Industrial demonstrators

Unique value propositions demonstrated

The unique value propositions for the industrial automation domain resulting from the application of the SOA paradigm implemented using Web Services, to be illustrated through the industrial demonstrators, can be summarized as follows.
• Vendors' product offerings can be simplified and the cost of their elaboration reduced accordingly, owing to the use of a unifying technology across the board.
• Agility at development time is enhanced and development cost and complexity are reduced, since: application programming is done at the highest possible level of abstraction; re-use of services is facilitated; services can be readily integrated with other services;  services can be readily composed into higher-level services; use of a unified technology allows reducing the number of tools required; the independence between services and their implementation enables incremental and parallel development paths; e.g., simulated hardware can be gradually replaced by real hardware; legacy technology can be encapsulated through service façades, according to a "wrap-and-re-use" rather than a "rip-and-replace" approach.
• Overall system complexity is reduced through the encapsulation of complexity inside each of the constituent building blocks that make up the system. As a result, scalability becomes a built-in feature. By the same token, manageability and maintainability are greatly enhanced.
• Building fault-tolerant systems out of self-reliant components is far less cumbersome than if using a set of tightly inter-related components.
• Run-time agility and adaptability to change are greatly increased as: devices are automatically recognized and identified through plug-and-play connectivity; devices can be readily reconfigured or replaced in case of failures or evolutions; service deployment can be conducted incrementally and scaling can take place over time; communicating entities can share and exchange resources and collaborate with each other through peer-to-peer communication; decision-making can thus be driven down to the source of the information acted upon, which, in turn, enhances responsiveness and efficiency, while improving configurability.
• Owing to the abstraction between service interface and service implementation, services can be materialized on heterogeneous software and hardware platforms. This opens unprecedented interoperability perspectives, in particular of being able to mix and match automation equipment from disparate vendors in the same manufacturing environment.
• The adoption of uniform communication protocols down from the lowest levels of the manufacturing device hierarchy up into the manufacturing enterprise's higher-level business process management systems holds the promise of bridging the gap between the shop floor and the top floor and, hence, of substantially improving the enterprise's overall efficiency.

Use of SOA based on Web Services makes it feasible to develop real smart collaborative mechatronic modules, i.e., intelligent modules based on a combination of mechanics, electronics, control software and communication gear. Such components can be pre-tested and directly delivered to the end user site for installation. In industrial applications, new machines may be installed, manufacturing lines may be duplicated, and more and more applications can be built just by assembling mechatronic modules. This collective functionality distributed across a set of mechatronic modules that can be readily reconfigured to suit evolving application needs replaces the programming of manufacturing sequences and supervisory functions in traditional production systems. Even manufactured pieces of equipment may be integrated into such architectures, as they can automatically detect the various machines and tools required for their production and find their way through the production line.

Industrial automation demonstrators

Three different industrial automation demonstrators have been developed in the SODA project, led by Schneider Electric, Loughborough University and FluidHouse, respectively. Moreover, the Schneider Electric demonstrator has a business process connection engineered by various Spanish partners.

Schneider Electric demonstrator
In collaboration with Geensys, SOGETI and ARC Informatique, Schneider Electric has developed a complete service-oriented industrial application that aims at demonstrating the added value, as perceived by users, brought by service-orientation, both at development time and at run time; indeed, this demonstrator illustrates the entire lifecycle of an industrial automation installation.
This demonstrator platform is a mock-up of a pharmaceutical packaging production line, depicted in the figure above. The platform is made up of four automation lines, each controlled by a Schneider “Premium” PLC communicating through IP over Ethernet. The demonstration focuses on the two SODA-enabled "dose-maker" machines. The role of each dose-maker is to fill small bottles with orange or white granules flowing from a tank. Each bottle is carried on a pallet and when the presence of a pallet is detected, an Ositrack RFID reader allows identifying the pallet and reading the production data associated to the bottle, in particular the type of granules required. A dose-maker comprises a motor that drives an endless screw, causing granules to leave the tank when activated, as well as a trap situated between the tank and the bottle to be filled; when the trap is opened, the granules can flow through. Sensors allow determining when the trap is opened or closed, the tank is empty, the carter opened, etc. In order to fill a bottle, the dose-maker needs to open the trap, run the motor for a certain number of turns, and then close the trap.

The dose-maker machine is a more advanced version of that developed by the SIRENA project; it has been completely redesigned, including the platform's physical wiring. It now incorporates Schneider's Ethernet-based distributed I/O devices (FTBs), to which DPWS and WS-Management have been ported and on which the application services are deployed and executed.
On the software side, the dose-maker incorporates components from Schneider Electric, Geensys, SOGETI and ARC Informatique. The extended ControlBuild tool chain from Geensys enables developing, modelling and simulating the application software running on the dose-maker. The implementation of these applications can be done using familiar automation programming languages of the IEC 61131 family. Applications thus generated are installed and configured on the FTBs using the dynamic deployment mechanisms developed by Schneider Electric. Both the deployment and the run-time monitoring and supervision of the dose-maker application use WS-Management, implemented by SOGETI. One type of run-time monitoring and supervision software is the PcVue SCADA package from ARC Informatique, which has been adapted to the SODA environment.

The use cases illustrated through this demonstrator are the following:
• UC1: The machine builder designs the dose-maker machine as a self-contained mechatronic component, followed by its programming and simulation to check the machine’s global consistency and performance. He/she designs this new machine by composing a motor mechatronic component (provided by a third-party manufacturer, black box component), by re-using a trap mechatronic component taken from a library, white box component), and developing the dosing component, which aggregates services offered by the motor and trap components. Once the design/programming phase is done, the machine builder is able to simulate this dose-maker machine.
• UC2: The machine builder deploys the application into the devices according to the hardware/mechanical requirements and constraints. Since the motor mechatronic component was provided by a third party, there is nothing to deploy into the motor device. The machine builder deploys the trap service in one FTB, and the dosing service into another.
• UC3: Once the OEM has assembled the mechanical and automation parts of the machine, he/she checks the device’s status and cabling through his/her application (running on a PDA) in order to speed up the cabling process and decrease cabling errors.
• UC4: The end user is able to diagnose the dose-maker machine, get its status, and tune its settings.
• UC5: Seamless integration with the SCADA environment for monitoring and supervising the dose-maker machine.

Business process integration
One of the interesting aspects of the SODA ecosystem is the incorporation of industrial information, provided by DPWS services, into higher-level business processes.
Normally, the industrial process and business process worlds remain separate due to, for example, technical or security reasons. The SODA project has created suitable conditions for tackling the integration of these two worlds. Once the industrial process exposes its services through the Internet cloud, the integration of relevant industrial data within the business repository information can be performed in a simple and natural way.
Actually, all possibilities are opened but, in the context at hand, we are not aiming for indiscriminate incorporation of industrial information: activities as regulation, control or supervision of the industrial process from the business processes are not considered here. The idea is to feed the business information repository with specific industrial information that helps system users improve the quality of business services offered to customers. For instance, knowing the rate of failures or the status of a customer's industrial process may assist the after-sales service in order to respond in a more diligent and more effective way. Other business activities, as pre-purchase services or marketing campaigns, as well as product design, may also benefit by improving their quality.

The following demonstration scenario outlines how such integration may be achieved. It is based on the use of DPWS-enabled cameras (developed by I&IMS and UPM) and of the CRM (Customer Relationship Management) system from B-kin.
Two cameras are continuously watching for any unauthorized access to a restricted plant floor area. The alarms originated by the detection of forbidden accesses are sent, via the Internet, to a Data Process Centre (DPC). This DPC is location-independent; in the present case, it is actually located in the USA. Once these data are available for CRM users, some CRM functionalities can be accessed through several front-ends.
Depending on the availability of Discovery proxy and Eventing proxy services, two different scenarios can be proposed.
Scenario A. In this case, both the DPWS Discovery Service and the DPWS Eventing Service have their corresponding proxies that allow extending the reach of the local network beyond the industrial installation. Thus, DPWS devices are directly visible across the Internet cloud. The pertinent industrial data are received by a configurable DPWS client application and fed into the DPC. Alarms coming from the industrial process become so-called CASES entities in the CRM context. CRM users will now be able to consult statistical, historical and trending information of the connected industrial process and use it to improve the quality of the business activities.
Scenario B. In this case, neither the DPWS Discovery Service proxy nor the DPWS Eventing Service proxy is available, so the DPWS network remains purely local. DPWS devices are not directly connected to the Internet cloud. The DPWS client application is installed, with the rest of the industrial devices, in the DPWS local network. Other components that convert this DPWS client into a proxy are also installed and will allow the indirect connection of the industrial process to the Internet. At the DPC side, the corresponding client component will receive the industrial data. From here, we are in the same situation as in scenario A.

Loughborough University demonstrator
Loughborough University's industrial demonstrator is a miniature version of a section of a Jaguar production line. It is made up of electro-mechanical units from Festo, and is often used by Ford to test new control techniques. In a previous version, this test rig was driven by Schneider PLCs, whereas in the current version these PLCs have been replaced by FTB I/O controllers (the same as those used in Schneider Electric’s demonstrator), of which the operations are driven by a PC-based orchestrator. The control programs deployed into the FTBs have been created with Loughborough's own engineering environment, but could as well be developed with the Geensys ControlBuild tools. Unlike Schneider's industrial demonstrator, this demonstrator has the advantage of being transportable.

Loughborough's demonstrator was first exhibited at the MACH 2008 fair, held in April 2008 in Birmingham, UK and was also shown at the ITEA2/ARTEMISIA 2008 Symposium in October, as the common ground between SODA and the IST/SOCRADES project. On the SOCRADES side the Loughborough platform was connected to an SAP system, whilst on the SODA side it was connected to ARC's PcVue SCADA environment.

A video presentation of this demonstrator is provided in the Documents section of this Web site.

FluidHouse demonstrator
The industrial demonstrator developed by Finnish partners FluidHouse, Rihotec and TUT is based on the Silent Power fluid automation monitoring system from FluidHouse, used in the heavy-duty paper & pulp industry. This is a compact, clean and quiet hydraulic power unit that complies with the EU-directive for environmental regulations. Its modular design enables flexible implementations and it is an ideal test bed for testing the benefits of the SODA approach. The prototype platform comprises networked embedded controllers running DPWS that interact with field instruments on one hand and with a PC-based fluid monitoring application on the other. The system also includes a wireless sensor network, engineered by Rihotec, for which technology from Sensicast was selected; the wireless nodes, connected to field instruments, are accessed through a gateway.