|
|
www.johnstark.com |
|
|
|
Key management requirements from EDM/PDM Management needs to get engineering data under control. Engineering data lies at the heart of the engineering function in today's computerized, information-based environment, yet few companies have it under tight control. Data is scattered in many locations. It is on paper, microfilm, aperture cards and computers. Most engineers do not fully trust 'the system' and, to protect themselves, keep several copies of data, and try to stop other people accessing their data. Often several groups of users will have different versions of what should be the same data. Each will claim to be the legitimate owner of the data, hence the owner of the correct version. As companies invest in more and more engineering workstations, thus creating a truly distributed computing environment, it becomes even more difficult to maintain control of the company's engineering computers, and to maintain control of engineering information, a valuable company resource. It becomes more difficult for companies that must comply with legal requirements on traceability to maintain audit trails - so that they can track back to the source of any product problems. It becomes increasingly difficult for companies to know which is the 'master' version of data - is it a computer-based model or a drawing, is it the scanned image of a document, or the document itself? As more data is shared with suppliers, similar questions arise as to which company has the master version. Management needs to make sure that the engineering data in use is of a high quality. Without reliable, timely and accurate information, managers and users can not work efficiently. It is not easy to run quality checks on data moving invisibly round networks. Instead, quality has to be built in. This can only be done through the right procedures and a company culture that penalizes poor quality work. Error creation and propagation must be prevented. It is only too easy for a user to introduce an error into data. Once the error is in though, it can be difficult to find. It can be even more difficult to remove its effects. Management needs to make sure that engineering data is reused, and that it is allowed to evolve. There is no point in continually reinventing the wheel, yet designs need to be revised so that they continue to meet market requirements. Reuse of information was supposed to have been one of the major advantages of CAD, but in practice there has been less reuse than expected. The reason has been the difficulty for users, even if they are aware of suitable existing data, to actually find this data. It may be somewhere on the CAD system or somewhere in a drawing store. Few users are prepared to spend hours, or even days, hunting for existing data. Instead, they will take a clean sheet, or screen, and design a new part. Management needs to ensure that its engineering data management strategy addresses both existing, 'legacy' data and data that will be created in the future. Engineering data management is not a green field activity, and should not be treated as one. Most companies have a vast amount of information tied up in existing data. One of the major challenges to management when developing an engineering data management strategy is to marry the ability to meet current and future needs effectively with the capability to reuse existing data. Management must make sure that engineering data is secure. There are many facets to the security needs for engineering data. They range from the individual user worried that a colleague might unintentionally overwrite a file, to the company that worries that access rights granted to trusted suppliers might somehow be discovered by unscrupulous competitors. Major multinationals transmitting design information by satellite between sites in different continents wonder how secure their data is. Similarly, defense organizations, while promoting increased use of digital data and electronic data repositories as a means of reducing project costs, are aware that major effort will be required to maintain data security. Management must ensure that confidential and proprietary information is protected from unauthorized access. Management will have to make data available to users when they need it, otherwise valuable time will be lost. If management really wants to reduce lead times, it will have to cut the waste out of the engineering process. Administrative paper shuffling will have to be abolished. Just In Time techniques, successfully applied on the shop floor, will have to be applied in engineering functions. Engineering data must flow smoothly through the organization. Old habits of spending hours, or days, looking for data will have to disappear. The data made available to a user will of course, have to be the right data, otherwise more time will be lost until the correct version is found. Within this context, a user may not only be someone working a few yards away from where the data is stored. It could be someone in another organization, on the other side of the world, in which case opportunities abound for wasting many hours of valuable time. Management needs to take control of the engineering workflow. In many companies, few people can describe the engineering workflow, and even fewer know why it has the shape it does. In most cases, the flow results not from a reasoned design, but from a long series of minor reorganizations that resulted from changes in departmental structures, product characteristics and human resources. Managers should carry out a review of the current workflow from initial product specification down to customer support. This will provide them with a solid base from which improvements can be made. Activities that do not add value should be removed, and activities that were previously carried out in serial should, wherever possible, be run in parallel. The management of changes, modifications, versions and variants will have to be improved. Unless configuration management is improved, the rapidly increasing number of changes, and of different versions of products, will lead to configuration control getting out of hand. Management must make sure that EDM/PDM fits into the overall engineering strategy. Improved management of engineering data and the engineering workflow will provide an integrated framework within which management will be able to adopt and support strategies to reduce lead times and cut product costs. On-time availability of data reduces lead times. Concurrent Engineering techniques, implying parallel activities, will not work efficiently unless the workflow is under control, and data ownership and usage rights have been defined. Design for Manufacture, and similar techniques that aim to involve Manufacturing staff early on in the design process, can only benefit from a clearly defined workflow and the certainty that all parties are working with the latest version of the data. |