In our first article on Saab Aeronautics, we discussed the development work surrounding the company’s multi-role fighter jet JAS 39 Gripen, and the partnership with Boeing to develop the Red Hawk T-7. We also examined what the PLM solution at Saab Aeronautics looks like today.
The central question is, how can the company produce a PLM platform that can better meet the demands of the future? There is no doubt that Saab Aeronautics has high ambitions in their hunt for an ultimate PLM system. The company’s leading PLM architects, Johan Tingström and technical fellow, Erik Herzog, have strong views on what they want with the Open Services for Lifecycle Collaboration (OSLC) format as one possible base for that future platform.
In part two of our story on Saab Aeronautics, I discuss the new suggested architecture known as the Genesis PLM model, and a related solution proposal being tested in the Heliple project.
There are three particular challenges in the current PLM lineup that Tingström and Herzog want to overcome: too many manual import and export activities, and limited traceability if or when proprietary interfaces cannot be used. How can they achieve this?
The answer is the Genesis model, where a key factor is the fact that the products Saab Aeronautics develops tend to live longer than the PLM system versions used to develop them.
Taking a closer look at the list of capabilities Saab Aeronautics aims to derive out of these systems, it becomes clear that the focus is on what the analysts at Gartner and CIMdata have jointly defined as so-called Product Innovation Platforms (PIP).
This means that the platform you build must be characterized by the ability to replace or “jump” between modular units in these systems, without extensive migration activities or adaptations. This function is seen as absolutely critical to success.
Here are the parameters:
- It must be possible to establish life cycle management throughout the system.
- There must be opportunities for tailor-made support in disciplines such as systems management, mechanics, software and electronics.
- A modular platform solution is required where individual domain environments can be replaced without integration problems.
- Saab Aeronautics, as a small supplier, will never be able to influence all suppliers to use the same system. Therefore, there is a need for a product data management platform based on international standards, such as Eurostep’s ShareAspace.
Fully Integrated, But Replaceable Modules
The architecture both Tingström and Herzog desires is a structure that provides support for all engineering and management activities within each discipline. At the same time, these must be fully integrated into a company-wide platform whose individual components can be upgraded independently of the others on the platform.
All processes related to engineering processes, such as system development, software development or mechanical development, must be able to be done in dedicated environments. In this, processes and flows must be able to take place with traceability, storage and divisibility, with relevant accessibility restrictions for everyone regarding project management, configuration responsibility and related engineering work.
“Genesis PLM is primarily applicable when we start with a new project for future products,” said Tingström. “If we look at how to carry out normal engineering work, it is the traditional sequence that applies: requirements management, design, realization, integration, verification and—specifically for what we develop, which are critical systems—declarations that everything is built according to design. This really applies to all engineering domains.”
What are the weak spots here and how does Saab Aeronautics want to improve it?
“We would like our systems engineers to be able to work all day in a systems engineering environment, our mechanical engineers in a mechanical environment and the software developers in an Application Lifecycle Management (ALM) environment. Other things may be added, but we would like to have this approach as a basic principle. On top of that, we would then need to ensure that we can keep together, make available and track product data,” Herzog said.
Looking at Saab Aeronautics' processes, four integration points are of highest importance:
- Check for traceability of requirements.
- Check the configuration structure.
- Check the changes.
- Check the realized configuration.
“Given that the products live so long, a long-term archive will also be needed where we can guarantee in a structured way that product data corresponds to what we have delivered,” Herzog adds.
A Cohesive Singular Environment
“If we would have such an environment instead, and if we should also be able to replace components independently of each other, then we know that we need some kind of archive where we can save data long-term. But at the same time, we want to keep the engineers in integrated environments,” Herzog says.
He claims that the modular approach is of great importance partly to maximize the automation that a supplier can give the company, and partly to ensure that you have all the engineers—who work with a node in the product structures—in the same environment. This also applies to leadership and configuration leaders, to keep the environments together.
With a modular structure, you can also keep everything together in a singular environment.
“What we are trying to create is a federation of equivalent environments, which enables employees to work in the same environment and see the same information,” Erik Herzog explains. “We accept that we duplicate capabilities, so that a software requirements capability may well differ from a systems engineering requirements capability, but the important thing is that we ensure we can track between the requirements.”
“The next part is that because we know that these environments will be born, live and die with different life cycles and different frequencies, we want to be able to replace a software where development takes place the fastest, and perhaps replace it with something else—and to do this without affecting the whole landscape.”
In short, it should be simple and easy to replace what is getting old.
A Modular PLM System on the Wish List
There are more aspects to the Product Innovation Platform theme. Tingström points to requirements related to collaborations in new constellations.
“One collaboration can mean that you work within the framework of one system, while another collaboration requires another system. Here we must have flexibility in this type of new constellation, as well. Therefore, the life cycle of the system itself is one thing, but in addition to this, you have to consider the possibility of being able to meet new partners' systems,” he says.
Modularity in PLM support is high on the wish list; at the same time, traceability between data in different product disciplines must be ensured. How can this be achieved?
“If we look at our own processes, there are four different dimensions we need to think about,” says Herzog. “Demand traceability, product structure, changes and realization. If we look back in time, it was enough to make a link between, for example, two different requirements. Then you were done. But for the future, this will not be enough. We must then also have configuration control of this link, so that it goes to the correct version of another structure.”
For this to be possible, there are a couple of things that must be achieved.
“Integration must be able to be realized at low cost. That is, few integration points combined with standardized mechanisms. Here we believe that the OSLC standard gives us the right capability to achieve this,” Tingström adds.
Tests Together with the Royal Institute of Technology
This is also where Saab has started to do tests in smaller projects, and where the Royal Institute of Technology’s (KTH) Jad El-Khoury comes into the picture. He has helped Saab Aeronautics create an integration between IBM's development environment, Jazz, and Siemens Digital Industries' ALM environment, Polarion.
“That's right. In a small experiment, we tested the technical pieces in four or five different scenarios that we got from Saab and managed to do it out-of-the-box and get the components to talk to each other. It went well; as always, there were some minor initial troubles, but they were quick to fix,” asserts El-Khoury. In short, the requirements and change scenarios were relatively easy to realize.
“On the other hand, we could not make the product structure in this situation because Polarion lacked a concept for this,” Herzog adds. "But what could work, worked. We did not have much money for this project, but it took only about 40 hours to establish the integration, where most of the problem was a web browser setting. "
"Exactly," adds El-Khoury. "It was about the security aspects of getting these machines and browsers to approve each other. Much in these contexts has to do with the security aspects in the communication between two different systems. In this case, we worked with IBM's cloud system for the Nordics on the one hand and KTH's own system on the other. In both cases, there are a lot of security barriers built in, which can be a little tricky to get around.”
The Way Forward and the OSLC Standard
The good news is that there was the possibility of linking two different systems out-of-the-box due to the fact that they both follow the OSLC standard, at least to a point. This is a good thing, and provides the core of how they intend to move forward.
“Typically, in a software environment—for example, Siemens’ ALM solution, Polarion—we could have a number of requirements for a product or product element,” Herzog explains. “The important thing is that those on the software engineering side could then track them up to the related requirements on the system engineering side and understand where they come from in a system environment. We need to be able to establish a product structure between the tool boundaries. Plus, change matters must also be able to flow back and forth, as well as the fact that when something is created, we must be able to keep track of what we have realized.”
Interaction Via Standards
Looking ahead to the next stage of the journey, what comes next is to weigh in on the role and contact points of the supply chain.
“Exactly,” says Tingström. “For example, we have Eurostep with its ShareAspace solution. In this area of collaboration, we generally see that we want to digitize ourselves more. Being able to take in granular data, regardless of source format, is very important to us.”
Of course, he adds, the large suppliers of aircraft can still "dictate" to their suppliers in the chain which tools they want them to use.
“But we will never be in such a position,” Tingström says. “We want our interaction to take place in standard format, such as STEP. If we then look at our collaboration environment, which could be ShareAspace, then we can do our product development in our internal environment. This means that for a component such as an Air Data Computer, we can produce stakeholder requirements that we send out to a number of different suppliers.”
“Of course, we must ensure that we track them to our overriding requirements, while we can move the requirements specific to a collaboration platform and make it available to our suppliers. They could then respond with their own requirements specification, design, verification and declaration. At the same time, in this collaborative environment, we could create traceability against the demands we have sent out, but also towards our own environment,” Tingström adds.
The Heliple Project: A Way Into The Future?
This is the use case Saab Aeronautics wants to make within the framework of the Heliple project—partly to ensure whether it is financially justifiable to ask a provider to invest in OSLC interfaces, and partly answer the question of what to expect.
“Will this work in a large-scale environment,” Erik Herzog wonders. “These are two of the goals of the Heliple project.”
“If we look at the picture of the development model above, we have things like that we will have to make change matters from the overall system level down to the more change-driven development,” Herzog explained. “And it will go from the upper system level down to software, where we will find smaller cases. Here, in accordance with the previous reasoning, this link will be an OSLC link (marked by the red arrow in the image above).”
The 18-month Heliple project now intends to investigate the possibilities of supporting knowledge about and using OSLC. The participants are Saab Aeronautics, The Royal Institute of Technology and Eurostep.
Tingström summarizes the scope of the Heliple project in brief:
- The pattern of the Genesis architecture, as described above, should be promoted.
- Gain experience creating OSLC interfaces.
- Improve tool generation in the OSLC interface.
- Demonstrate the power of OSLC.
As a bonus, the result of this project could be that it will be easier for smaller players in the system market to enter, he says. "Today it can be difficult for smaller companies to, despite innovative solutions, enter larger companies or organizations, as these often have high demands on their suppliers in the form of long-term stability. Of course, this is something we must also include in our evaluations. The companies we choose to work with need to be large enough to be able to continue to develop their system and have enough money to not disappear.”
Challenges for Every Actor in Aerospace & Defense
In summary, this is what Saab Aeronautics want to achieve, and the whole thing is illustrative of the complexity encountered when it comes to developing fighter aircraft.
It should also be said that Saab Aeronautics is far from alone in these challenges.
In a recent article, engineering.com took a closer look at the challenges facing one of the world's leading players in the aerospace & defense industry segment and in the development of fighter aircraft, Lockheed Martin, and the development of the F-35 Lightning II or Joint Strike Fighter (JSF).
This project is one of Lockheed Martin's major investments. Over time—the project started in 2006—a typical problem for all model programs, including the F-35, is the tendency to end up with a number of unconnected solutions. Specific software for specific domains often results in the need for manual intervention when transferring results from one end to another.
Nothing is easy, but with the Heliple project, Saab Aeronautics has provided a plausible way to meet the demands and interaction opportunities needed create the more flexible structures required to cope with complexity, combined with traceability, security aspects and the ability to maintain data integrity over time, which is a core issue when talking about 30 to 50-year life cycles.
All in all, this is an interesting project that will be very exciting to follow.
For More on this, read, The Hunt for the Ultimate PLM System, Part 1: Saab Aeronautics and the Gripen Fighter Aircraft.