The emerging research challenges for PLM effective digital collaboration

Introduction

This article is an article based on my presentation to the PLMx PI Chicago 2018 event the 5th of November, 2018. It is a short recap of my past years research activities within the PLM domain, with a strong focus on secured collaboration and interoperability.

The industrial context as reflected by the PLMx PI Chicago 2018 context

More and more domains are adopting PLM through digitalisation journey: in addition to traditional domains such as Automotive, Aerospace&Defence, Energy or Building, you can find now Food or Medical Industry. An impressive case concerns Doors and Windows producers (more than 1.36 Decillion configurations to be managed).

PLM can also be used by used for large IT infrastructure such as the one set up by Facebook.

Model Based System Engineering is progressing, with usage of SysML models as digital threads (Whirlpool), proposed transformation of traditional V cycle to Diamond cycle in order to deal with Digital Twins (Boeing) and inclusion of MBSE in digital transformation approaches (Boeing).

Actual experimentations are going far beyond the prospective proposed by the Gartner, with attempts to experiment AI, Big Data Analytics, Blockchain, Gamification or Full digital development in order to create value.

PLM should become agile. It is explored in the presentation "Can Agile Methodology Deliver Successful PLM Projects? " by Paul Empringham (PI), which complete and explore complementary ways referring my article "Agile: from drawing to models with SPEM?". In the Paul's presentation, it appears that adopting agile for some PLM artefacts (e.g. new workflow) seems to be of a particular interest. Some are still subject to Waterfall. So it could justify emerging hybrid engineering processes as explored in my article.

Also, several presentations are covering Engineering and Production processes as well, with coupling between PLM solutions, ERPs and MES, in particular those related to IIoT. Usage of Open Standards for Product Data and IoT are still requested by many enterprises. I consider it as a key for addressing non interoperability and obsolescence risks. DMDII project is pushing the standard for the Industrie 4.0 made in USA.

When some consultant are proposing some approaches for ensuring profitability, some service and solution providers are relying on more integrated solutions (3DExperience by Propel) or multi-tenant PLM backbone (Open PLM).

Actual experimentation of emerging technologies for PLM (AI, Big Data Analytics, Full Digital Development with 3D printing, Blockchain, Gamification) seems to go far beyond the prospective proposed by the Gartner. However, the heterogeneity of the PLM mature is to be considered within the industry. Agile approaches at scale could help some enterprise to recover the delay quicker.

I had the pleasure to be invited of a presentation of the open results of my past years as a researcher in the domain of PLM interoperability and Digital collaboration. It is the purpose of what follows in this article, which collect several topics already addressed in previous article, but making some links with the output of the PLMx PI event, and presenting what could be the new research challenges in this context.

My research activities context

Today, for most of the manufacturing enterprises involved in Aeronautic, Defence and Space, but also in other manufacturing area (Building, Automotive, Transportation, etc), the on going required Digital Transformation can't be achieved without Digital Continuity within the enterprise and within the extended enterprise.

Digital continuity required when dealing with digitalisation

In addition, the delivered products have to be connected to the operational environment in order to provide new maintenance and exploitation services. So if considering that the delivered product is the (complex) system of interest, it is in addition supported by complex engineering environment, complex production environment and complex exploitation&maintenance environment.

Growing complexity of the considered systems of systems

The digital continuity in such a context will be based on digital threads and digital twins, which are consistent multiple representations of the product which can be enriched consistently all along the different phases of the life cycle of the manufactured product.

Digital threads and Digital Twins as enablers

These three points are fully reflected by the business cases presented during the event.

As a consequence, the supporting Information System is itself more and more complex, in particular because of the continuously accelerating peace of change of Information and Communication Technologies. In addition to continuously improved modeling, simulation and monitoring capabilities, enterprises have to deal with emerging and disruptive technologies such as Industrial Internet of Things, Blockchains, Artificial Intelligence, Big Data with associated analytics, Cloud as a service platforms, Agile approaches, serious gaming network, etc. Various presentations of the PLMx PI events are reflecting this, and are even showing that some actual usage (with restricted business cases in terms of scale but very impressive results - cf. Hackrod presentations from which the last picture is extracted) are made with value creation and associated business models. Of course, a lot is still to be made before it becomes common.

But all this confirmed the following:

Emerging disruptive technologies to be continuously integrated

The business context of my research

PLM Interoperability issues for effective collaboration

The issues addressed by last years research activities are those preventing an efficient and effective collaboration:

1. secured collaboration
2. Exchange of Product & Process data
3. Continuous evolution with continuously rising pace of change
4. Growing Complexity of the different systems of systems to be considered

Mastering Information Systems in such a context is key, and can't be achieved at an acceptable price without considering standards for exchange, sharing, linking and long term archiving of Product and Process data. This was the motivation of the development of the Federated Interoperability Framework over the last years.

The Federated Interoperability Framework

The FIF has been a continuous applied research effort performed within the frame of Airbus Group Innovation, for preparing and building operational PLM interoperability. The continuous evolution of the FIF is reflected by the following pictures, capturing the different successive research projects which contributed to the FIF, the considered enablers and open standards, the addressed brakes which had to be unleashed and the version of the associated collaborative experimental platform versions which have been produced.

Open standards are at the intersection of operational, research and standardisation activities which have been continuously performed an iterative way. In particular, it was identified that creation of silos between these activities are going against the objective in terms of continuous interoperability at an acceptable price.

The FIF was applied to establish Digital Continuity for Model Based System Engineering and Virtual Manufacturing as well.

It was done through for:

• Dynamic Manufacturing Networks (IMAGINE project), addressed in the research paper "Digital Factory System for Dynamic Manufacturing Network supporting Networked Collaborative Product Development", N.Figay et Al. Data Knowledge Engineering 105

Assessment of PLM standards and associated implementation with an associated Interoperability test bed (SIP project), addressed in the research paper "Model Based Enterprise Modeling for Testing PLM Interoperability in Dynamic Manufacturing Network", N. Figay et al, Chapter Enterprise Interoperability Volume 213 of the series Lecture Notes in Business Information Processing

• Assessment of IIoT and Virtual Manufacturing standards, addressed in the research paper "Alignment of the PLM federated interoperability framework with IoT and virtual manufacturing". 7th International Conference on Information Society and Technology, ICIST2017

In addition, future business scenarii to be addressed by the FIF have been defined related to connected cyber systems and connected autonomous intelligent systems.

Main research results

Identified issues and associated proposals for innovation, which constitutes research results, are the following:

1.? Multiple representations of the same thing with preservation of the semantic" is required

=> Proposition of the Extended Hypermodel for Interoperability

2. The interoperability is to be considered not application to application, but in a more global and holistic system if willing to identify created value from the usage of open standards such as Application protocols

=> Dynamic Manufacturing Networks which are multi-layered representations of organizations working on a product, with underlying processes, supporting information system and technologies used for their realization.

3.Open standards that can’t be assessed (adherence to business) and for which implementation can’t be tested are useless

=> Interoperability test bed

4. Languages and solutions supporting Multi-scale and multi-granularity representations are required

=> Modeling over UML2/SysML

5. Visual semantic representations are required for mastering emerging complexity

=> Enterprise Navigation System with interactive advanced visualization.

Enterprise navigation systems take advantage of System/Ontology modeling coupled with advanced interactive modeling technics, as illustrated by the following figure.

Conclusion: still a lot to be done

In order to support PLM effective collaboration, a lot was achieved through the development of the FIF

• Identification of enablers which have to be combined for building interoperability
• Identification of brakes to be unleashed which prevent interoperability
• Proposal of an evolutionary framework based on the usage of a combined set of relevant PLM standards
• Proposition of new solutions for addressing identified scientific gaps

However, a lot is still to be done.

• Digital continuity requires through life cycle PLM interoperability, with data architects using the appropriate open standards for shared semantic and processes for ensuring data quality
• Through Lifecycle PLM interoperability is still to be prepared, considering circular economy impact as soon as possible (several lives for a product)
• New processes with short loops back relying on emerging IIoT have to be defined, relying on digital threads and digital twins

Many FIF principles and enablers remain valid, but many brakes remain ignored, creating important risks for industry

• Insufficient investment on open manufacturing standards by the industry, and still too much overlapping and competing open standards
• Still research to be done when having to combine systems of systems with semantic preservation, or producing standards that can be easily aggregated and combined. New ways to deal with MBSE combined with digital threads and twins are required, as illustrated by the presentation of Boeing with the Diamond model.
• Too much closed solutions and technologies preventing effective collaboration by creating silos, with in particular the issue of licensing complexity and costs.
• Agile and traditional approaches are to be reconciliated (cf. my article "Agile: from drawing to modelling with SPEM" or the presentation of Paul mentioned earlier).

This will lead to future research papers and LinkedIn articles contributing to the extension of the FIF.

For knowing more