MODEL DRIVEN ENGINEERING: AN OPEN FORUM††††††††††††††††††††††††††††† Oct 26, 2012





Herman Migliore, PSU Director Systems Engineering

After more than 40 years as a working engineer, I have retired. Well, semi-retired anyway! I still have the privilege of directing PSU's systems engineering program.



Model Based Systems Engineering

SysML (OMG/UML) seems to be the most popular approach to MBSE.There is a rich history of modeling in EE, ME, and SW Engr.  Systems Engineers can learn them.  Furthermore, Systems Engineers are responsible for coordinating the efforts of the domain specialists (and the information they provide).  Let's see what we can learn from each other.†† This forum will be a test run to prepare for a one day, regional conference -- after another yearís worth of progress.Much of my material comes from presentations at the INCOSE Workshop on MBSE, January 2012:




James Godfrey, CH2M Hill

James Godfrey has masters in electrical engineering and 14 years of experience including manufacturing, critical system design, and system integration.His experiences range from data centers in Panama to nuclear remediation in Hanford to semiconductor manufacturing in Israel.He has been with CH2M Hill for 13 years and currently works with the Industrial and Advanced Technology business group.



UML & SysML for Industrial Automation

CH2M HILL has been applying UML to our manufacturing software requirements efforts for a number of years.We are now starting to apply SysML to our systems versus the typical construction drawings and specifications.Pros and cons for SysML will be discussed.James' Visuals




Andy Byers, ANSYS

Andy has been with ANSYS for 8 years as an application engineer, account manager and regional sales director. Before that he worked at Tektronix designing next-gen test and measurement equipment. His
educational background is in computational electromagnetics - BS/MS at University of Colorado Boulder.

Multi-physics, Multi-scale System Simulation
Sub-Title: Methodology that links together component, sub-system, product, and operating environment levels and changes traditional engineering teams and functions.

In the past 40 years, engineering simulation has revolutionized the product development process. By minimizing costly physical testing, accelerating time to market, and enabling design innovations in a low-risk virtual environment, simulation has helped businesses in every industry achieve significant competitive advantages. However, new challenges arrive that are forcing the engineering community to broaden its view of the product design process. For example, the new generation of smart products, including consumer offerings like phones, tablets and automobiles as well as industrial products such as wind turbines, is engineered to sense and respond to user needs and the surrounding environment. This creates interesting new challenges for engineering teams because these products are comprised of many interconnected subsystems that rely on the performance of one another. To keep pace, engineering teams must shift from a component or subsystem view to a higher-level perspective that considers performance at the product and operating environment levels. To achieve this they need
to apply multiple physics, multiple scales and a collaborative engineering approach. This talk will look at a multi-scale systems design approach and focus on the interactions that need to happen between the component, sub-system, product, and operating environment levels. How engineering tools can traverse these different levels and the resulting impact on engineering teams and functions will be included in the discussion. Andy's Visuals




Ryan Slaugh, Engineer, Pacific Northwest National Laboratory

Ryan Slaugh has been an engineer with PNNL for eight years.In that time he has worked on a multitude of hardware and software integration projects.His skillset includes software and firmware development as well as hardware design including discrete circuit design, printed circuit board layout, and design for manufacturability.Ryan is currently working on modeling solutions for rapid prototyping in a research and government environment as well as integration challenges in secure environments.Ryan is also a current student in the Masters of Systems Engineering program at Portland State University.



Modeling of Development at a National Lab

Ryan will discuss the challenges present in a research environment such as those found at a national lab.While manpower and brainpower are plentiful, associated constraints can create a unique environment for development.One aspect that can be used more in this environment is modeling - both in terms of process modeling and sub-systems modeling.In his masterís project for PSU, Ryan is exploring the possibilities of hardware and software modeling and applying it to the laboratory culture and processes.




Neil Chung, SAIC

Chief systems engineer for the Defense & Homeland Solutions operation, approximately 2,000 employees - Assistant vice president for Science & Technology


Multi- Disciplinary Modeling (Without SysML)

Topics will start with 3D modeling for mechanical systems, to 3D modeling for major architectural systems, to camera performance modeling for video surveillance coverage, to radar performance modeling for detection range in varied environments.Neilís visuals.



Bill Chown, Mentor Graphics

Bill Chown is product marketing director for the system-level engineering division at Mentor Graphics. He worked as an electronic designer, systems engineer and group leader before joining the EDA vendor community. He is a board member for the Object Management Group, a Senior Member of the IEEE, and Secretary of the Cascade Chapter of INCOSE.


Model Driven Engineering

The adoption of title model-driven versus model-based engineering for this forum was a result of conversations with Bill.He will clarify in terms of model driven development covering systems design challenges, model abstractions, active requirements, and drive implementation. Bill's visuals.



Ken Propst

Mechanical Engineer with experience in design and manufacture of electro-optical systems.


Development of a SysML Educational Example

As part of his masters project in the systems engineering program, Ken learned about SysML and applied it to the development of an educational engineering system.He will discuss the education use case as well as the HW-SW aspects of the example engineering system, an accelerometer.Kenís presentation was based on his masters project on SysML applied to an accelerometer.



William ďIkeĒ Eisenhauer, National Director, Veterans Engineering Resource Centers

VERC is a Veterans Health Administration program to integrate industrial and systems engineering into the fabric of healthcare delivery.Prior to that position, he was the Chief of Systems Engineering at the Portland VA Medical Center. In addition to his role as overseeing the VERC program he is a professor of Systems Engineering and Engineering Technology and Management at Portland State University. His research interests include Adaptive Belief Management, Reconstructabilty Analysis, Shared Resource Constrained Data Envelope Analysis, Conflict Under Deceptive Irrationality, and Sustainable Quality Management Program Development. His previous experience has been in the areas of business process redesign, management science, and analytics in a number of various industries, including financial, entertainment, publication, and political surveying. He previously held positions as a Senior Risk Manager and Head of Loss Mitigation at Wells Fargo Home Equity.



Educating the Systems Engineering Student in Modeling
Introduce the topic of modeling in terms of purpose, objective, and context of modeling as a skill required of the systems engineering graduate, while also addressing industry need for competent and effective modelers.SysML/UML/MSBE points to a possible direction.What we can do different or better in the future, starting from what we classically do in Operations Research to Systems Dynamics to Simulation to SysML/UML/MSBE.
Ike's visuals.



Wayne Wakeland, PSU, Director System Science
From 1978 to 2000, Dr. Wakeland held fulltime managerial positions in information systems or manufacturing at local high tech firms (Tektronix, Photon Kinetics, Magni Systems, Epson, and Leupold & Stevens). Research interests include: system dynamics, systems thinking, discrete systems simulation, process modeling, manufacturing systems, information systems, and strategic planning, including:1) Manufacturing:process flow modeling, resource allocation, operational scheduling, maintenance optimization; 2) Environmental: fisheries management, ecological sustainability; 3) Healthcare: system of drug diversion & abuse, patient, flow, traumatic brain injury, animal colony management, health systems policy; 4) Information Management: data mining.His modeling and simulation courses also support the PSUís System Engineering Masterís Program. He also teaches systems thinking and sustainable operations at the Bainbridge Graduate Institute, in Seattle.

System Science Modeling and Simulation

Systems science methods focus on understanding the general properties and behavior of complex systems by creating models and finding patterns in data.Specific methods include :1) system dynamics which focuses on modeling the underlying feedback structures with differential equations which are solved to simulate behavior over time, 2) discrete system simulation which uses a Monte Carlo approach to analyze how the variety/randomness in systems impacts their performance with special emphasis on business processes and manufacturing systems, 3) agent based simulation which is used to study how low-level interactions between individual agents influences overall system behavior/performance, and 4) the use of algorithms to find patterns in large, complex datasets in order to better predict and control the systems which generated the data.Systems scientists strive to interact effectively with collaborators from other disciplines by using clear and well annotated graphics and diagrams to present models and data analysis results.Wayne's visuals.