The Mechatronics Design Process System.
The traditional electrome chanical-system design approach attempted to inject more reliability and performance into the mechanical part of the system during the development stage. The control part of the system was then designed and added to provide additional performance or reliability and also to correct undetected errors in the design. Because the design steps occur sequentially, the traditional approach is a sequential engineering approach. A Standish Group survey of software dependent projects found.
• 31.1% cancellation rate for software development projects.
• 222% time overrun for completed projects.
• 16.2% of all software projects were completed on time and within budget.
• Maintenance costs exceeded 200% of initial development costs for delivered software.
The Boston-based technology think tank, Aberdeen Group, provided key information on the importance of incorporating the right design process for a mechatronic system design. Aberdeen researchers used five key product development performance criteria to distinguish “best-in-class” companies, as related to mechatronic design. The key criteria were revenue, product cost, product launch dates, quality, and development costs. Best-in-class companies proved to be twice as likely as “laggards ” (worst-in-class companies) to achieve revenue targets, twice as likely to hit product cost targets, three times as likely to hit product launch dates, twice as likely to attain quality objectives, and twice as likely to control their development costs. Aberdeen’s research also revealed that best-in-class companies were.
• 2.8 times more likely than laggards to carefully communicate design changes across disciplines.
• 3.2 times more likely than laggards to allocate design requirements to specific systems, subsystems, and components.
• 7.2 times more likely than laggards to digitally validate system behavior with the simulation of integrated mechanical, electrical, and software components.
A major factor in this sequential approach is the inherently complex nature of designing a multidisciplinary system. Essentially, mechatronics is an improvement upon existing lengthy and expensive design processes. Engineers of various disciplines work on a project simultaneously and cooperatively. This eliminates problems caused by design incompatibilities and reduces design time because of fewer returns. Design time is also reduced through extensive use of powerful computer simulations, reducing dependency upon prototypes. This contrasts the more traditional design process of keeping engineering disciplines separate, having limited ability to adapt to mid-design changes, and being dependent upon multiple physical prototypes.
The mechatronic design methodology is not only concerned with producing high-quality products but with maintaining them as well—an area referred to as life cycle design. Several important life cycle factors are indicated.
• Delivery: Time, cost, and medium.
• Reliability: Failure rate, materials, and tolerances.
• Maintainability: Modular design.
• Serviceability: On board diagnostics, prognostics, and modular design.
• Upgradeability: Future compatibility with current designs.
• Disposability: Recycling and disposal of hazardous materials.
We will not dwell on life cycle factors except to point out that the conventional design for life cycle approach begins with a product after it has been designed and manufactured. In the mechatronic design approach, life cycle factors are included during the product design stages, resulting in products which are designed from conception to retirement. The mechatronic design process is presented in Figure 1-4.
FIGURE 1-4 MECHATRONIC DESIGN PROCESS
Because of their modularity, mechatronics systems are well suited for applications that require reconfiguration. Such products can be reconfigured either during the design stage by substituting various subsystem modules or during the life span of the product. Since many of the steps in the mechatronics design process rely on computer-based tasks (such as information fusion, management, and design testing), an efficient computer-aided prototyping environment is essential.
Important Features
• Modeling: Block diagram or visual interface for creating intuitively understandable behavioral models of physical or abstract phenomenon. The ability to encapsulate complexity and maintain several levels of subsystem complexity is useful.
• Simulation: Numerical methods for solving models containing differential, discrete, hybrid, partial, and implicit nonlinear (as well as linear) equations. Must have a lock for real-time operation and be capable of executing faster than real time.
• Project Management: Database for maintaining project information and subsystem models for eventual reuse.
• Design: Numerical methods for constrained optimization of performance functions based
on model parameters and signals. Monte Carlo type of computation is also desirable.
• Analysis: Numerical methods for frequency-domain, time-domain, and complex-domain design.
• Real-Time Interface: A plug-in card is used to replace part of the model with actual hardware by interfacing to it with actuators and sensors. This is called hardware in the loop simulation or rapid prototyping and must be executed in real time.
• Code Generator: Produces efficient high-level source code from the block diagram or visual modeling interface. The control code will be compiled and used on the embedded processor. The language is usually C.
• Embedded Processor Interface: The embedded processor resides in the final product. This feature provides communication between the process and the computer-aided prototyping environment. This is called a full system prototype.