INFORMED MANUFACTURING APPLIED AT TAILOR WELDED BLANK PRODUCTION Tavares, J.J.P.Z.S, Santos, M. dos e Silva, J.R.
Resumo: Nowadays, production systems based on mass production paradigm have been characterized more by the control of the final items than the control of its process. The attempts to invert this situation, stimulated by the increasing of quality and personalization demand, is constantly hindered, or by information lack on the new product processing in its several transformation stages, or by the difficulty to identify which information would be relevant in a future analysis, or both. The effective analysis of the productive process facilitates the identification of specific and prompt failures in the products even after its manufacture, making possible an oriented recall action, less traumatic and rough for the companies' finances and image. However, even more could be inspected and reported, as the cause-effect relationship among uncontrollable events and the resultant product, or the existence of repetitive and/or redundant steps, and so on. The increase of information follows the product complexity, principally in an automated shop floor, and to keep all the process description to those products is a challenge. This work consider a solution committed to deterministic and automated manufacturing processes, establishing a product standard behavior after manufacturing and a system that looks for atypical events, capturing the available contextual information either in the product, through an automatic data support as RFID; or in the manufacturing process, extracting sensors signals and machine states, expressed in Petri Nets. In this case, a perception system would register statistically the pertinent information related to the standard case and would capture details only about the standard misaligned occurrences, resulting in a considerable reduction in the amount of information and enabling a better environment for posterior analyses. Based on EPC Network proposal, we named this tool Savant-2 and EPCIS-2. This new approach brings automatic context to data capture and can assist invariant analysis of exception. Savan-t2 and EPCIS-2 applied in a manufacturing process, providing an information feedback mechanism, is called Informed Manufacturing. A case study of Informed Manufacturing in Tailor Welded Blank Production is presented.
|