Upload
khangminh22
View
0
Download
0
Embed Size (px)
Citation preview
Grant Agreement No. 767498
Innovation Action Project H2020-FOF-12-2017
D5.11 Third Parties Experimentation Support
WP5 – Cross-Border Industrial Experiments Version:
Due Date:
Delivery Date:
Type:
Dissemination Level:
Lead partner:
Authors:
Internal reviewers:
1.0
30/09/2020
10/11/2020
Report (R)
PU
Innovalia
All Partners (See List of Contributors below)
Susanne Kührer, Sergio Gusmeroli
Ref. Ares(2020)6563066 - 10/11/2020
Disclaimer This document contains material, which is the copyright of certain MIDIH consortium parties, and may not be reproduced or copied without permission.
The commercial use of any information contained in this document may require a license from the proprietor of that information.
Neither the MIDIH consortium as a whole, nor a certain part of the MIDIH consortium, warrant that the information contained in this document is capable of use, nor that use of the information is free from risk, accepting no liability for loss or damage suffered by any person using this information.
MIDIH has received funding from the European Union’s Horizon 2020 research and innovation programme under Grant Agreement no.
767498.
Date: 10/11/2020 D 5.11 – Support for Third-Party Experimentation Page 3 of 78
Version Control:
Version Date Author Author’s Organisation Changes
0.1 08/09/2020 Daniel Echebarria INNOVALIA ToC first draft
1.0 10/11/2020 Daniel Echebarria
Guillermo Herrera INNOVALIA
Final version and submission
Contributors:
Contributor Partner
Daniel Echebarria INNO
Guillermo Herrera INNO
Deliverable Title: D5.11 – Support for Third-Party Experimentation
Deliverable Number D 5.11
Keywords: Open Calls, Cross-Border, DIH, Mentoring, CPS/IoT, Digital Technologies
Date: 10/11/2020 D 5.11 – Support for Third-Party Experimentation Page 4 of 78
Executive Summary
The present document describes the experiments held in the two Open Calls that took place for
the MIDIH project, describing Open Calls motivation and main objectives, as well as topics in
which they are based on, reporting main actions programmed to ensure the experiments
management, and providing a monitoring plan to evaluate the result and impact of those
actions. Also, it includes some description of the experiments and the reporting their results,
both technical and economic, and elaboration on the material deployed to guarantee an optimal
dissemination.
D5.11 introduces the main goals and objectives of MIDIH Open Calls. Thereafter, it describes the
monitoring plan programmed in order to manage these experiment and ensure their correct
execution. For this purpose, the templates for the two deliverables required for each
experimenter to fill are consequently included in the Annexes. The first one is programmed for
the second progress month of the experiment and corresponds to an earlier phase of the
experiment in which current “As-Is” scenario and expected impacts are described. The second
one is focused on the results reporting, exploitation and dissemination plans at the end of each
experiment, so, it is programmed for its end (Month 6). Additionally, and as indicated, both
deliverable templates can be found at Annex I, at the end of this D5.11 document.
While D1.10: Open Call Package 1 and D1.11 Open Call Package 2 contained the requirements
for the SMEs to apply for these calls for experiments and provided details about the application
and proposal assessment, the main body of the present document regards the experiments
description. Accordingly, this section is divided for the two Open Calls in order to describe each
one of the experiments and report the results achieved. In this description, both the technical
and economic/business impacts derived from experiment deployments are also described.
Finally, dissemination material is briefly reported, as well. Each Open Call owns a dissemination
strategy through different tools, as Flyers for Open Calls advertisement, leaflets, interviews,
video reports and power point presentations. So, in the final sections, all these tools are
identified for each call and reported with some screenshots.
Date: 10/11/2020 D 5.11 – Support for Third-Party Experimentation Page 5 of 78
Table of Contents
Executive Summary ....................................................................................................................... 3
List of Figures ................................................................................................................................ 6
1 Motivation ............................................................................................................................. 7
1.1. First Open Call ............................................................................................................... 8
1.2. Second Open Call .......................................................................................................... 8
2 Methodology for monitoring Open Call Winners ............................................................... 10
2.1. D1. Scenario, Business processes and requirements of the experiment .................... 10
2.2. D2. Experiment Report and lessons learned deliverable ............................................ 11
3 Summary of Experimentation and Results by MIDIH Open Call winners ........................... 12
3.1. Open Call 1 winners .................................................................................................... 12
3.2. Open Call 2 winners .................................................................................................... 22
4 Dissemination Material ....................................................................................................... 35
4.1. First Open Call Dissemination Material ....................................................................... 35
4.2. Second Open Call Dissemination Material .................................................................. 38
List of Acronyms and Abbreviations ............................................................................................ 41
ANNEX I: Deliverables submitted by OC winners ........................................................................ 43
Content of the M2 Deliverable ................................................................................................ 43
Content of the Final (M6) Deliverable ..................................................................................... 59
Date: 10/11/2020 D 5.11 – Support for Third-Party Experimentation Page 6 of 78
List of Figures
Figure 1. Open Call 1 Flyer .......................................................................................................... 35
Figure 2. SWARM experiment (CEESA) results poster for First Open Call .................................. 36
Figure 3. First Open Call winners interviews ............................................................................... 37
Figure 4. First Open Call winners presentation for INART solution (D-CUBE) ............................ 37
Figure 5. Second Open Call webinar ........................................................................................... 38
Figure 6. Open call 2 Flyer ........................................................................................................... 38
Figure 7. V-TREV experiment (Izertis) results poster for Second Open Call ............................... 39
Figure 8. Results reporting videos for Second Open Call ............................................................ 40
Figure 9. Second Open Call winners presentation for V-TREV solution (Izertis) ......................... 40
Date: 10/11/2020 D 5.11 – Support for Third-Party Experimentation Page 7 of 78
1 Motivation
The MIDIH project has had two calls for external participants for the development and
implementation of 32 industrial experiments of six months duration, funded with an amount of
up to 60.000€ each.
This deliverable is covering the results and socio-business impact of the Third-Party MIDIH
experiments, developed by the winners of those two calls for participants closed, respectively,
on 29th June 2018 and August 6th, 2019. More details on the single experiments can be found in
the Annexes to the Period Report, for OC1 in Period 1 and for OC2 in Period 2.
The objective of MIDIH Open Calls lies in the configuration, across Europe, of different Cross-
border experiments that complement the work done at the Lighthouse Experiments as
developers and users of Open Source-based components from the MIDIH Open Platform.
The MIDIH project addresses the Manufacturing Industry, and these Cross-border experiments
leverage the versatility of the MIDIH solutions. The Lighthouse Experiments inside MIDIH are
addressing, respectively, three industrial sectors: Automotive, Cutting tools and Steel Industries.
However, the objective of MIDIH, when organising the two rounds of Open Calls, is to
demonstrate that these type of CPS/IoT solutions mapped along the MIDIH Reference
Architecture can produce impressive results in other Manufacturing sectors as well.
A secondary objective is to make the Competence Centers, which are MIDIH beneficiaries in the
vectors or catalysts for the adoption of these CPS/IoT technologies, by helping the Open Call
winners in the technical side of their development activities; to each OC experiment a Mentor
was assigned. These impacts very positively on one of the main goals of MIDIH: the creation of
a network of European DIHs and CCs which may act not only as awareness-raisers, but also as
accelerators of the Digital Transformation of the European SMEs, by providing access to funding
while acting as honest brokers, connecting the Industrial users of these technologies with the
providers which can develop and fine-tune this CPS/IoT components and tools.
The requirements and evaluation process are thoroughly covered in the deliverables D1.10 –
Open Call Package 1 and D1.11 – Open Call Package 2, and the development of the third-party
experiments took place in the following periods:
- Open Call 1 experiments run from October 2018 to March 2019, and the results were
presented in the Open Demo Day in Milan, on May 20th, 2019, an event hosted by
POLIMI and reported in the Annex to the Periodic Report for Period 1.
- Open Call 2 experiments run from December 2019 or January 2020 to July 2020 or
August 2020, most of which were being slightly delayed due to the COVID-19 pandemic.
The results and demonstrations were held online at the Demo Day on September 18th,
2020 and reported in the Annex to the Periodic Report for Period 2.
The socio-business assessment of the OC winners – see template on the last page of the Content of the Final (M6) Deliverable – has been conducted in WP2 and is reported in Del. 2.6 Socio-Business Assessment and Lessons learned v2.
Date: 10/11/2020 D 5.11 – Support for Third-Party Experimentation Page 8 of 78
1.1. First Open Call
The 1st MIDIH open call targeted the development of data driven applications, by IT SMEs as
technology providers, and experiments in CPS/IoT by Manufacturing SMEs.
The Open Call aims at complementing functionalities around MIDIH reference architecture and
performing experiments in CPS/IoT based on the components provided by the architecture. The
experiments must cover one of the three main Business Scenarios: Smart Factory or Smart
Product or Smart Supply Chain.
The types of activities to perform that qualify for receiving financial support are data driven
experiments in CPS/IoT under some Technological and Experiment topics.
Regarding Technological Topics, expected applicants are IT SMEs as technology providers. This
topics are extensively covered in the deliverable D1.10 – Open Call Package. However, a brief
summary is deployed below:
T1. Modeling and Simulation innovative HPC/Cloud applications for highly personalised Smart Products
T2. Smart Factory Digital Twin models alignment and validation via edge clouds distributed architectures
T3. Advanced applications of AR / VR Technologies for Remote Training / Maintenance Operations (Smart Product and Smart Factory)
T4. Machine Learning and Artificial Intelligence advanced applications in Smart Supply Chains management and optimisation
Referring to Experimentations topics, these ones must cover one of the three main scenarios:
Smart Factory, Smart Product or Smart Supply Chain. Furthermore, the usage of components of
the reference architecture is mandatory and expected applicants are manufacturing SMEs.
Experiment topics are the following ones:
E1. Integrating CPS / IOT subtractive production technologies in Additive Manufacturing experimental facilities
E2. Integrating CPS / IOT factory automation technologies in Robotics experimental facilities
E3. Integrating CPS / IOT discrete manufacturing technologies in Process Industry experimental facilities
E4. Integrating CPS / IOT factory logistics technologies in Warehouse management experimental facilities
1.2. Second Open Call
The 2nd MIDIH open call targeted the development of data driven applications, preferably by IT
SMEs as technology providers, and experiments in CPS/IoT preferably by Manufacturing SMEs.
Date: 10/11/2020 D 5.11 – Support for Third-Party Experimentation Page 9 of 78
The open call aims at complementing functionalities around MIDIH reference architecture and
performing experiments in CPS/IoT based on the components provided by the architecture.
The types of activities to perform that qualify for receiving financial support are data driven
experiments in CPS/IoT under some Technological and Experiment topics. The experiments must
cover one of the three main scenarios: Smart Factory or Smart Product or Smart Supply chain.
Regarding Technological Topics, expected applicants are IT SMEs as technology providers. These
topics are extensively covered in the deliverable D1.11 – Open Call Package 2. However, a brief
summary is deployed below:
T1. Modelling and Simulation innovative HPC/Cloud applications for highly personalised Smart Products, Smart Factory and Smart Supply Chain
T2. Smart Factory and Smart Product Digital Twin models alignment and validation via edge clouds distributed architectures
T3. Advanced applications of AR / VR Technologies for Remote Training / Maintenance Operations (Smart Product and Smart Factory)
T4. Machine Learning and Artificial Intelligence advanced applications in Smart Product, Smart Factory and Smart Supply Chains management and optimisation
Referring to Experimentations topics, these ones must cover one of the three main scenarios:
Smart Factory, Smart Product or Smart Supply Chain. Furthermore, the usage of components of
the reference architecture is mandatory and expected applicants are manufacturing SMEs.
Experiment topics are the following ones:
E1. Integrating Additive manufacturing into legacy production system for experiments CPS / IOT production technologies
E2. Integrating CPS / IOT technologies to bridge factory automation and robotics
E3. Integrating CPS / IOT discrete production technologies in Process Industry
E4. Integrating CPS / IOT factory logistics technologies in internal/external logistic scenario
Date: 10/11/2020 D 5.11 – Support for Third-Party Experimentation Page 10 of 78
2 Methodology for monitoring Open Call Winners
Each Open Call experiments have been programmed for a six months period. With the aim of
having a careful monitoring and management of the experiments deployment, the
documentation of the experiment has been structured in two chapters, or deliverables. Thus,
both chapters have to be completed with a deliverable in order to report progress and achieved
results.
The first deliverable was programmed to be submitted in the second month of each
experiments’ progress, while the second one target was the end of the experiment (month 6).
Both of them can be found at Annex I, at the end of this document. However, a brief description
about their contents is exposed below.
2.1. D1. Scenario, Business processes and requirements of
the experiment
As commented, this deliverable is programmed to be submitted at the end of the second project
month. The aim of this deliverable was to define the main scope of the experiments, by
contextualising them onto two available scenarios: “As-Is” and “To-Be”. “As-Is” refers to the
current situation and context in which the experiment is deployed, whereas “To-Be” scenario is
oriented to the target situation which is expected to reach once the experiment has finished
successfully.
For this purpose, the deliverable contains a first section where its own scope is defined.
Thereafter, an extensive description and the scope of the experiment are exposed. Therefore,
this section begins with a study about the Business process dimension of the area under
experiment. Then “As-Is” and “To-Be” scenarios are described. On the one hand, “As-Is” scenario
is contextualised by an introduction about the business processes involved in. After being
introduced, its current functionalities and ICT Architecture are described and weaknesses and
bottlenecks are identified in order to contextualise the initial environment in which the
experiment is developed. On the other hand, to be scenario is exposed with the modified
business processes expected, its functionality and updated ICT architecture. To conclude with
this section, Business Objectives and KPIs are reported.
In addition, it is deployed an exhaustive Verification and Validation plan of the solution, followed
by the Experiment impact, where are commented the expected results and benefits.
Finally, the report content ends with a final section, in which deliverable conclusions are
discussed.
The deliverable template, as already mentioned above, is included in the annexes at the end of
this document.
Date: 10/11/2020 D 5.11 – Support for Third-Party Experimentation Page 11 of 78
2.2. D2. Experiment Report and lessons learned
deliverable
This deliverable was focused on reporting the experiment results and main conclusions at the
end of the experiments (Month 6).
It contains an Executive Summary which is based on reporting the experiment deployment, as
well as solutions and MIDIH components followed by a Lessons Learnt discussion section with a
few recommendations about its applications.
In addition, it is exposed an Exploitation and Dissemination plan for both medium and long term
and the report ends with a conclusions section about the results and the whole performance of
the experiment.
Furthermore, this deliverable contains three annexes, namely:
• Annex I: Technical assessment questionnaire
• Annex II: Skills needed for the replication of the experiment
• Annex III: Socio Business Assessment
The deliverable template, as abovementioned, is also included in the annexes at the end of this
document.
Date: 10/11/2020 D 5.11 – Support for Third-Party Experimentation Page 12 of 78
3 Summary of Experimentation and Results by MIDIH
Open Call winners
This section contains summary of the experimentation and the results reporting of each
experiment deployed by MIDIH Open Call Winners. For this purpose, each experiment will be
reported in the context of the Open Call to which it belongs.
3.1. Open Call 1 winners
The summary of submitted and selected proposals for the Open Call 1, divided by Technological
topics, which address technologies around the MIDIH architecture, and Experimentation topics,
is available in this section.
Detailed information for each experiment can be found here. Each summary contains a brief
description of the experiment and both technical and economic/business impact report.
PeRsOnalised design by manufacturing LifEcycle feedback loop (PROFILE) – [I-DEAL S.R.L.]
PROFILE experiment aims demonstrating the potential for design and manufacturing by
exploiting Knowage tool and Orion Context Broker to implement the industrial feedback loop
through the whole clothing product life. The experiment is related to Smart Product reference
scenario. Cross border experimentation have been carried out in collaboration with VTT
Technical Research Center for Real Time Stream Data Analytics.
The simulations are based on 3 datasets, acquired by i-Deal:
• Consumer morphology 3D measures and preferences (acquired by ISizeYou app)
• Clothing production measures and fiting trends (acquired from the clothing
manufacturers)
• e-commerce filtering and analysis (developed in Somatch H2020 project)
The experiment enables this loop by providing 3 levels of modeling and simulation:
• Current design VS present consumers (reference success rate of present product)
• New design VS present consumers (simulated success of the new collection)
• New design VS new target consumers (simulated success rate of new collection in a new
market)
The components developed during the MIDIH experiment are fully described in D2. [PROFILE].
OC2 – Experiment Report and Lessons Learnt deliverable.
The obtained results demonstrated that experiment had success and it is possible to offer a
business service to clothing designers in order to reduce the collection failures.
As commented, PROFILE experiment has led to the implementation of a real time delivery of the
3 levels of simulation required to support the design a new clothing collection on the basis of i-
Deal services. The real time release of the results of them has removed a significant bottleneck
to i-Deal service exploitation, which will start from its established customers: Piacenza (active in
Date: 10/11/2020 D 5.11 – Support for Third-Party Experimentation Page 13 of 78
the field of traditional clothing), Sparco (sport technical apparel) and Grassi (worker protection
clothing).
Demonstrated the technical feasibility the additional efforts required will be focuses on the
creation of the proper user interface, in the first period dedicated to the internal operators of i-
deal to better define and test them. In a second time they will be adapted for the eventual direct
use by customer designers.
Robot Optimisation Platform in Real-Time (Robo-OptimAl) – [INFINITE FOUNDRY LDA]
Infinite Foundry developed a cloud platform enabling engineering to raise on demand
computational capacity to execute 3D design and virtual product simulation and testing, through
web browsers. Their innovation is a cloud service that virtualises the access to the main
commercial CAD, CAM, CAE and simulation software through the browser on a pay per hour
pricing, which allows physics-based modelling technology to be used before, during, and after
production of a physical product e.g. a vehicle for complete lifecycle analysis. The way it works
is by storing IoT (internet of sensors) data measured in service in a hybrid configuration: Artificial
Intelligence (AI) in the edge computing + physics in the cloud computing. In this way, this
platform can be applied to run a physics-based model of the industrial robots in a body shop,
simulated in a unique hybrid approach of AI + physics.
Tecnomatix Plant Simulation software provides discrete event simulation and statistical analysis
capabilities to optimize material handling, logistics, machine utilisation, and labour
requirements. In this experiment, Tecnomatix Plant Simulation technology is brought as a
desktop prediction tool, to be used during operation in a cloud environment to quickly answer
to asymmetric changes in market needs that have impact in production. To achieve that MIDIH
components in edge computing extract the movement data from robots using OPC UA protocol
and send it to an IoT data repository, where the instance with Tecnomatix Plant Simulation can
collect the robot movement data it needs.
Although plant and process simulation technology are known for a long time to allow industry
to optimize its production and resources, it has not been widely implemented due to the
hardware and software costs involved. This is due to the fact an industry does not need to
optimize every day its production tasks, so the ROI of plant and process simulation technology
is questionable. This experiment was very innovative as it showed the potential of using the
power of cloud and its pay as you go nature to allow industries to optimize their production
process without needing an on-premises engineering infrastructure.
Dynaback the t-shirt to alleviate back pain (Dynaback) – [MADESIGN LTD]
The Dynaback solution in the MIDIH experiment consists in two main parts: On the one hand,
the further development of the hardware: the Dynaback garment. A prototype of the Dynaback
garment was finalised in a previous European project, and within MIDIH, the garment was made
more comfortable, precise, and closer to manufacturing. On the other hand, the backend
development to collect data from different sensors, to store them on servers, and to retrieve
them for processing and visualisation.
Date: 10/11/2020 D 5.11 – Support for Third-Party Experimentation Page 14 of 78
The Dynaback garment is one facet of the project. It allows to collect data. However, the whole
other objective of the project is to collect this data seamlessly, to store is and to process it.
The Dynaback garment underwent some considerable changes since the beginning of the MIDIH
project. The size of sensors was reduced, the manufacturing process of electronics was brought
closer to industrial production, the garment was designed more comfortable, accurate and
compatible with the work environment. Besides, the architecture has changed since the
beginning of the MIDIH project. While originally the Mobile app was mean to communicate
directly with the WARP 10 platform, now it includes another server between the two.
The Warp10 platform was chosen because it maps well to MIDIH architecture. Many of MIDIH
components are found in Warp 10: Kafka, HBAse, Zeppelin, Warp10 can be accesses from spark,
Warp10 can be accesses from pig, Warp 10 can be accesses from storm…
The growing regulations might explain the need for companies to protect their workforce better.
For instance, in France in the coming couple years, new legislations will force companies to
assess the painfulness of the work of their employees. Dynaback could be a good candidate to
fill in the gap of missing technology. It is not only growing regulations that are pushing the
companies towards increased safety. The realisation that injuries have an unneglectable cost for
the company is encouraging them to make sure their employees are safe. Work-Related
Muskulo Skeletal Disease account for Europe spending 500 billion euros a year.
Bouma’s manufacturing 4.0 (BOUMA) – [SOCIÉTÉ BOUMA]
Currently all outgoing Quality Control inspection operations are performed manually. These
operations include physical manipulation of each of the parts of the finished parts batch. Results
are manually keyed in a word processing software and all conformity document are printed to
be sent to the customer along with the parts and the invoice. Thus, the project includes three
Tasks:
• Task 1: Robotic Arm for Finishing and final QC operations
• Task 2: Automatisation of Quality contractual documentation process and transfer to
customer Extranet
• Task 3: Proof of concept for Data Quality files collection, curation, formating and
transfer to customer under confidentiality and IPR framework
Three process have been implemented inside Project:
• Process 1: Physical automatisation of Outgoing QC controls. Implementation of a robotic
arm for the QC control process.
• Process 2: Dematerialisation of Outgoing QC controls. An automatic control machine
has been commissioned and integrated to outgoing QC workflow Due to the various
formats required by different customers the process is a two-step process: 1. Design
and development of master common format output and 2. Customisation of output for
each major customer
• Process 3: Dematerialised interface with key customer PoC. A basic concept using the
Reference Architecture of the International Data Space was elaborated
Date: 10/11/2020 D 5.11 – Support for Third-Party Experimentation Page 15 of 78
Customer requirements are becoming more important on the quality and quality documents
attached to the delivered parts. The automation of the finishing and control operations, as well
as the digitalisation of the documents represent a fast ROI, and an improvement of the quality.
Manufacturing Industry Data-Driven Digital twin (MID3) – [TECHNOLOGY TRANSFER SYSTEM S.R.L.]
MID3 aimed at implementing a reconfigurable Digital Twin and Simulation avatar of the
Polytechnic of Milan didactic factory fully integrated with the MIDIH-RA. The I4.0Lab factory is
an assembly of several reusable and general-purpose functional components, therefore, the
experiment focused on creating their elementary Digital Twins and interfacing them with the
real counterpart through the MIDIH infrastructure deployed at PoliMi.
Technical impact can be resumed as a combination of the following items:
• Full simulation model of the experimental facility synchronized with the actual factory,
performing constant status monitoring and performance forecasting
• The data-driven simulation of the I4.0Lab deployed on the MIDIH infrastructure
• TTS Simulation Engine equipped with MIDIH Unit Databus Connector
• Real applications of advanced I4.0 technologies create immediately benefits in terms of
perception and understanding from the companies’ side
• Openness and extensibility of the frameworks creates value for the whole ecosystem
The experiment deals with the creation of a digital twin for the I4.0Lab of the Polytechnic of
Milan, meaning that there is not a direct link with a business plan, or an evaluation of costs
needed to bring the solution on the market. Once deployed, the MID3 solution represents by
itself a relevant showcase of the potentialities of the I4.0 Technologies provided by TTS and of
the opportunities deriving from the integration with an Arrowhead Framework. The real use
cases of application of advanced I4.0 technologies creates immediately benefits in terms of
perception and understanding from the companies side, especially SMEs, that have the
opportunity to touch working systems and evaluate the timings to implement real solutions.
Now, the fruitful collaboration with the Polytechnic lays the basis for a continuation of the
activities with the training of the I4.0Lab technical staff to use the simulation synchronized with
the real plant to carry on future research activities.
3D recognition inititing IoT data for industrial training (3D IoT) – [REALMAX OY]
Realmax Finland, within the MIDIH experiment, developed a technical design and system demo
for “3D recognition initiating IoT data for industrial training”. The experiment focused primarily
on the development of “Connected Industrial Worker”. In this way, Realmax is able to utilise
artificial reality, 3D and remote connectivity/mentoring technologies to support the modern
industrial worker. Use case evolves around the maintenance management of the industrial site
and equipment, “Smart Factory”. From the architectural point of view, both “Data-in-Motion”
and “Data-at-Rest” are utilised in this project in a meaningful manner.
The technical solution is about developing a UI version for the HMT-1, voice controlled wearable
smart device, with the following features:
• Enabling HMT-1 voice commands for the application control.
Date: 10/11/2020 D 5.11 – Support for Third-Party Experimentation Page 16 of 78
• Scanning QR codes to recognize industrial facility and/or machines
• Receiving live IoT data from the factory servers and rendering it to HMT-1 display, based
on the QR code scanned. For a while, the data will remain at nominal values, then it
starts going off-track and an alert is generated.
• Launching an external remote mentoring application.
• After remote mentoring session, the new parameters are set.
• Once set up of new parameters is finished, the IoT data returns to nominal values
The few leading Industry 4.0 businesses are already investing vastly in augmented reality and
enhanced MRO (maintenance, repair, and operations) practices, while conservative industries
still observe the action. Augmented reality in maintenance will play a very important role in the
industrial sector. It has a direct influence on performance, productivity and product quality and,
also profit and reputation. Augmented reality allows users to enhance their field of view with
the real-time digital information. However, it is also a valuable solution for many of the
challenges which surround the industrial maintenance and operations.
On-site training of Industrial workers using AR Technology (INART) – [D-CUBE PRIVATE COMPANY]
The INART experiment demonstrates a novel way of Augmented Reality (guided assembly in
manufacturing environments through MIDIH ecosystem. For this purpose, INART Key Points are
the followings:
• Hands free solution with technical drawings and training material available in view at all
times
• Blending of real execution, support and training makes equipment and assembly
operations easier to learn/use
• Gamification increases worker engagement and potentially skills
• Time and cost efficient industrial training
• Assembly recordings can be analysed and allow for sequence optimisation, support
decision making and drive growth
The INART architecture is built on the MIDIH RA standards: Data in Motion (DiM) concept & Data
at Rest (DaR) concept.
INART two major components are the AR Hololens Application and the immersive Framework
and are described extensively in D2. [INART]. OC2 – Experiment Report and Lessons Learnt
deliverable.
In the scope of INART, the content authoring and the analytics modules are custom components
developed and fully integrated in the Immersive Framework. While presenting the virtual model
to the worker, the HoloLens device streams a video of what the worker is seeing to a remote
workstation for the assembly expert (i.e. the supervisor). This exchange of information is
following the Data in Motion (DiM) concept of the MIDIH Architecture. After each assembly
session is finalised, information on the performance of the trainee (worker) is synchronized and
stored to the cloud through the Immersive Framework following the Data at Rest (DaR) concept
of the MIDIH Architecture.
Date: 10/11/2020 D 5.11 – Support for Third-Party Experimentation Page 17 of 78
The AR driven assembly process is drastically changing the way an assembly is presented to the
workers as well as the way workers and supervisors interact in an immersive way in the
manufacturing process. Instead of referring to a manual that might contain hundreds or
thousands of pages, or siting through hours for training courses, new hires and trainees use
headsets that broadcast the pertinent information directly in front of their eyes. Apart from
making the education process easier and more exciting, INART provides to senior executives a
holistic framework to improve and assess workspace industrial training, using cutting-edge
technologies.
D-cube’s goals for INART is to provide Immersive Training Experiences to its existing and/or
potential customers as part of the Immersive Framework Ecosystem. Apart from providing
Artificial Intelligence and Deep Learning Solutions in the context of Industry 4.0 services, vertical
AR industrial training sessions could be developed to support assembling or machine operating
processes.
A Wearable Expert Augmented Reality System (AWEAR) – [RESEARCH STUDIO FORSCHUNGSGESELLSCHAFT MBHY]
The AWEAR platform enables the creation of 3D maps (development of 3D models) of complex
industrial facilities using low-cost mobile sensors. The AWEAR platform is, based on the created
offline models and current sensor data (RGB + IMU), able to accurately localise the worker
(position and orientation) without the need of further expensive sensing infrastructure and
provide assistance in the form of augmented navigation guidance, e.g. for maintenance workers
or remote expert applications.
The challenge of the AWEAR project was to realise the integration of cutting-edge awareness
technology (object tracking, real-world interaction, user localisation in complex environments,
augmented reality) and to provide a unique combination of technology and thus functionality.
The modular technical augmented reality (AR) platform can be adapted to most diverse
industrial assistance applications enabling both, adaptivity and scalability.
The realisation of the AWEAR platform consists of three main components: Generation of a point
cloud 3D model of the relevant environment, Localisation of the user in the recorded
environment and tracking motion of user or users and Displaying AR guidance markers.
Environment Model Generation – SLAM (Simultaneous Localisation and Mapping) is an
approach towards solving the problem of localising a mobile device in a potentially unknown
environment and keeping track of its position. The localisation of the camera in its environment
is performed by extracting 2D features from the RGB image and associating them with 3D
locations in the environment. 2D features are saved to a database, once the environment is
mapped. Visual guidance markers were created in Unity, a cross-platform real-time 3D engine
developed by Unity Technologies. The application creates markers and places them in 3D AR
space following the localised position from the SLAM / odometry information. This application
was deployed to the smart phone and used in the testing and demonstration cases.
The number of embedded and wearable computers increases exponentially, it is being
introduced into more and more types of everyday applications, and a wide span of algorithms
tackle increasingly complex problems. However, the nature of man-machine interaction is still
Date: 10/11/2020 D 5.11 – Support for Third-Party Experimentation Page 18 of 78
largely dominated by decades old paradigms. The latest developments in interactive ICT systems
are approaching truly immersive and intuitive interaction. Especially augmented and mixed
reality technologies offer interaction potentials which will shape the future of interaction design.
Multi-tenant Active Deep Monitoring Platform for AVN Trust Management and Optimisation (MIDAS) – [ENEO TECNOLOGÍA S.L.]
The deployed MIDAS solution consists of a continuous network monitoring tool designed for a
shopfloor environment that will allow to autodetect risks and abnormal situations, based on
manually tagged and categorized events and traffic patterns, and initiate some mitigation or
resolution actions. The solution uses some of the MIDIH components combined with newly
developed modules and has been tested in a demonstration environment. For this purpose,
MIDAS has combined existing technologies that are present in MIDIH architecture with new
Open Source technologies to create the new active monitoring platform.
The objective of this experiment is to develop and test the tools needed to monitor and control
the connections and data flows around a specific connector or component. This will be done by
improving the network monitoring software called Redborder, owned by ENEO Technologies.
The two main components that have been developed are the MIDIH Dashboard and the one
that allow capturing and analysing the network traffic in order to feed the dashboard. Both are
fully described in D2. [MIDAS]. OC2 – Experiment Report and Lessons Learnt deliverable.
The solution developed in this project will have a very interesting business perspective, as
security is one of the major concerns of industries to deploy Smart solutions for Industry 4.0.
The active collaboration between ENEO and the International Data Spaces Association to make
the solution deployed compatible with the IDS Reference Architecture will allow Redborder to
provide security services to those companies that use the IDS connector.
Manufacturing optimisation with ARtificial INtelligence Advanced planning & scheduling (MARINA) – [MANOGEM SA]
Industry requires intelligent systems that turn raw data into useful information for optimisation
of operations. Besides ERP, SCM or MES, Advanced Planning & Scheduling (APS) is the key to KPI
better real-time decisions and improvements as it brings intelligence on top of the ERP. APS is
an area with high potential and APS usage is growing at about 7% CAGR worldwide. The concept
of the MARINA project proposal is to “use AI to fine tune the AI”. I.e. to build an AI and ML layer
on top of an existing multi-solvers and multi-heuristics APS optimisation system in order to
automatically find optimal heuristics parameters and to self-learn from historical data how to
adapt to changing datasets and constraints.
Nowadays, fine tuning of solver parameters is done by hand by an expert in the Objective Screen,
this is a tedious and error prone procedure. Then, without optimized parameters, finding an
optimal schedule can take a very significant time, even hours. In the new process, solver
parameter tuning is done using the implemented ML (Machine Learning) framework that is
connected to FIWARE / MIDIH architecture through the Context Broker and performs an
automatic learning of parameters. Thus, it provides mapping of models stored in FIWARE /
MIDIH and Process real-time manufacturing information captured by IoT devices.
Date: 10/11/2020 D 5.11 – Support for Third-Party Experimentation Page 19 of 78
Having reached the project results of an enhanced ORITAMES APS improves the market position
of ORITAMES. Already a very completive package compared to SOTA, the AI brings better
optimisation, more manufacturing KPIs improvements and thus better ROI for MangoGem’s
customers. In this way, using “AI to optimize the AI” is the key concept and the benefits of this
approach are, on the one hand being able to plan, schedule and explore optimisation scenarios
better than traditional approaches and on the other hand, being able to make implementation
of SCM and APS (Advanced Planning & Scheduling) systems much easier.
ENERgy Saving Platform Experimentation in ALuminium Melting Process Industry (ENERSAL) – [GHI HORNOS INDUSTRIALES SL]
In the ENERSAL context, GHI is working on the design and development of an intelligent and
autonomous energy management monitoring system that will favour the optimisation of the
operation of the tilting rotatory furnaces integrated on the aluminium melting process. For this,
GHI has defined an architecture, both at hardware and software level, for the development of
this solution. The components that have been performed are extensively described in D2.
[ENERSAL]. OC2 – Experiment Report and Lessons Learnt deliverable, namely:
• A high-speed edge-powered furnace control
• Big data simulation-based predictive framework
• Prescriptive Energy Management Module
Under this context, the development of this experiment with these new components has led to
a series of improvements, as a reduction of hardware requirements on the edge and data
storage required on the edge, Cheaper/free opensource solutions and safety on data
transmission through MQTT with SSL increased over VPN tunnel.
Thanks to the incorporation of digital technologies in turnkey plants, high added value services
are being developed, providing GHI a huge improvement on their competitiveness. In addition,
through these services, a new way of marketing and exploitation of GHI products and services is
being discovered, which opens up new access routes to the Market.
Data driven experiment for Energy saving and quality Assessment (DELTA) – [TERA SRL]
The main goal of Delta experiment is realisation of a decision support system for energy
efficiency and zero-defect manufacturing of production line capable of supporting the
production manager with information concerning the energetic state of physical resources of
the production line. The system is able to introduce appropriate measures for the energy
efficiency of the line. The Delta experiment aims to improve the energy efficiency in buildings
and factories, and to monitor the manufacturing process through the IoT gateway. Further, the
product quality assessment is done utilising a distributed control network and implementing
complex sensor fusion, quality prediction and defect detection algorithms.
The DELTA system is composed of various sensors which provide different measurements (e.g.,
energy consumption, power consumption and power production), both from the machineries
and in the plant. These sensors share the collected data with GIoE (Gateway for Internet of
Everything by TERA). Finally, the data are transferred via MQTT to the MIDIH platform for
advanced data processing and visualisation. The realised system can support the production
Date: 10/11/2020 D 5.11 – Support for Third-Party Experimentation Page 20 of 78
manager in the monitoring of the energetic state of physical resources of the production line
and providing an adequate decision support to intervene in the event of loss of efficiency.
Cloud-based solution obtained by the integration of reliable frameworks and tools (such as
MIDIH components) were of great assistance during the development of the experiment system
application for energy efficiency of the production line.
The important experience and the know-how deriving from the R&D activities on the quality
prediction and energy efficiency experiment performed for the DELTA project have been
exploited by the company for other projects. The areas studied during the project cover Industry
4.0 topics applied to the production processes for quality prediction and zero-defect
manufacturing, energy efficiency and predictive maintenance and automation and IT for the
large plants.
Energy Consumption Prediction Based on IoT and CI Techniques (IoTandCI) – [PAUFEX PRESOV, S.R.O.]
The aim of IoTandCI experiment was mainly concentrated on solutions of two problems:
• To design and validate energy consumption prediction models with improved prediction
accuracy.
• To integrate monitored and predicted data into MIDIH Platform with access for
researchers in the field of energy consumption optimisation and also for end users in
companies supplying heat.
The whole experiment in energy consumption prediction was based around the data obtained
through the monitoring system of respective heat sources. The experiment is mainly interested
in daily electrical energy and gas consumption with the possibility of prediction in horizon of
several days (possibly a week).
The experiment was so far concentrated on the analysis of time series characterising the gas
consumption in three different types of non-residential buildings which were used as a starting
point. Thus, It has been increased the number of monitored processes, and the improved
technical infrastructure of the company has allowed an increase in the number of monitored
heating processes by at least 2 times, and by using a combination of advanced signal processing
techniques and CI modelling approaches the prediction accuracy is improved by more than 5%.
The problem of energy consumption prediction in buildings has become a very important issue
with regard to energy management and planning as well as to possible reduction of
environmental impact. The essence of the experiment is designing and verification of the
prediction model of energy consumption at different loads based on data obtained by on-line
monitoring (using IoT devices) of the heat sources for heating different types of facilities (e.g.
schools, public buildings).
Analytics of sensor data of stock taking drones (anAIRlyze) – [DOKS.INNOVA TION GMBH]
The anAIRlyze experiments improved the existing solution inventAIRy® that implements an
indoor drone-based stock taking solution for warehouses, with a data analysis component. The
generated raw data streams of the drone(s) may generally be separated into telemetry relevant
data and observed product/object relevant data (Data-in-Motion). This data was visualised but
Date: 10/11/2020 D 5.11 – Support for Third-Party Experimentation Page 21 of 78
not yet analysed and processed. Those data streams in the experiment have been analysed and
transformed into higher level logistical KPI datasets (Data-in-Rest). The goal was to visualise the
KPIs in a specific dashboard for the end user who is then enabled to take necessary
countermeasures for process optimisation.
During the anAIRlyze experiment conducted by doks. images of flights and sensor values (e.g.
from battery) were collected and saved to rosbag files for later offline replaying and processing
capability. In this way, upon further experimentation these datasets were used to find empty
positions with TensorFlow which gave good results. The playback of other sensor values helped
in the battery prediction which was also tested with good results on additional flights after the
development and battery model creation.
The impact of such a solution is in general less errors been made, so that less search processes
for specific objects are necessary. No more lost items which should be at a specific position but
are somewhere else. With automated optical inspection and additional sensors as attached to
such a solution even qualitative aspects may be affected.
Distributed Warehouse Management Experiment (SWARM) – [CEESA, S.A.]
CEESA’s SWARM experiment proposes to validate a service that enables manufacturers and
retailers to efficiently manage distributed warehouses in the scenario of new added-value
services for e-commerce business model. The SWARM main challenges were 1) the optimisation
of the multiple warehouses management, some of them virtual since the products in transit are
owned by the manufacturer or distributor till the moment the client accepts them, 2) the
integration with enterprise management tools and 3) the efficient interoperability with
management tools of different logistic agents within the supply chain, including logistic agents
and carriers.
The SWARM experiment solution is mainly based in the use of one FIWARE component, the
Orion Context Broker. It relies on the open source MongoDB technology and will receive the
request using the NGSI communication protocol to manage the context information about the
context entities of the experiment, as requested in the Open Standards for Context Information
Management of FIWARE. Therefore, SWARM experiment has been useful to demonstrate that
MIDIH technology can be actually instrumental in delivering very significant business KPI
improvements, and specifically:
• A better connection between resource management tools and the rest of the actors
involved in the delivery have improved the data exchange and reduce the incidences,
with a reduction of delivery incidences around 20%.
• Real time access to physical and virtual stock level of an item, enables a correct
management of the products availability for sales and deliveries, avoiding stock break
by 25%
• Automation of labelling via courier’s web services, and providing a return label to clients,
reduce the labelling times and improve quality of service for clients’ satisfaction, with a
reduction of labelling time by 20%
In order to bring the product to the market, CEESA will have to differentiate from existing
competitors. CEESA strategy is to enable the increased flexibility offered by MIDIH technology
Date: 10/11/2020 D 5.11 – Support for Third-Party Experimentation Page 22 of 78
enablers and architecture, but also to address the need of small and medium enterprise marker,
where the company already holds a strong position.
3.2. Open Call 2 winners
The summary of submitted and selected proposals for the Open Call 1, divided by Technological
topics, which address technologies around the MIDIH architecture, and Experimentation topics,
is available in this section.
Detailed information for each experiment can be found here. Each summary contains a brief
description of the experiment and both technical and economic/business impact report.
DuraTag AR – [MOONSTRUCK]
The main goal of the DuraTag AR experiment is to provide generic Augmented Reality module
as a service for training, repairing and maintenance of the machines and its components
facilitating MIDIH technologies to evaluate the developed tools with real end users and
machines, reaching the appropriate maturity level before market release. The proposed solution
could cover most of the scenarios to address the most challenging of problems across different
industries creating a connection and end-to-end experience between the customer and the
product. Facilitating digitized datasheets for the machines and having different environmental
parameters collected using the existing platform (working hours, temperature, previous
malfunctions) the AR tools could provide comprehensive data visualisation and training.
The developed platform is enabler for the machinery industry that can be used for creation of
digital twin for machines, facilitating unique item level identifiers labelled as QR code for
collecting data about usage, service, repair, etc. The Augmented Reality module is provided to
support presentation of the AR view, which on QR code scan presents 3D models with
augmented data about machine and its machine parts. Developed components are supporting
product passport, mobile application as an enabler for scanning the smart tags and AR content
presentation as well as web application with different user roles for stakeholders, companies
and management of product’s and users’ data. The existing complementary DuraTag platform
building blocks based on Microsoft Azure are completely replaced with Apache open source
technology and extended with new features to support augmented reality. Now Apache Server
it’s being used and there have made experiments with Apache Kafka, Zeppelin and Flink. The
final project architecture of DuraTag platform consist of MIDIH (Apache) components and
additional custom-developed components as modular functional blocks with common API.
AR technology is proving highly flexible and is showing great potential in industry sector. The
main benefits of the proposed tools will be optimisation of costs, lowering the risk work related
injuries, rationalisation of operation costs and maintenance, and extended operational life of
the equipment. As an additional benefit, the platform use will lower the carbon footprint of the
heavy machine industry and will lower the amount of waste by decreasing the manufacturing of
new units, reducing transport for repair or landfilling, by extending operational life, recycling
and reuse of mechanical parts. Thanks to implementation of MIDIH technologies the quality of
services offered has improved, and this should increase number of customers and expected
revenue.
Date: 10/11/2020 D 5.11 – Support for Third-Party Experimentation Page 23 of 78
AdaptabLe TubE pRocessing (ALTER) – [IGIT]
ALTER within the MIDIH experiment enhanced a robotized pipe bending cell with AI-based
capabilities based on the MIDIH Reference Architecture. Two main areas of improvement were
targeted:
• Real-time operation of the robot including proper identification of the raw materials to
be processed (the length and the material of a pipe) as well as finding the required
grasping point on the selected element – the aspect depending on the Data-In-Motion
part of MIDIH.
• Long-term learning for optimal use of resources and reducing waste - the aspect relying
on the Data-At-Rest part of MIDIH.
The original workcell has been developed using the HORSE Software Framework and the
foreseen improvements had to be incorporated without reengineering the whole solution.
Therefore, ALTER had an ambitious goal of integrating two software frameworks targeting
different aspects of Industry 4.0. This has been accomplished by replacing the HORSE
middleware with Apache Kafka belonging to the Apache toolchain of the MIDIH Reference
Architecture. The proposed solution involved both the server-side and edge deployment. The
first focuses on the long-term learning, stock optimisation, and data registration and storage.
The latter is responsible for RGB-D image registration and near real-time processing necessary
to control the robot – raw material identification and localisation of the grasping point. Finally,
additional components, not belonging to neither of the frameworks, had to be integrated to
provide the required functionalities – OpenCV+Python has been used to control the RealSense
RGBD camera and pyTorch has been used to develop the learning-based capabilities of the
system.
Besides, IGIT, as a technology provider will continue the development with the main goal of
increasing the accuracy of grasping and the efficiency of the long term raw materials planning
component. Companies involved within experiment see the commercial benefits of the
developed solution. Once the system is completed it will significantly enhance the portfolio of
products so a 30% increase of new customers is expected.
Advanced Predictive Maintenance (APEMAN) – [MASTA]
The goal of the experiment was to develop and validate in operational settings a Digital Twin
model of a machine running on an edge device, which can be used for predictive maintenance.
The proposed solution is based on the Apache toolchain of the MIDIH Reference Architecture
and uses both the Data-in-Motion and Data-at-Rest processing pipelines. The first one is
responsible for real time acquisition of data and reasoning with a pre-trained deep neural
network model in order to predict potential failures. The second manages storage of annotated
data and periodical re-training of the network with the goal of continuous improvement of the
predictions’ accuracy. The developed system has been deployed and tested in close-to-
operational settings and achieved the expected technical KPIs.
The APEMAN experiment allowed to develop an embedded device able to collect measurements
directly from a machine on a shopfloor via a wide array of sensors (e.g. current, voltage,
frequency, vibration, temperature etc.) and to test them against a digital twin of the machine to
Date: 10/11/2020 D 5.11 – Support for Third-Party Experimentation Page 24 of 78
predict and identify imminent failures of the machine. The data collected during the normal
operation of the system is continuously transmitted to the server side of the system where it is
used to periodically retrain the model and improve its accuracy over time. Such an approach
allows using ISAB as a portable device which can be easily moved from one machine to another,
retrained and used for diagnostics while continuously increasing its performance. This
achievement was possible only because of the duality of the MIDIH architecture supporting both
the real-time Data-in-Motion processing and offline Data-at-Rest analysis.
The developed system allows prediction and detection of machine malfunctions before a critical
failure occurs, which creates obvious benefits for the manufacturing companies. This has been
confirmed by 3 external companies willing to become early adopters and rent out the next
iteration for test trials in their facilities. The technology provider – MASTA – sees the system as
a breakthrough product, which will support the intended transition from a service-based
integrator to a knowledge-based product provider and is willing to continue investing in its
further development.
Model Learning for Cloud-Edge Digital Twins (MoLe) – [NEC Laboratories Europe GmbH]
MoLe aims to facilitate the development and use of digital twins for smart factories, so factory
stakeholders can enjoy all the benefits of digital twin technologies with little effort. In order to
achieve this, it have been connected a data translator, a Scorpio Broker, a Knowledge Infusion
and a FogFlow to create a Digital Twin Execution Framework by combining them. This allows
factory stakeholders to easily implement Digital Twins of machines and processes.
During MoLe, it has been successfully implemented the data translator and made it available as
open source software to the community, being a part of the Scorpio Broker. The translator has
a worst-case performance of 1ms, easily keeping up with all the data from the production line
in real-time. Thanks to Scorpio, it is now possible to access data of the didactic factory in real-
time by consuming it from an online endpoint. In addition, it has been implemented the software
architecture to allow the easy creation of digital twins. This architecture is tested by two digital
twins of two stations in didactic factory: the Front Cover Magazine and the Press Station. These
digital twins are able to monitor all the data from the real twin (each station) and infer their
state and energy consumption. Knowledge Infusion has been used to support the monitoring of
the station’s state with an accuracy of 82.33%, ML had the role for predicting energy usage,
obtaining a MAE of 4% predictions.
Overall, the NGSI-LD Translator is a generic, programmable tool for capturing JSON-based data
and convert it to NGSI-LD. This is a common task during the Knowledge Acquisition phase on any
AI system. The insights gained in the MoLe experiment were shared with various business groups
within NEC, especially the Smart Industry team but also to a new team targeting Smart
Agriculture.
Automatic Vibration Diagnostics and Prescriptive Maintenance Service for Industrial Equipment in Industry (VibroBox) – [Sitel LLC]
Sitel’s goal was to make an integration of VibroBox Service and its improvement with MIDIH that
will help companies (mostly European Industrial Companies and manufacturing SMEs) to
become more competitive, become more eco-friendly and save maintenance and operation
Date: 10/11/2020 D 5.11 – Support for Third-Party Experimentation Page 25 of 78
costs by improving the processes of technical management of industrial machinery. As a result,
all these effects will positively influence manufacturing process planning, supply chain planning,
and equipment life cycle management (incl. maintenance & operation) by means of the MIDIH
platform. During the experiment, parts of the Smart Factory scenario have been executed. For
the Smart Factory scenario the aim is the creation of innovative industrial processes, where data
are acquired and analysed to allow reach the goals.
The project allowed to expand the knowledge of VibroBox developers of open infrastructures
like FIWARE and MIDIH reference architecture. During the MIDIH project VibroBox team have
made only a demonstration of the possibility of the VibroBox Vibration Diagnostics and
Prescriptive Maintenance service working with the FIWARE components.
Integration with FIWARE has created opportunities for the future by giving VibroBox customers
a separate interface (API) for sending processing tasks and requesting / receiving results /
updates on equipment status. To use it, customers will have to add their software and, as in
other cases, enter their equipment into Sitel database and configure processing (this is done
jointly with VibroBox engineers). The Sitel team will promote the idea of integration among its
partners as an option for the most optimal solution from the perspective of universality, scaling
and future support.
Platform for immersive training and activity evaluation based on virtual reality and artificial intelligence (V-TREV) – [IZERTIS]
As a result of V-TREV experiment a virtual reality system for the on-site training of workers in
the industrial sector using VR technology has been designed and developed. Moreover, it has
been evaluated in a real environment with the collaboration of a large chemical company.
This solution has a double purpose. First, to ease the training in technical processes by providing
an extensible and flexible system. Secondly, the real-time evaluation of the trainee performance
to provide insights to their supervisors on which parts of the process seem to be harder for
employees or which ones must be modified.
This project takes advantage of the last VR headset technologies, which allow standalone six-
degrees-of-freedom (6DOF) tracking (which means that the user can move, turn, bend, crouch
and much more) with wireless goggles. The wireless communication feature is used to send data
to an independent platform in which the evaluation process is conducted.
In the technical aspect, V-TREV comprises three clearly separated modules:
• The first module includes a virtual reality application running in a standalone virtual
reality head mounted device (Oculus Quest) with wireless capabilities
• The second module takes advantage of the wireless feature of the former module to
build a platform for the evaluation of the data gathered by the application running in
the first module. This module is executed in a cloud environment, creating a system
which receives and processes data streams from the first module, can immediately
feedback results if required, and stores the data
• The third module consists of a dashboard web application for monitoring, evaluating
and presenting the data stored in the second module
Date: 10/11/2020 D 5.11 – Support for Third-Party Experimentation Page 26 of 78
V-TREV makes use of innovative virtual reality technologies such as the Oculus Quest device and
the last VR toolkits and frameworks as well as open-source software from the MIDIH reference
architecture such as Apache Kafka.
V-TREV is IZERTIS’ proposal is led by virtual reality along with data monitoring: virtual scenarios
for training employees in a safe and comfortable environment, with interactive guided
procedures that the trainees can repeat as much as they desire. These procedures can be
extended, modified, or updated anytime it is required, creating a valuable tool for companies
that embrace a new employee training paradigm. As a result of the dissemination and
exploitation activities performed within the project, 54 entities, have been individually reached
to inform them about V-TREV. Indeed, five of them have shown their commercial interest in the
application of V-TREV results in their training processes.
Machine Learning Application for Motion Capture (MAMOC) – [Dmc-smartsystems GmbH)
MAMOC’s open call project’s goal was to show that an AI application enables automatic
annotations of video material in the sense of « recognising the activities that happen in the video
». To achieve this, it combined object detection and hand pose estimation, enriched with
position data. With this data it trained a neural network to recognize the actions happening. The
solution is particularly intended for the optimisation and digital support of workplaces in manual
production.
MAMOC gained a much better insight in the MIDIH-Architecture and managed also to integrate
its solution with the MIDIH Apache components: Nifi, Kafka, and Cassandra. In this area it was
found that it is very simple and advantageous to use standardized, reusable and easy to
integrate components to support standardisation in the industry, even with open source
products. Thus, it was possible to implement the components in a very short time.
As of now dmc-smartsystems has a prototype setup for showing their expertise in the topic of
modern workplace supervision. It also strengthens their skills and introduced to some new
topics of modern machine learning foundations. They will still focus on product integration to
video analysis toolset from the dmc-Ortim and try to advertise their solution to their customers
in the field of industrial engineering to get a better feedback and even some more business
opportunities.
PG Plant – [PROCESS GENIUS OY]
The PGplant solution provides the Industrial Internet with an attractive facade, which is based
on the advanced Digital Twin concept. To improve the compatibility of the PGplant solution with
standard IoT platforms and frameworks, this experiment has focused on building connectivity
with MIDIH reference framework to extend the architectural capabilities of PGplant. At the same
time, a strong effort has been put into optimising the technical scalability, automating the
deployment process, and completing market launch readiness to enable scalable growth on
global markets. The outcome of this combined effort has resulted in a cloud platform-based
product offered as an easy-to-deploy solution, which integrates with any standard frameworks
via open APIs. Through the implementation of MIDIH pilots, PG Plant solution has been validated
in the Smart factory environment outside the domestic market and has raised larger awareness
about PGplant.
Date: 10/11/2020 D 5.11 – Support for Third-Party Experimentation Page 27 of 78
By building connectivity with MIDIH reference framework, the architectural capabilities of
PGplant has been expanded. Deployed automation solutions has reduced the time developers
spend in server maintenance tasks. The time required to setup a new customer platform into
cloud have been reduced from a 1-day job to less than 2 hours job. Data Stream (DS) service
architecture and DS service API documentation created provides guidance and understanding
of functionality and the integration of data from sites into the PGplant digital twin visualisation
tool/platform. Technical results from the MIDIH project has enabled the platform to be used by
wider audience as the platform is no longer as tied to specific hardware or measurement tools
than it previously was.
The results in Midih has enabled the transfer of business model from the Service base product
to more Software base product, as this enables to expand to the global markets via sub-
contractors and co-operators rather than only by using in-house resources. The first Prokosch
pilot in Stuttgart has already led to sales negotiation in Germany to produce the new PGplant
platform to other companies. The Second pilot platform made for Jacquet in Finland has
increased the interest of their mother company, so the platform is already spreading to other
facilities around Europe.
Improving asset and process management in Offshore gas production using MIDIH industry data analitycs (OffshoreLytics) – [Rolloos Oil & Gas B.V.]
Offshore process monitoring in the drilling process are generating GBs of(useful) data daily, in
this experiment, Rolloos main goal have been to perform a feasibility study, Proof of Concept,
to test the suitability of MIDIH data analytics Platform for resolving monitoring of alarm
situations, to enable (proactive) detection of the new alarm situations. The main result of this
experiment is the D2LabRig Portal, based on MIDIH industry data analytics platform, which
visualises OffshoreLytics service. The Portal/service provides an automatic overview of the
dynamics in the realisation of the rig operation/processes and the main task is to enable an
automatic analysis and reporting of important processes in a rig operation scenario. The service
uses rig sensor data to automatically generate an objective description of the performances
(KPIs) of the drilling process using an advanced hybrid analytics approach. The focus is on
understanding the behaviour of the drilling process, to be able to detect situations (and their
causes) when the process behaves in a suboptimal way.
The usage of MIDIH components was very successful, both from the point of view of the easiness
of usage and the effectiveness of services (D2Twin). Data-driven process monitoring and control
(as implemented in OffsoreLytics service) can be applied in various use cases, specifically for the
domains where the processes are ill defined. Test were performed to validate completeness,
efficiency and the accuracy of created data-driven process model, which fulfilled all experiment
KPIs.
During the experimentation, it become clear that the provided service for automated reporting
as well as the Portal can be used in various domains. For the near future Rolloos has agreed to
work with 1 existing customer that has granted Rolloos the opportunity to validate its current
data pipeline on two existing drilling rig for which all real time data acquisition is supported by
Rolloos. For Rolloos, the main focus areas will be to provide the following services and
integrations:
Date: 10/11/2020 D 5.11 – Support for Third-Party Experimentation Page 28 of 78
• Performance analysis: Post well analysis tool to identify well performance on historical
data
• Reporting service: Add Daily KPI to existing Rolloos drilling reporting service
• Roll@caster: Deliver drilling KPI on Edge computation device
• Rolloos Redzone Services: combine drilling performance KPI output with Deep learning
vision based performance tools
By having an automated process that is consistent across all the drilling fleet, Rolloos will be able
to offer a service that will help identify hidden lost time efficiency’s utilising digital solution and
reducing time spent by manual inputs.
Energy Efficient Milling using Data Driven Models for Cement Manufacturing (CEMTEC-FIWARE) – [Linz center of mechatronics GmbH]
The goal of the CEMTEC-FIWARE project was to implement a FIWARE based condition
monitoring and expert system platform at the CEMTEC cement milling pilot plant in Enns,
Austria. A system has been put in place that automatically records process measurements and
operator set-points during batch-production in a cloud database. Methods and tools for data
analysis and the concept of a physics based and data driven expert system have been developed
and prepared for future deployment. Once enough production data is collected, this expert
system can be calibrated with it, so it will then suggest set-points to the pilot-plant operators,
with the aim of these set-points being optimal for energy efficient cement production.
The condition monitoring and expert system platform is built around FIWARE components
running in pods on a cloud server. Administration of the containers is performed using
Kubernetes. Connection to PLCs and installed sensors is realised via the industry standard OPC-
UA using a secure LTE connection. Production data and set points are stored in a database
(crateDB) using QuantumLeap and visualised with the open source tool Grafana. Under this
context, Key features and benefits of the developed platform are:
• Cloud-centric platform for automatic process monitoring and data analysis
• Production data are automatically streamed into a secure cloud database
• LCM and company partner can easily access and analyse production data
• Plug and play solution via secure LTE connection
• Connection to plant using the industry standard OPC-UA
• No need for the company partner to manage local software
• Extremely cost-efficient in terms of deployment and operation
For LCM, the acquired know-how regarding cloud based FIWARE implementations together with
the developed framework for condition monitoring, process modeling and generation of expert
systems allows to offer similar solutions to LCM customers in different industries. The cloud-
centric implementation allows LCM to directly reuse the implementation work done in CEMTEC-
FIWARE, acting as an enabler for data analytics and AI projects. For company partner CEMTEC,
the pilot plant FIWARE implementation for condition monitoring, data storage and data analytics
will help to efficiently acquire new insights into the cement grinding process, required for
competitive product development.
Date: 10/11/2020 D 5.11 – Support for Third-Party Experimentation Page 29 of 78
Data Space Platform for the management and maintenance of Polyurethane Foam plants (SmartPoly - Smart) – [IPF – Ingeniería del Poliuretano Flexible, S.L.]
Thanks to SmartPoly IPF has improved the functionalities of MyFoamPlant, by designing and
developing a SmartData space to collect and exploit, maintaining the proper data sovereignty,
all the data related to the manufacturing process of the polyurethane plant. The SmartData
platform will allow to correlate the process data with the characteristics and performances of
the final product. With the integration of all the data derived by the manufacturing process and
their correlation to the final product, MyFoamPlant, bases on MIDIH reference architecture, will
be a platform able to manage the manufacturing process in all phases and will provide useful
information to perform predictive maintenance on the process plant.
Through SmartPoly users could access data or reports from anywhere and from any device in
real time, raw and aggregated data is sent securely to the cloud, for more demanding processing
and to enable its visualisation through Internet. Thus, three improvements have been
implemented inside SmartPoly:
• Exploitation of external information to integrate analytics algorithms for the
management of the production process
• Visualisation and control of external and internal data to have a deeper knowledge of
the process
• Exploitation of external information to integrate analytics algorithms for the
predictive maintenance of the machines
Depending on the type of industry, which is directed polyurethane foam, they will have to be
manufactured with very specific characteristics, as each sector requires the foam possesses
specific properties that meet their needs. The essence of SmartPoly is the design and
development of an improved management platform that will make possible the optimisation of
the polyurethane foam manufacturing process and also will improve the knowledge, expertise
and control over the most important parameters of the manufacturing process. Thanks to
SmartPoly results, IPF will reinforce this new line of business, expand its product portfolio for its
customers, and target new customers who want to improve the operations and maintenance of
the equipment in their plants through a profitable investment.
Dual-level Recognition for Environment Aware Mobile Robot (Best Route) – [Beck et al. Services SRL]
Initially, order-related parts were picked using an automatic SAP generated picking list, which
was printed out and handed over to logistics employees. Employees used this list without
prescribed picking sequence with a route chosen at own discretion. This frequently leads to an
inefficient picking path, especially for inexperienced employees, who needed a long training
period. After picking, a scan of the part number via manual barcode scanner confirmed the
executed picking order. However, there was no system to cross check and provide feedback if
the picked parts were the correct ones. In this way, the main goal of the experiment is to assess
how the MIDIH platform can support improvements in the Picking procedure, by speeding up
the process and by reducing the number of picking errors.
The application developed as part of the experiment will be used productively by the customer
in the future, who is evaluating the possibility to use this application in a different area of the
same factory and in other factories. The Best Route architecture followed the MIDIH reference
Date: 10/11/2020 D 5.11 – Support for Third-Party Experimentation Page 30 of 78
one using the Apache toolset. Several MIDIH elements were not needed in the context of the
Best Route experiment. It should be remarked that the Data Analytics layer already existed
before the Best Route experiment, which become one additional data provider (the customer
had a Data Lake in the Azure Cloud already before this experiment). The Best Route application
runs within the infrastructure of Deutz-Fahr, in a separate VLAN. The application will be
connected with the LAN used e.g. by SAP and by office workers. However, the mobile devices
(smart phones or tablets) will not be able to route data to the office LAN.
The Best Route experiment showed that an optimized process– based on MIDIH technology –
helps achieving the following results:
• Reduce the picking time from about 35-40 minutes to about 20-22 minutes
• Reduce the number of errors from 2-3 per day to 2-3 per month
• Improve the flow of information about missing parts.
The Experiment will have a clear positive impact for Beck et al and its customers. The most
immediate exploitation for Beck et al. will be three folds:
• Take advantage of the positive relationship established with SDF to acquire new
projects which benefit of the MIDIH technological stack.
• Start up a marketing campaign to identify and address customers who rely on manual
picking activities and propose similar solutions, using the outcome of the MIDIH
experiment as positive reference.
• Address existing customers with production facilities and evaluate if and how the
MIDIH technology could facilitate their digitalisation initiatives.
Dual-level Recognition for Environment Aware Mobile Robot (DreamBOT) – [TRACTONOMY ROBOTICS BVBA]
DREAMBot will develop a proof-of-concept edge-cloud application based on the MIDIH
reference architecture (RA) to enhance the performance of the Omnit AMR, where both levels
use machine learning approaches of a different complexity. The use case is based on standard
cart handling procedures where an important task element is for the AMR to identify a set of
carts as it approaches the cart parking area in a factory or a warehouse. The AMR operate in a
workspace shared with workers and capable of docking with the carts regardless of their
arrangements on the shopfloor.
In Dreambot, data are processed from onboard vision sensors, in near-real-time to generate a
probability map of carts in the viewport of the robot as it approaches the parking area. This will
be based on lightweight machine learning classifiers implemented and running on the onboard
PC, forming the Data-in-Motion or Edge layer. The AMR uses the information to plan the
approach to the carts. Once in range and while in motion, the AMR pushes 3D point cloud data
to the Data-at-Rest level (cloud) where parallel pose estimation algorithms verify and return
pose estimates to the AMR for capturing the carts. All edge-cloud processing occurs during AMR
motion. The DREAMBot architecture uses technologies selected to be compliant or compatible
with the MIDIH reference architecture. ROS2 is used for end-to-end data streaming middleware,
providing data security while guaranteeing Quality of Service (QoS) during edge-cloud
communication, even in very unreliable wireless environments. The main objective is to perform
a set of experiments to validate this architecture in a realistic demonstration environment
including resilience of the edge-cloud application to network latencies and losses
Date: 10/11/2020 D 5.11 – Support for Third-Party Experimentation Page 31 of 78
The dual-level recognition application, proposed in DREAMBot, is crucial to TRACTONOMY’s
value proposition – delivery of an affordable, environment-aware and easy-to-use robot.
Therefore, implementation of DREAMBot will have not only technical, but also business-related
outcomes – including reaching out to new potential customers and establishing a demonstration
showroom in TRACTONOMY’s facilities. Now are taking place discussions with another major
retailer in Belgium and global e-commerce brand in the US.
Augmented Reality and predictive Maintenance applications for robotic machines in the cable assembly industry (E-ROBOTIC) – [ALLBESMART, LDA]
The main objective of the E-ROBOTIC experiment was to explore and promote the usage of AR
technology on specific robotic machines targeting this industry. This goal was achieved by
bringing an AR-based User Interface for a Robotic Seal Inserter, allowing an operator/technician
to access and visualise machine data/information required for maintenance operations,
training-on-the-job, real-time monitoring of machine KPIs. The AR interface includes a predictive
maintenance scenario to explore further automation possibilities.
The FIWARE ORION Context Broker, with Mongo DB, has been used and proofed suitable for
typical shop floor data exchanges. When requiring access to large files (e.g., machine manuals
with hundreds of high-quality images) or streaming services (e.g., training video tutorials), a
mixed approach is needed for accessing the Data-at-Rest (DaR). For such scenarios, and taking
into consideration scalability reasons, the volume of data that traverses the context broker must
be minimized - specific DaR resources (e.g., documents, images, videos), should be configured
in the MongoDB with minimal information, sufficient for retrieving the resource from other
location(s).
The targeted manufacturing companies will benefit from a decreased machine downtime,
decreasing cost of training or an extended machine lifetime among others. The overall impact
for Allbesmart business is:
• Expected increase in the number of new customers (up to 40%)
• Expected increase of revenue (up to 30%)
• Expected number of new jobs created (up to 20%)
Innovative Industrial 3D Services (II3Ds) – [NOVITECH A.S.]
Innovative Industrial 3D services (II3Ds) is a project in the domain of Smart Factory applying the
MIDIH Architecture to the existing Industrial 3D (I3D) product developed and commercialised by
the Novitech a.s. The I3D product represents digitalisation of processes in industrial company,
combines virtual reality (VR) and augmented reality (AR) to provide guidance for a worker
through the process of training and the process of work execution. The worker is trained in VR
to get familiar with the manufacturing equipment maintenance process. The same set of data is
used in AR to help the worker to execute the work and report the progress in real-time.
Accordingly, II3Ds integrates the FIWARE components (Wirecloud Widget Mashup) to display
and visualise on a dashboard the actual status of the execution of workorders by workers in field
(data-in-motion / DIM). The information collected during the previous execution (data-in-rest
/DIR) of the same workorder are analysed and visualised by FIWARE KNOWAGE. QR codes
detected and identified by Kurento Media Server are used for failsafe identification of the
equipment. The recognized QR codes serve two purposes: to verify the position of the worker
Date: 10/11/2020 D 5.11 – Support for Third-Party Experimentation Page 32 of 78
and to provide for the worker real-time information collected from IoT systems (DIM). In VR
mode, the QR codes are represented by hotspots, behind which the same data are available for
the worker, as in the AR execution mode.
Two components out of the tested three are going to be used in I3D also in the future. SpagoBI
will be provided as a customizable BI tool to analyse data collected by I3D. A potential business
opportunity is also to use the tool to create customized reports for clients. The second
component, which is going to be used is Kurento Media Server. The tested scenario proved its
usability. The next implementation will include also other filters, e.g. face blur to protect
personal data in videos created during work order execution. The development will focus on
KMS also to verify its performance in Remote Assistance as a recording server.
Monitor the energy consumptions (PROOF) – [Energy@Work]
PROOF experimentation was strategically oriented to the experimentation of the new
Energy@Work IoT gateway integrated with original solutions for energy efficiency at the specific
plant for the production of hybrid composite material components for the aeronautical sector
in Brindisi (IT). The idea was to set up a sort of “plug-and-play” sustainability assessment system
and test its feasibility on a real setting. Energy@Work has then integrated the existing
equipment already in use at the plant, mainly related the internal variables of autoclave during
the polymerisation process (temperature and pressure for each of the component), with other
intelligent measurement systems to address a global assessment of energy and gases (nitrogen,
propane) consumption of the whole production process.
The final goal of this application experiment has been to maximize process sustainability and
thus, as a result, to achieve the monitoring of energy and gas consumption with the minimum
impact over generic production settings, providing at the same time a concrete support to the
production managers and operators to optimize the production improving the awareness of
global energy and gas consumptions of the production process and of their relationship with the
costs of the production process.
PROOF is a decision support tool for production of hybrid composite material components for
the aeronautical sector able to optimize the production process and the consumption of the
energy and gas based on MIDIH Reference Architecture. In particular, the system is composed
of sensors which enable the gathering of data related to gas and electricity consumption
collected through the E@W Smart Gateway. Under this context, MIDIH reference architecture
has been exploited for data gathering (Orion Context Broker), advanced stream data
processing (Apache Flink) and visualisation (Knowage).
The results of the PROOF experimentation will be used to improve and complete the company
asset offer, in terms of the set of Cloud-based services. In particular, E@W will show the PROOF
innovative functionalities and results among the existing clientele, and not only, in an effort to
consolidate and expand the E@W business activities, improving the network of relationships,
increasing the number of potential partners and opening a new road for opportunities for
innovation in terms of Cloud-based services for industry 4.0.
Smart Monitoring for Energy Efficiency and Predictive Maintenance – Application to Electric Motors Retrofitting (SUPREEMO) – [NCSR “DEMOKRITOS”]
Date: 10/11/2020 D 5.11 – Support for Third-Party Experimentation Page 33 of 78
The SUPREEMO experiment proposed a data-driven approach using CPS and IoT technologies to
deliver solutions for energy efficiency and predictive maintenance. The experiment involved the
non-intrusive load monitoring (NILM) of industrial Electric Motor-Driven Systems (EMDS). EMDS
are commonly found in pumps, fans, compressors and material handling equipment. For this
purpose, custom sensors were installed on various devices ranging from pumps to centrifugal
units, pressure chambers and boilers. The sensors were designed, printed and tested during
SUPREEMO, and they collected high-frequency electric load data to enable advanced analytics
of the obtained signals.
The experiment created a 60TB database of hard-to-find industrial data, which enabled the
development, training and testing of custom ML and DL algorithms to analyse high frequency
energy load data. It finally presented a suite of robust tools for fault prediction in industrial
equipment operation (with prediction windows ranging from a few days to a few weeks), as well
as a general methodology to propose tangible energy efficiency measures.
In summary, the experiment:
• Developed a robust system for the early detection of signal anomalies, possibly leading
to equipment malfunction, and therefore energy or production losses, and eventually
unexpected device breakdowns.
• Linked this system to the DSS, where the facility personnel had access to realtime
equipment status information, and alerts when signal anomalies occurred, so that
inspections and (if needed) predictive maintenance actions can be planed.
• Applied systematic procedures for the identification of energy efficiency potential and
practical solutions to improve energy efficiency and energy costs, and delivered a set of
solutions based on the constraints and practicalities of the pilot process.
The SUPREEMO solution utilised Cloud – Edge architecture to enable the collection, processing,
transmission and storage of the large volumes of electricity datastreams. The implementation
of this system had several challenges. Using the MIDIH components it was possible to reduce
the cost, increase the integration potential with 3rd party services and the scalability, and
deliver a robust and compact working version in less than 4 months. Among the different
components of the MIDIH architecture, the Kafka broker was used to handle the data exchange
between the fog device and the cloud infrastructure, and MongoDB was used for the databases.
The selection of these components allowed the transfer and storage of the very large volumes
of data, and the real-time visualisation of the collected measurements. SPARK enabled the
creation of the live data streams to feed the analysis modules, and ensured the necessary speed
in data flow for real-time analysis.
The initial business concept was to offer advanced, flexible and affordable Industry 4.0 solutions
for the reduction of energy costs, and the improvement of process efficiency by monitoring the
device health and predicting future malfunctions. The experiment met all three key performance
indicators, which were:
• Energy saving solutions with over 3-5% reduction of total energy costs
• Equipment fault prediction and classification accuracy over 90%
• A user-friendly Decision Support System (DSS) featuring an overall user satisfaction
score above 80%
Date: 10/11/2020 D 5.11 – Support for Third-Party Experimentation Page 34 of 78
And enabled improvements on various business aspects, including energy efficiency and process
sustainability, equipment efficiency and maintainability, and production losses due to
equipment unavailability. All these translate to overall reduction of operational and
maintenance costs if the pilot plant continues to use the tools tested and developed during
SUPREEMO.
Now that the experiment is completed, the developed tools have been installed and tested in
real industrial conditions, and their results have been validated against real observations
(showing for instance that all the predicted signal anomalies resulted to malfunctions). The
collaboration with the industrial pilot ELSAP S.A. was excellent, and their input was extremely
important to understand the real needs of future customers, and how to place the SUPREEMO
solution within an industrial perspective.
The procedures and systems developed under SUPREEMO can be easily applied to other
industrial plants in the future. The DEMOKRITOS team is now working into the next steps to
improve these tools and bring them closer to the market, and significant progress has been
made in seeking additional pilots and funding.
Date: 10/11/2020 D 5.11 – Support for Third-Party Experimentation Page 35 of 78
4 Dissemination Material
Dissemination material for these experiments consists of the advertising material to
acknowledge content and information the target audience can utilise in order to promote them.
Accordingly, it was proposed an extensive material for each Open Call, as flyers, leaflets, videos
and presentations. Whereas in the first Open Call several posters and video interviews were
performed, in the second Open Call the dissemination material produced were focused on
results reporting videos.
4.1. First Open Call Dissemination Material
In order to explain the MIDIH project and the target of the first Open Call it was provided a leaflet
(Figure 1), which is available for the participants to distribute it at international events,
meetings, etc.
Figure 1. Open Call 1 Flyer
In addition, several posters have been produced by Open Call Winners in order to report results
and support dissemination. These posters were used during the Open Call Showcase held in
Milan and at the FIWARE Summit held in Genoa. Figure 2 shows an example of a poster
performed. All the posters have been stored in the MIDIH repository but also can be found at
MIDIH website (https://midih.eu/project.php).
Date: 10/11/2020 D 5.11 – Support for Third-Party Experimentation Page 36 of 78
Figure 2. SWARM experiment (CEESA) results poster for First Open Call
Furthermore, several video presentations (Figure 3) have been produced during the last year. In
these videos can be found some winners interviews of the first Open Call experiments. All the
videos have been stored in the MIDIH repository and the OC1 interviews are also accessible in
at the Link: https://www.youtube.com/channel/UCeggravnGqkVcflBNn36oGA/, which belongs
to the MIDIH Youtube Channel.
This material will be used for the Twitter and LinkedIn Dissemination actions and Video and
Presentation will be uploaded in the MIDIH web site under a dedicated Media section. Here at
Date: 10/11/2020 D 5.11 – Support for Third-Party Experimentation Page 37 of 78
Figure 3, some examples of some video miniatures for interviews to OC1 experiments can be
seen:
Figure 3. First Open Call winners interviews
Finally, a main presentation of the experiment development and results was performed and
shown at the first Demo Day on 20th May 2019 in Milan (Figure 4) and can be found at MIDIH
website (https://midih.eu/project.php).
Figure 4. First Open Call winners presentation for INART solution (D-CUBE)
Date: 10/11/2020 D 5.11 – Support for Third-Party Experimentation Page 38 of 78
4.2. Second Open Call Dissemination Material
For the second Open Call, a new webinar uploaded under the Youtube MIDIH channel was
produced (Figure 5).
Figure 5. Second Open Call webinar
In the same way, MIDIH produced Flyers for the advertisement of the Open Call 2, which were
used in several events (Figure 6).
Figure 6. Open call 2 Flyer
Date: 10/11/2020 D 5.11 – Support for Third-Party Experimentation Page 39 of 78
Besides, some posters were also produced in order to acknowledge the results results, as can
be found in the figure 7 (Izertis experiment).
Figure 7. V-TREV experiment (Izertis) results poster for Second Open Call
In addition, several video presentations (Figure 8) have been produced during last year, in order
to show the results of the Open Call 2 winners' experiments. All the videos have been stored in
the MIDIH repository and the Open Calls experiments’ demos and interviews are also accessible
at the link: https://www.youtube.com/channel/UCeggravnGqkVcflBNn36oGA/, within the
MIDIH Youtube Channel.
Date: 10/11/2020 D 5.11 – Support for Third-Party Experimentation Page 40 of 78
This material has been used for the Twitter and LinkedIn Dissemination actions and the Video
and Presentation have been uploaded in the MIDIH web site under a dedicated Media section.
Figure 8. Results reporting videos for Second Open Call
Furthermore, as well as in the first Open Call, a main presentation of the experiment
development and results has been performed and shown at the Second Demo Day on 18th
September 2020, which took place online and is located at the MIDIH website
(https://midih.eu/project.php).
Figure 9. Second Open Call winners’ presentation for V-TREV experiment (Izertis)
Date: 10/11/2020 D 5.11 – Support for Third-Party Experimentation Page 41 of 78
List of Acronyms and Abbreviations
Acronym Meaning
1OC (or OC1) First open Call
2OC (or OC2) Second Open Call
AI Artificial Intelligence
API Application Programming Interface
APS Advanced Planning and Scheduling
AR Augmented Reality
CAGR Compound Annual Growth Rate
CC Competence Center
CI modelling Continuous Integrating
CPS/IoT Cyber Physical System / Internet of Things
D1 Deliverable 1
D2 Deliverable 2
DaR Data at Rest
DIH Digital Innovation Hub
DiM Data in Motion
DoF Degrees of Liberty
DSS Decision-Support System
DT Digital Twin
EMDS Electric Motor Driven Systems
ERP Enterprise Resource Planning
GIoE Gateway for Internet of Developing
ICT Information and Communication Technology
IDS Industrial Data Space
IPR framework Integrating, Planning and Reporting Framework
IT Information Technology
KPI Key Performance Indicator
LAN/VLAN Local Area Network/Virtual Local Area Network
LTE Long Term Evolution
MAE Measuring Absolute Error
MES Manufacturing Execution System
ML Machine Learning
Date: 10/11/2020 D 5.11 – Support for Third-Party Experimentation Page 42 of 78
MQTT Message Queuing Telemetry Transport
MRO Maintenance, Repair and Operations
NILM Non-Intrusive Load Monitoring
OPC UA Unified Architecture from Object Linking and Embedding for Process Control
PLC Programmable Logic Controller
PoC Proof of Concept
QC Quality Control
QoS Quality of Service
QR Quick Response Barcode
RA Reference Architecture
RGB+IMU Red, Green Blue sensor + Internal Measurement Unit
RoI Return of Investment
SAP Systems, Applications and Products in Data Processing
SCM Supply Chain Management
SLAM Simultaneous Localisation and Mappings
SME Small Medium Enterprise
SSL Secure Sockets Layer
TTS Test to Speech simulation engine
UI User Interface
VPN Virtual Private Network
VR Virtual Reality
Date: 10/11/2020 D1.OC1 – Scenario, Business Processes and Requirements of the
Experiment Page 43 of 78
ANNEX I: Deliverables submitted by OC winners
Content of the M2 Deliverable
OPEN CALL 1 Grant Agreement No. 767498
Innovation Action Project H2020-FOF-12-2017
D1._____.OC1 – Scenario, Business Processes and Requirements of the
Experiment Version:
Due Date:
Delivery Date:
Type:
Dissemination Level:
Lead partner:
Authors:
Internal reviewers:
x
30/11/2018
xx/xx/xx
Report (R)
CO
xxx
_
_
Date: 10/11/2020 D1.OC1 – Scenario, Business Processes and Requirements of the
Experiment Page 44 of 78
Disclaimer This document contains material, which is the copyright of certain MIDIH consortium parties, and may not be reproduced or copied without permission.
The commercial use of any information contained in this document may require a license from the proprietor of that information.
Neither the MIDIH consortium as a whole, nor a certain part of the MIDIH consortium, warrant that the information contained in this document is capable of use, nor that use of the information is free from risk, accepting no liability for loss or damage suffered by any person using this information.
MIDIH has received funding from the European Union’s Horizon 2020 research and innovation programme under Grant Agreement no.
767498.
i
Date: 10/11/2020 D1.OC1 – Scenario, Business Processes and Requirements of the Experiment
Page 45 of 78
Version Control:
Version Date Author Author’s Organisation
Changes
Annexes:
Nº File Name Title
i
Date: 10/11/2020 D1.OC1 – Scenario, Business Processes and Requirements of the Experiment
Page 46 of 78
Contributors:
Contributor Partner
Deliverable Title: D1.___.OC1 – Scenario, Business processes and requirements of the experiment
Deliverable Number D1.___.OC1
Keywords:
i
Date: 10/11/2020 D1.OC1 – Scenario, Business Processes and Requirements of the Experiment
Page 47 of 78
Table of Contents _______________________________________________________________________
Table of Contents47
List of Figures48
List of Tables48
Executive Summary:50
1 Introduction51
1.1 Scope of the deliverable51
2 Description and scope of the experiment52
2.1 Business process dimension of the area under experiment52
2.2 «AS IS» scenario52
2.2.1 Introduction: business processes involved52
2.2.2 Current Functionalities52
2.2.3 ICT Architecture52
2.2.4 Weaknesses and bottlenecks52
2.3 « TO BE » scenario52
2.3.1 Introduction: new / modified business processes53
2.3.2 To be functionalities53
2.3.3 To be ICT architecture53
2.4 Business Objectives53
2.5 Business KPIs54
3 Verification and validation plan of the deployed solution55
4 Experiment impact56
4.1 Expected results and benefits56
5 Conclusions57
List of Acronyms and Abbreviations58
i
Date: 10/11/2020 D1.OC1 – Scenario, Business Processes and Requirements of the Experiment
Page 48 of 78
List of Figures _______________________________________________________________________
i
Date: 10/11/2020 D1.OC1 – Scenario, Business Processes and Requirements of the Experiment
Page 49 of 78
List of Tables _______________________________________________________________________
i
Date: 10/11/2020 D1.OC1 – Scenario, Business Processes and Requirements of the Experiment
Page 50 of 78
Executive Summary: _______________________________________________________________________
i
Date: 10/11/2020 D1.OC1 – Scenario, Business Processes and Requirements of the Experiment
Page 51 of 78
1. Introduction _______________________________________________________________________ 1.1. Scope of the deliverable
i
Date: 10/11/2020 D1.OC1 – Scenario, Business Processes and Requirements of the Experiment
Page 52 of 78
2 Description and scope of the experiment _______________________________________________________________________
2.1. Business process dimension of the area under experiment
Give a description about the business dimensions, context and opportunities of the area
under experiment
2.2. «AS IS» scenario
Prepare a text developing the following concepts:
1. Intoduction: business processes involved
2. Current functionalities
3. ICT Architecture
4. Weaknesses and bottlenecks
2.2.1. Introduction: business processes involved
2.2.2. Current Functionalities
2.2.3. ICT Architecture
2.2.4. Weaknesses and bottlenecks
2.3. « TO BE » scenario
Prepare a text developing the following concepts:
1. Introduction: new/modified business processes
2. To be functionalities
3. To be ICT architecture with a list of IT requirements and those IT requirements
linked with MIDIH architecture and components
i
Date: 10/11/2020 D1.OC1 – Scenario, Business Processes and Requirements of the Experiment
Page 53 of 78
2.3.1. Introduction: new / modified business processes
2.3.2. To be functionalities
2.3.3. To be ICT architecture
2.3.3.1. List of IT requirements
2.3.3.2. IT requirements linked with MIDIH architecture and components
2.4. Business Objectives
Business objectives are defined on the following table:
BUSINESS
OBJECTIVE
List the
Business
objectives
expected for
the Business
Scenario/Use
Case
DESCRIPTION
Give a short description about the
Business Objective
IMPACT
indicate the expected impact
of the business objective for
the company
EFFECT IN VALUE1
Rate from 1 to 5
(being 1 no significant
impact and 5 very
high impact)
Cost
Efficiency
Quality
Flexibility
Innovation
Sustainability
Cost
i
Date: 10/11/2020 D1.OC1 – Scenario, Business Processes and Requirements of the Experiment
Page 54 of 78
Efficiency
Quality
Flexibility
Innovation
Sustainability
Cost
Efficiency
Quality
Flexibility
Innovation
Sustainability
Table 1 – Business Objectives.
2.5. Business KPIs
KPIs are defined on the following table:
ID
BUSINESS Indicators
List the Business objectives
expected for the Business
Scenario/Use Case
DESCRIPTION
Give a detailed description of the indicators
Unit* Current
value
Future expected
value
1
2
3
Table 2 – Business KPIs.
i
Date: 10/11/2020 D1.OC1 – Scenario, Business Processes and Requirements of the Experiment
Page 55 of 78
3 Verification and validation plan of the deployed solution _______________________________________________________________________
GANTT diagram and the scheduled activities proposed:
i
Date: 10/11/2020 D1.OC1 – Scenario, Business Processes and Requirements of the Experiment
Page 56 of 78
4 Experiment impact _______________________________________________________________________
4.1. Expected results and benefits
Comment the expected results and technical and economical/business impact
i
Date: 10/11/2020 D1.OC1 – Scenario, Business Processes and Requirements of the Experiment
Page 57 of 78
5 Conclusions _______________________________________________________________________
i
Date: 10/11/2020 D1.OC1 – Scenario, Business Processes and Requirements of the Experiment
Page 58 of 78
List of Acronyms and Abbreviations _______________________________________________________________________
ID Comments
i
Date: 10/11/2020 D2.OC2 – Experiment Report and Lessons Learned deliverable Page 59 of 78
Content of the Final (M6) Deliverable
OPEN CALL 2
Grant Agreement No. 767498
Innovation Action Project
H2020-FOF-12-2017
D2.[Project Name].OC2 – Experiment Report and Lessons Learned
deliverable
i
Date: 10/11/2020 D2.OC2 – Experiment Report and Lessons Learned deliverable Page 60 of 78
Version:
Due Date:
Delivery Date:
Type:
Dissemination Level:
Lead partner:
Authors:
Internal reviewers:
0.1
31/05/2020
Report (R)
CO
i
Date: 10/11/2020 D2.OC2 – Experiment Report and Lessons Learned deliverable Page 61 of 78
Disclaimer This document contains material, which is the copyright of certain MIDIH consortium parties, and may not be reproduced or copied without permission.
The commercial use of any information contained in this document may require a license from the proprietor of that information.
Neither the MIDIH consortium as a whole, nor a certain part of the MIDIH consortium, warrant that the information contained in this document is capable of use, nor that use of the information is free from risk, accepting no liability for loss or damage suffered by any person using this information.
MIDIH has received funding from the European Union’s Horizon 2020 research and innovation programme under Grant Agreement no.
767498.
i
Date: 10/11/2020 D2.OC2 – Experiment Report and Lessons Learned deliverable Page 62 of 78
Version Control:
Version Date Author Author’s Organisation
Changes
Annexes:
Nº File Name Title
i
Date: 10/11/2020 D2.OC2 – Experiment Report and Lessons Learned deliverable Page 63 of 78
Contributors:
Contributor Partner
Deliverable Title:
Deliverable Number
Keywords: …
i
Date: 10/11/2020 D2.OC2 – Experiment Report and Lessons Learned deliverable Page 65 of 78
Table of Contents _______________________________________________ Table of Contents ........................................................................... Error! Bookmark not defined.
List of Figures .............................................................................................................................. 66
List of Tables ................................................................................... Error! Bookmark not defined.
Executive Summary: ....................................................................... Error! Bookmark not defined.
1 Introduction ........................................................................... Error! Bookmark not defined.
1.1 Scope of the deliverable .............................................................................................. 69
2 Deployed Solution and MIDIH components ........................... Error! Bookmark not defined.
3 Experiment Report ................................................................. Error! Bookmark not defined.
4 Lessons learnt and recommendations ................................... Error! Bookmark not defined.
5 Exploitation plan .................................................................... Error! Bookmark not defined.
6 Dissemination report and future plan .................................... Error! Bookmark not defined.
7 Conclusions ............................................................................ Error! Bookmark not defined.
8 Annex I: Technical assessment questionnaire ....................... Error! Bookmark not defined.
9 Annex II: Skills needed for the replication of the experiment Error! Bookmark not defined.
10 Annex III: Socio Business Assessment ................................ Error! Bookmark not defined.
i
Date: 10/11/2020 D2.OC2 – Experiment Report and Lessons Learned deliverable Page 66 of 78
List of Figures _______________________________________________
i
Date: 10/11/2020 D2.OC2 – Experiment Report and Lessons Learned deliverable Page 67 of 78
List of Tables _______________________________________________
i
Date: 10/11/2020 D2.OC2 – Experiment Report and Lessons Learned deliverable Page 68 of 78
Executive Summary: _______________________________________________
i
Date: 10/11/2020 D2.OC2 – Experiment Report and Lessons Learned deliverable Page 69 of 78
1 Introduction: _______________________________________________
1.1. Scope of the deliverable
i
Date: 10/11/2020 D2.OC2 – Experiment Report and Lessons Learned deliverable Page 70 of 78
2 Deployed Solution and MIDIH components _______________________________________________ Instructions (to be deleted in the final document)
Please provide a list of the deployed MIDIH components and the solution
i
Date: 10/11/2020 D2.OC2 – Experiment Report and Lessons Learned deliverable Page 71 of 78
3 Experiment report _______________________________________________ Explain the achieved KPIs from a technical and business perspective
i
Date: 10/11/2020 D2.OC2 – Experiment Report and Lessons Learned deliverable Page 72 of 78
4 Lessons learnt and Recommendations _______________________________________________ Prepare a text answering the following questions:
1. What worked well during the experimentations? Why?
2. What did not work well during the experimentations? Why?
3. What unexpected issues occurred and how did you fix them?
4. Did any opportunity emerge during the experimentation? Which kind?
5. Were the project goals attained? If not, what changes would help to meet goals in
the future?
6. What are the three most important lessons learned on the experiments?
7. What recommendations would you make to others doing similar projects?
i
Date: 10/11/2020 D2.OC2 – Experiment Report and Lessons Learned deliverable Page 73 of 78
5 Exploitation plan _______________________________________________ Outline the commercial perspective of the solution developed/tested including an estimation of the additional costs needed to bring your service/product into the market.
i
Date: 10/11/2020 D2.OC2 – Experiment Report and Lessons Learned deliverable Page 74 of 78
6 Dissemination report and future plan _______________________________________________ Actions to disseminate the results of the project
i
Date: 10/11/2020 D2.OC2 – Experiment Report and Lessons Learned deliverable Page 75 of 78
7 Conclusions _______________________________________________
i
Date: 10/11/2020 D2.OC2 – Experiment Report and Lessons Learned deliverable Page 76 of 78
Annex I: Technical assessment questionnaire _______________________________________________ Please fill the following table, where you find statements related to the MIDIH architecture and their available software components (FIWARE and Apache). In the scale column you should indicate your agreement. In the comments field you can put additional free text.
Indicator Statement Scale low (1-5) high
Comments
Fulfillment of user requirements
The MIDIH solution contributed to fulfill my experiments objectives.
Learnability It is easy to start and use the MIDIH solution and learn functionalities.
Understandability The MIDIH solution is easy and self-clear to understand and the concepts and terminology are understandable.
User attraction level The MIDIH solution is attractive to the user. I feel satisfied and comfortable when using it.
Efficiency
The time and resources required to achieve the objectives of the solution are reasonable, the solution is fast enough and does not require too many steps.
Openness I had access to all relevant information to get started quickly.
Interoperability and maturity
Capabilities of the MIDIH software components were sufficient to interact with my own or other systems.
Applicability
The required effort to integrate and adopt the MIDIH software components was reasonable for the development of experiment specific solution.
i
Date: 10/11/2020 D2.OC2 – Experiment Report and Lessons Learned deliverable Page 77 of 78
Annex II: Skills needed for the replication of the experiment _______________________________________________ For each competence area of the Table below that is relevant in your experiment, please indicate 2-4 skills, and the associated proficiency level (Foundation – Intermediate – Advanced – Extremely specialised), that in your opinion are necessary for the successful replication, implementation and daily use of your solution.
Please, for each skill, add a short comment to explain why that skill and its associated proficiency level is important.
If you are a Manufacturing SME, please also fill the “Industry 4.0 Readiness Check” questionnaire at www.beincpps.eu. If you are not, please flag the type of Manufacturing SME you refer to.
Industry 4.0
Competence Area
Technology providers Manufacturing SMEs
Sector: ________________________
Lean Management
□ Yes □ No
Process standardisation level
□ Low □ High
Autonomy and Responsibility of operators
□ Low □ High
Manager/
Professional
Technician/
Operator
Manager/
Professional
Technician/
Operator
Strategy and Business Models
Smart Operations and Maintenance
Smart Product-Service Engineering
Smart Supply Chain
IT-OT integration
(IoT platforms, architectures, etc.)
Big data
Other
i
Date: 10/11/2020 D2.OC2 – Experiment Report and Lessons Learned deliverable Page 78 of 78
Annex III: Socio Business Assessment _______________________________________________ MIDIH is doing a Socio-Business Assessment for the whole project, which includes also all experiments. The aim is to demonstrate the impact the project is having in terms of the growth of jobs and revenues due to the EC funding. The overall impact of all public funding has then to report to the European parliament.
Please fill the following table with related KPIs to the socio-business assessment based on which impact the successful execution of your experiment could have on your business. We perfectly understand that it might be difficult to make a clear relation between the result of your experiment and potential new business, but we would appreciate to have at least an indication. In the comments field you can put additional free text.
KPI Relative number Comments (optional)
Expected number of new customers
(in % compared to all customers - within the next three years)
Expected increase of revenue
(in % compared to overall revenue - within the next three years)
Expected number of new jobs created
(in % compared to overall number of employees - within the next three years)