The German Research Center for Artificial Intelligence (DFKI) carried out the 3-week tests (from September 4 to 23, 2023) of the PHYSICS platform to assess the goals targeted by the Smart Manufacturing Use Cases:

1. Deployment of substitute service in the Cloud
2. High Confidence Quality Control

(see Blogpost: Industrial Use Cases of FaaS: The basics you need to know for more details)

The evaluation phase was held from 8:00 a.m. to 8:00 p.m., Monday through Saturday. The test consisted of calling the High Confidence Quality Control PHYSICS Cloud deployed function, resulting in approximately 20000 requests per week. This gave the total number of 60177 requests with 59425 successful responses (98.75%), even though PHYSICS is still in beta testing.

Only one major problem with the PHYSICS platform occurred on Monday afternoon during the first week of testing, which, once noticed, was fixed relatively quickly on Tuesday. This can be seen in the weekly results where the successful response rate was 96.47% in the first week, 99.82% in the second week and 99.96% in the third week.

Considering the first Smart Manufacturing Use Case PHYSICS Cloud Deployment is only meant to be used as a backup solution in case when the DFKI’s PHYSICS Edge Deployment goes out of service. Therefore, the availability of the PHYSICS platform (percentage of successful responses) at this stage of development was assessed as satisfactory.

In addition, for every single request the action time[1] and the total time[2] were recorded. Both Smart Manufacturing Use Cases required an average of less than 5 seconds for a Quality Check. The performance of PHYSICS fulfilled this condition which can be seen in Figure 1.

[1] only function execution time

[2] total request round-trip time

It is important to highlight that the total time for the successful request is consistent as shown in the histogram presented in Figure 2. The only outliers are the required cold starts of the functions each day after not being used overnight (not shown in the histogram, since they last longer than 10 seconds).

Because of the pay-as-you-go model of Function as a Service in general, our secondary goal of reducing costs was also achieved. About 20000 requests per week, each lasting less than 2 seconds would cost less than 1€ per month in operational costs[1]. Of course, if we need more requests or even more complex quality checks, operating time will increase, and therefore cost will scale linearly with it. Due to the simplified development and deployment, replacing our functions in the future will be straightforward, further reducing (development) costs.

To sum up, the tests conducted for Smart Manufacturing Use Cases proved the potential of the PHYSICS. Considering that the platform is still at the testing stage, its availability, performance, and chance to reduce development costs were evaluated as satisfactory.

[1] Estimated cost based on comparable FaaS platforms on the market.

The post Assessment of goals targeted by Smart Manufacturing Use Cases appeared first on PHYSICS.

PHYSICS aims at enabling European Content Services Platform to enhance their offerings from baseline computing services to more advanced business and operational models such as the Function as a Service paradigm. To this end, it will bundle the produced technical solutions into 3 major tools.

In addition, and in combination with these three items, PHYSICS is producing a Reusable Artefacts Marketplace Platform (RAMP), in which internal and external entities (developers, researchers etc) are able to contribute fine-grained reusable artefacts and assets (functions, flows, controllers etc), for incorporating added value offerings of different types (controllers, optimizers, flow patterns, abstracted function nodes etc.).

The main aim of the RAMP is bringing contributors and buyers from the cloud computing environment around one central artefacts marketplace, in order to include standalone cross-platform artefacts and assets. Considering this project, various stakeholders among large enterprises, small and medium enterprises, and research partners, show great interest and show promise of a maturation path for the RAMP. Because of the large number of artefacts that we expect to make available and the knowledge of the platform acquired during its development, companies delivering consulting services could provide clear, tangible solutions for the processes of their clients in the form of specific artefacts and integrate these artefacts in larger macro strategies.

The main characteristics of the Reusable Artefacts Marketplace Platform are the following:

• Providing one central place to bring together all actors of the cloud computing environment
• Commonly defined interfaces between similar types of elements, so that different controlling flavours can be included in a plug-and-play manner and with no functional difference in an operational flow
• Easy incorporation through packaging them in the target code/application design framework.
• Easy reference of external sources that needs to be included in the code/description segment.
• Means of evaluation and feedback, as well as rating of the respective element, in order to enable community feedback, indicate element usability and reliability, features that are needed also from the business aspect of the ecosystem such as developer compensation for marketplace participation

As of April 2022, the public website for contributing to the marketplace is up and running and can be found here.

Thanks to the effort of INNOV-ACTS and the consortium, it is already possible to register to the RAMP and contribute with assets. We are looking forward to have you all joining us. You can register here and create your asset here

The marketplace will incorporate reusable solutions across the various fields of PHYSICS, such as cloud patterns implementations, controller/optimizer algorithms, management schedulers, and more.

Each added artefact is defined by Name, Category, Release Date, field of use, type of License, keywords, Owner and any useful link to help/support/show the artefact.

In the next few months, we aim to enlarge the functionalities of the RAMP with new possibilities for interaction, a feedback and evaluation system and video tutorials to support the external contributors that are willing to upload additional assets to the RAMP.

The post PHYSICS Reusable Artefacts Marketplace Platform – An ecosystem of assets appeared first on PHYSICS.

The SmartFactoryKL e.V. technology initiative was established in 2005 in Kaiserslautern and now consists of more than 50 organizations from different fields of research. The initiative aims to bring together industrial and research partners within a common network in order to implement joint Industry 4.0 projects for the factories. The following demonstrator was chosen to integrate PHYSICS related components and increase its usability. With lot-size-one production, it assembles user-customized USB pen drives using different autonomous and interoperable modules, each dedicated for a single step of the production. Modules are self-contained and independent of others. Their capabilities are abstracted by skills, which are then orchestrated by a higher-level software component (Production Flow Control, PFC). The software architecture of the overall infrastructure maintains how the product is produced, by generating a recipe for manufacturing, scheduling the products by their priorities, production time, etc. The software architecture is designed with a service-oriented approach, which enables decoupling and an easier conversion into Function as a Service (FaaS) approach in PHYSICS project.

This demonstrator produces USB sticks customized for user needs. A user can choose the color and the data to be inserted into the stick.

After an order is placed, the transport rail moves the product tray between the modules to complete the production.

After the production is complete, a quality check (QC) using artificial intelligence (AI) is performed in the QC module (Figure 3, bottom right). Current implementation gives possibilities for two improvements:

1. Currently, if a software failure at QC module occurs, no automatic QC is performed. In that case, the operator must check the integrity/quality of the product, manually. Maintenance user should also fix the software problem to enable automatic QC for later.
2. If the output of the AI QC’s certainty level is below the threshold, the QC must be done manually by the operator to cover for edge cases.

Both issues decrease the production rate. Therefore, DFKI introduced two use cases dealing with these two issues. With PHYSICS project, DFKI aims to improve the production rate in case of software failures and/or low QC certainty levels.

For the first issue, a failover scenario is designed. A local Edge infrastructure with PHYSICS components was set up and connected to the central PHYSICS platform available in Amazon Web Services (AWS). This in combination with FaaS enables QC to be performed regardless of its physical location. In case of an issue, the maintenance user will be informed about the problem, and they can continue working on it without time pressure. It is expected that there will be no downtime, as long as an active Internet connection is available.

For the second issue, if a QC certainty level is below a specific threshold, the PHYSICS platform will invoke an additional QC service at AWS to perform more complex computations using more computing resources. It is expected that certainty level improves such that a manual inspection is rarely required. Due to the pay-per-use nature of FaaS, this additional QC is very economical.

Although as a test pilot plant, SmartFactoryKL does not have a requirement on production rates, these use cases enable us to test the FaaS approaches and transfer the knowledge into the industry to solve similar problems in the real factory environments.

The post Industrial Use Cases of FaaS: The basics you need to know appeared first on PHYSICS.