When an average vehicle boasts more than 1 million lines of software code and dozens of ECUs (engine control unit) and common household appliances like dishwashers and refrigerators come standard with mini on-board computers, it doesn’t take a rocket scientist to see something about the traditional engineering process has to give.
The complexity of today’s products has added another dimension to product development as companies from car manufacturers to appliance makers struggle to integrate mechanical, electrical and software components into their products and, at the same time, uphold their business goals of improving time-to-market and meeting ever-higher quality standards.
Unlike traditional design processes where electrical engineers worked on their components relatively unconnected to the design efforts of mechanical engineers or the software development team, this new kind of product development — mechatronics as it’s sometimes called — demands a wholesale shift to a more concurrent, systems engineering approach where all the disciplines collaborate early on in the process. The reason for this cross-discipline collaboration is simple: With companies facing fierce competitive pressures, they can no longer afford to bump up against possible design flaws or incompatibilities between systems at the end of the design cycle when making changes is expensive and an impediment to shipping products in a timely manner.
Wait a second. If you’re thinking this scenario has a familiar ring, it does. But while these engineering principles have been around for decades, particularly in industries like aerospace and defense, the truth is companies have done a lousy job of collaborating across disciplines, mostly because of the lack of cross-functional knowledge among engineers and the availability of tools to facilitate this level of coordination.
“Electromechanical design has been around for a while, but what that’s really meant is a bunch of people over here doing mechanical design and others over there doing electronics and software and someone bringing them together, but it wasn’t very well-coordinated and not very well-managed,” says Ken Amann, director of data for CIMdata Inc., a research and consulting firm specializing in engineering practices. “Under the banner of mechatronics today, we’re trying to tie those processes more closely together so there really is interaction going on between the areas and one can see where one might influence the other more quickly.”
A number of things need to coalesce to make this happen. Development organizations need to train the different engineering disciplines to understand — and care about — the fact that a design alteration they make may have impact on areas of the product design that are outside of their domain. That necessitates cultural changes and, oftentimes, modifications to the power structure, neither of which is an easy task. Companies also need to place more of a premium on engineering disciplines such as requirements management and testing and simulation in order to fully close the loop and catch problems early on in the process. Finally, adoption of new design tools and business processes around Product Lifecycle Management (PLM) can help organizations along the systems engineering path and create efficiencies when designing for mechatronics.
How does this translate for the average engineer? It means a move away from a silo mentality when it comes to the tools, processes and parts of a design they’re involved with. At a higher level, the changes mean someone in the organization has to take ownership of decisions at a systems level and be able to make the tough call that might not sit well with the vice president of electrical engineering or the head of mechanical engineering. “The traditional mechanical engineer now has to be more aware of control systems electronics, perhaps even embedded software to understand how the software will be controlling the mechanism he or she is developing,” says Chad Hawkinson, vice president of product strategy for electronics, at PLM provider PTC. “Those kind of skills are not yet well-established in academic training.”
One of the areas demanding a cross-functional focus is requirements management. Here, companies are striving to do a better job at gathering the requirements for the total product, not just from the standpoint of what’s necessary for an electronics or mechanical design point of view. Beyond scoping requirements at a systems level, part of the big difference with a mechatronics approach is tying those requirements to every aspect of a design and managing them throughout the entire life cycle.
“When you think about requirements at the beginning, it’s like a check list, trying to put the puzzle pieces together as the products get more complex,” says Chad Jackson, research and service director at Aberdeen Group and one of the authors of a new report, “System Design: New Product Development for Mechatronics.” "You can’t have a narrow focus on these are my requirements — I’ll do what I have to do and nothing more.”
Performing simulations and testing from the standpoint of predicting system behavior prior to building physical prototypes is another critical element to successful mechatronics design and an area not yet fully addressed by most engineering organizations or by their design tool providers, according to Jackson. Consider a product as simple as a CD player: Engineers should be able to simulate using a virtual model what the CD player will do if a user presses the open button. While that sounds fairly basic, it involves a cross-discipline simulation that models how the motion of pressing the open button triggers a signal from a circuit board, which, in turn, translates into how the door mechanism opens.
Several of the PLM vendors, including Dassault Systémes, are working on just that capability. Dassault touts its on-going integration efforts between CATIA, ENOVIA and its SIMULIA brand of simulation applications along with its 2006 acquisition of Dynasim, a maker of a Modelica embedded software simulation tool as steps moving it in that direction. Dassault is also working toward defining a universal model that can be applied across all disciplines and is positioning its new 3DVIA online community as a way for companies to more effectively engage customers in gathering requirements and soliciting feedback on early, digital iterations of products. As such, mechatronics, as Dassault sees it, is becoming a base for product virtualization. “Most companies today have to deliver a physical prototype in order to test the product,” says Laurent Cherprenet, director of high-tech industry solutions for Dassault. “What’s critical is a unified model that allows them to test quickly and validate the optimal product architecture for many different dimensions, including safety, regulatory compliance and performance.”
PLM to the Rescue?
Dassault and its design tool competitors envision PLM as the core platform for facilitating this cross-functional collaboration. Beyond PLM’s existing functionality as a respository for all kinds of product-related materials and 3-D data, the platforms typically offer a variety of collaboration capabilities and have evolved over the years with workflow functionality for creating new and streamlining current cross-functional business processes. Still, PLM as it exists today in most vendors’ product offerings falls short in terms of being a true platform for facilitating the systems engineering view required by mechatronics product development. “Companies have invested a lot in PLM in terms of the mechanical world, but a higher percentage of what they’re now worried about is software and electronics and that’s not part of the PLM environment,” says Mike Burkett, vice president, PLM research at AMR Research Inc.
PLM vendors like PTC, Dassault and Siemens PLM Software are working to address those gaps. In its WildFire 4.0 release, PTC, for example, is offering the ProENGINEER ECAD-MCAD Collaboration Extension, which automatically identifies incremental changes between MCAD and ECAD versions of a PCB board design and creates tighter interfaces between the previously disconnected systems. Siemens PLM Software offers Teamcenter Embedded Software Manager, which allows engineering teams to manage software as individual components of a product. And National Instruments is collaborating with SolidWorks to integrate their respective products so engineers can test control systems developed in LabView on virtual mechanical models instead of having to wait months until a physical prototype is ready.
On top of these efforts, PLM vendors are aiming to build on this level of integration to create relationships and interdependencies between the various domains like ECAD and MCAD so when a change is made to one area — say, a printed circuit board design — it’s reflected in other areas that are affected like the mechanical design or in the software.
“We’ve gotten good at managing multidisciplinary information, but the next thing we need to do is relate the modules to each other so I know the knee bone is connected to the thigh bone,” says Eric Sterling, vice president, portfolio marketing for Siemens PLM Software.
Even without these next steps, companies like The Alloy Ltd. are already leveraging 3-D design tools and new development processes to bring a more systems-level view to their mechatronics projects. Consider a project Alloy undertook to redesign the Argus 3, a thermal infrared camera used by firefighters to see through smoke. Based on feedback from firefighters about how they used the camera in action, it became clear a remodel was in order to allow pairs of firefighters to pass it back and forth, according to Gus Desbarats, chairman and founder of The Alloy. The old version had one pistol grip handle which wasn’t sufficient for a handoff in a fail-safe way. To complicate matters, The Alloy needed to marry a common external look and feel with three separate electronics components to be in compliance with the different export rules in the different markets. There was no way, Desbarats says, The Alloy could be successful in this kind of design initiative unless there was close collaboration and real-time information sharing between mechanical and electrical engineers involved in the project.
“This is what mechatronics is all about,” Desbarats says. “We had to agree on a way of visualizing the electronics in terms of headroom and board sizes and model that early on. It wouldn’t have been possible without modern 3-D tools.”
Using Siemens PLM Software’s NX and Teamcenter and fostering a cultural change that encouraged the two disciplines to collaborate and share “smart guesses” about their early design parameters, The Alloy was able to come up with a radical, new two-handed design, which it was able to show off to its client virtually, without building a prototype, in fairly short order. “We were able to show them where the electronics fit in in this radical new concept,” Desbarats says. “We couldn’t have done that if the inside and outside were not designed by concurrent teams sharing data.”
If these students are any indication, the future of mechatronics is bright.
Latest blog Post: The tragedy of over-compromise can make products ill-equipped. When will the design process become more fluid in the introduction of diversified products? Stefan discusses an example and relates it to technology.
Posted in: Mechatronics on Campus | 11.27.2008.