate a feasible assembly pro-cess; (ii) optimum locator/clamp layout; (iii) joining process parameters' loop; (iv) work-station optimisation loop with robot simulation and path planning. The Fixture Layout Analyser & Optimiser can be used as interactive/collaborative framework among process and product design engineers. The developed GUI offers interactive tools to facilitate user's data input and visualisation of results. The software is now used by the students (currently 1 UG and 5 PhD) I supervise or co-supervise to develop their own research topics. This has led to a community of users and developers including: University of Naples (ITALY); IIT Kharagpur (INDIA); RWTH-Aachen University (GERMANY); West University (SWEDEN); Universidad Politécnica de Madrid (SPAIN). |
| Type Of Technology | Software |
| Year Produced | 2015 |
| Open Source License? | Yes |
| Impact | Multi-disciplinary Fixture Optimizer for Compliant Assembly (based on the multi-disciplinary variation simulation). The first worldwide technology to fully optimize digital assembly of compliant/deformable parts. The uniqueness of this technology is its capabilities to optimize assembly process for a batch of non-ideal parts (instead of optimization of process for a sample of one ideal/CAD parts in a subassembly). The conducted industrial case studies of the developed methodology of the developed methodology eliminated 25% of engineering changes, reducing equivalent development time from 6 weeks to 1 week for door assembly process design as compared to current industry best practice.. |
| URL | https://www.researchgate.net/project/Fixture-Analyser-Optimiser |
| Title | Part Monitoring and Control |
| Description | The tool has been developed to detect dimensional and geometrical faults of manufactured parts or assemblies. Quality practitioners can perform statistical process control (SPC) utilising cloud of points data. It has the capability: (i) to develop monitoring chart using correlated and multidimensional parameters; (ii) for in-process quality improvement through closed-loop feedback. |
| Type Of Technology | Software |
| Year Produced | 2015 |
| Impact | (i) Enable quality engineers to take decisions on the product quality and potential in-process adjustments? (ii) Facilitate use of non-contact scanners for quality monitoring and control of stamping and assembly process |
| Title | Part Variation modeler |
| Description | The tool generates virtual part or assembly based on CAD data (including GD&T specifications) and measurement data (i.e., cloud of points). It has capability for: (i) Variation Simulation Analysis for deformable sheet-metal parts; (ii) part error characterisation for single part and batch of parts. The Part Variation Modeller implements innovative methods to simulate "within batch" and "batch-to-batch" variation. Tool's integration capabilities are: (i) calculation of part fit-up to satisfy joint performance; (ii) definition and optimisation of locator/clamp layout; (iii) extract significant deformation patterns from high dense cloud of points. |
| Type Of Technology | Software |
| Year Produced | 2015 |
| Impact | This is a stand-alone tool or is also integrarted as part of the 'Fixture Optimizer for Compliant Assembly' (please see impact listed in the 'Fixture Optimizer for Compliant Assembly' ). The tool adds value to the product/process simulation by: (i) interactive/collaborative framework between process and product design engineers; (ii) capability to emulate production error at design stage; (iii) capability to reduce number of engineering changes during installation and commission. |
| Company Name | Warwick Analytics |
| Description | The company's products are based on sophisticated computer algorithms, developed at WMG at the University of Warwick, that allow companies to zero in on product faults and analyse process failures. Warwick Analytics' patented technology ("RCASE" - Root Cause Analysis Solver Engine) is developed from more than a decade of academic research. Our Vision To help manufacturing companies move towards zero defects and to focus on their key competitive advantages of designing and manufacturing superior products. Whilst our primary product development is located at Warwick University, the site where our technology was born, we also have offices in London, the US and Europe. Our team comprises both academic and business specialists to ensure the products we deliver truly meet the needs of our customers. Our software is qualified on all major databases such as SAP, Teradata and Microsoft. |
| Year Established | 2011 |
| Impact | Within our analytics platform there are several innovations: - Distributed architecture suitable for big data and Internet of Things (ioT) data, running in-memory or in-database analytics, compatible with all main database and infrastructure vendors - Proprietary algorithms, including RCASE (see below) - Information Retrieval algorithms meaning that unstructured data can be mined and that data don't have to be cleaned - Automated algorithm selection based on data structure and problem to be solved - Transparent post-processing validation and output THE PLATFORM: Automated Advanced Analytics (A3) The critical differentiators that set our platform apart from existing analytical tools are: - The automation of the data gathering step - The structured selection of algorithms depending on the problem and data available - The platform can be built either in the cloud or on-premise (indeed in a private cloud). A3 Architecture The nine analytical process steps associated with analytics platforms are automated to the maximum possible extent. This makes the platform able to be used by a business user with minimal necessity of involvement of IT or data science in the process. OUR ALGORITHMS (RCASE) Warwick Analytics has commercialised proprietary algorithms originally developed after decades of research. These algorithms are designed to deal with all types of dataset and problem design. They were designed from the ground up to cope with dirty and/or incomplete data, and to extract statistically meaningful signals which are not able to be extracted from other forms of predictive analytics techniques such as regression, neural networks or decision-trees. https://en.wikipedia.org/wiki/RCASE |
| Website | https://warwickanalytics.com/ |