|
|
www.johnstark.com |
|
|
|
In the typical company with large numbers of people involved in the engineering and associated functions, many problems arise from the lack of control over engineering data and the engineering process. People are sometimes unable to find the information they need. If they can find the information, it may not correspond to the actual state of the product. For example, a facility drawing may not correspond to the physical facility layout. Developers are unable to rapidly access a particular design among the mass of existing designs. To find specific information, they may have to search through tens of pages of listings. They lose valuable time. Studies show that design engineers spend up to 80% of their time on administrative and information retrieval activities. They develop new designs that may be almost identical to existing designs with the result that unnecessary additional costs are generated as the new designs are taken through the various activities necessary for manufacture, and then supported during use. As more and more data is generated on computer and other electronic systems, it becomes more and more difficult, with manual control and management procedures introduced when data volumes were twenty or even fifty times lower, to track the location of data, to prevent unauthorized access and to maintain up-to-date product configurations. Large companies hold many hundred TB (Terabytes) of data (1TB=1,000 GB). Companies hold thousands, or even millions, of drawings. 3D CAD part models may run to many MB. One company calculated that it needed 250,000 pages to describe a new product, and that, on average, each of these was reproduced 30 times. This can be thought of as many tons of paper, or many Gigabytes of data. Data entry is poorly controlled. Data is lost and can not be retrieved. It is re-created and errors are introduced. The wrong product goes to a customer. Product configuration data is not up-to-date. When a defective part is found in the field, many more products than necessary have to be recalled. Design history is not maintained, so it is impossible to draw on previous experience. Due to gaps between incompatible application programs, data is transferred manually, and errors occur. They have to be corrected, and their correction has to be managed. This costs time and money. An error slips through and is not discovered for several weeks. Correcting it leads to months of delay. Part descriptions and Bills of Materials developed on a CAD system may need to be manually transferred to an ERP system in a computer that is not linked to the CAD system. The Manufacturing Bill of Materials may be different from the Engineering Bill of Materials. The two systems may be the responsibility of different departments or organizations. The change processes in the two organizations may be different and out-of-step. At a particular time, a given change may have been made in one system, but not in the other. As a result, not all users will have immediate access to the most up-to-date information. Thousands of dollars can be wasted if the Manufacturing organization works on the wrong version of a design for several weeks. Several copies of the information describing the same part are maintained. Nobody knows which is the master copy. When a change is needed, not all copies are changed, and not all the downstream functions are alerted. Old, unwanted revisions of parts are machined, while the new, required versions are ignored. When there is no agreed master version of a particular item of information, and no agreed owner, all users of the information will behave as if they were the owners. Each user will define the item to suit their particular requirements. All the definitions may be different. Such wasted effort leads to confusion when information is transferred between users. |