- Welcome Page 2 - Update Page 3 - News Page 4 - Brief lines | Welcome to this issue of 2PLM, an e-zine distributed every two weeks on a free subscription basis. This week's lead article CAE Data Management is from Alhad Joshi of Altair, Inc. But first, further to mention of the "new PLM" in the previous issue of 2PLM, several readers asked for more details of the old PLM. OK, so a few words about the old PLM, but first a few more words about the new PLM. * PLM : a business activity * The new PLM is first and foremost the business activity of managing products across their lifecycles, from cradle to grave, in a joined-up way. It aims to increase revenues from new and existing products. The new PLM is holistic, simultaneously taking account of components as different as processes, applications, work methods, people, information, products and services. The new PLM application One of the components of the new PLM is IS, but it's not clear yet whether there will be just a single integrated PLM application, or a set of PLM applications. Some people see the IS solution for PLM as a single integrated application bringing together the various individual applications that address a company's products - similar to the way that the ERP application integrated all sorts of code and programs that addressed different parts of a company's finance, administration and human resource activities. Other people see the IS solution for PLM as a set of discrete best-of-breed applications linked to a common backbone. Old PLM Old PLM existed in companies in various shapes and forms. In the old PLM environment, companies did, of course, manage their products across their lifecycles, although "manage" may not really be the right word. For example, in one company we worked with, the PLM business activity was split between more than 50 different processes, departments, companies and systems. The number of times the responsibility for the product, or some part of it, changed hands was even higher - so it wasn't surprising that product development times were far too long, that known problems with existing products were not solved and found their way into new products, or that golden opportunities to extend the lives of existing products were lost. This was old PLM - there were products, but neither the lifecycle nor the management were very visible. Old PLM existed in many forms, probably in more forms than there are companies. Usually it was not "across the lifecycle", not integrated, and not holistic - but addressed just one component, such as product data, in each of its manifestations. Old PLM didn't get on the CXO agenda, but was seen as a low-level technical problem to be solved by departmental managers. More on New PLM To find out more about new PLM, and old PLM, read Product Lifecycle Management: Paradigm for 21st Century Product Realisation. This looks like being the book that establishes new PLM on the top management agenda. It's received strong support from both PLM users and vendors of PLM products and services. However it's not easy to find. The first print run of September 2004 has sold out, and the second print run isn't yet out of the printer's doors. Negative points noted in reviews include: too many bullet points ; too long ; a lack of graphics, but in spite of these, reviewers claim it's a must-read. * CAE Data Management * As companies look for ways to reduce the cost of designing better products, Computer Aided Engineering (CAE) has started playing a significant role in the product development process. However, the growth in the number of CAE simulations performed, and an absence of data and process management tools to track these simulations, has resulted in inefficiencies in the CAE process. CAE includes activities in areas such as Statics and Normal Modes, NVH, Impact, Acoustics, Fluids, Thermal, Durability etc. Most of this analysis is done ad-hoc. Analysts access product geometry and materials data, set up specific analysis, execute, post process and draw conclusions. Almost 90% of CAE is done this way. CAE analysis tends to generate large amounts of data. Often this data is stored in local file system directories. Maintaining data traceability is a major issue. Even when there are few users, it is difficult to keep track of all inputs, outputs and results for every simulation and every iteration. Currently there are no CAE-specific data management systems. However, the PDM systems used for managing CAD data could be leveraged to manage the CAE process as well. But there are many issues to be addressed:
1. Enable automated CAE setup The objective of this approach is to improve productivity by setting up rules, mesh-criteria, expected loads and boundary conditions in a generic way for different analyses with different product configurations. These rules are set on part-families or packages specific to the product line of the corporation. As an example, it will be possible to define a CAE BOM configuration rule for a specific analysis type that determines if a part of a particular type or category should be included in the CAE Bill of Materials. In case of a vehicle Body-in-White for example, the CAD structure can easily have over 2000 distinct parts. However, a corresponding CAE build could include no more than 400 parts, with the other parts either completely ignored, or modeled as simple masses. Once rules defining these criteria are defined and captured, it will be possible to configure the CAE build quite easily by applying these rules. Such rules can also define the kind of models required for specific analysis, such as coarse or fine meshes, with or without all CAD features, etc. Capturing such rules for CAE setup will address the very labor-intensive task of selecting the right parts and appropriate mesh types for building CAE assemblies. Although the set-up of such rules could be an extremely time-consuming task, some corporations are in the process of investigating the approach. 2. Manage CAE execution The objective of this approach is to improve productivity by managing all inputs such as CAE analysis structure, mesh files for all parts and their variants, solver decks, solver results, and reports ; manage multiple iterations of a simulation ; track input modifications, etc. This approach addresses the current situation in which - in the absence of a data management system - CAE teams gather all required data, set up the analysis, perform simulations and generate reports. In addition to this, analysts typically use specific processes for setting up CAE analysis. There is a need to capture these "process parameters" in addition to managing file associations. A CAE data management system A data management solution to address these CAE needs must satisfy the current practice outlined above. The primary objectives are:
Implementation Altair's HyperWorks Data Manager (HDM) is an example of such a system. The current release is designed to manage all aspects of CAE execution as described above. It is built in 3 layers:
For more information contact (Technical) Alhad Joshi, Director, Software Consulting, Enterprise solutions, alhad@altair.com 248.709.6146, or (Business) Keith Parmentier, Director, Business Development keithp@altair.com 248.321.3776  * PLMIG News * Among the events planned by the PLMIG in April 2005 are:
|
