PDQ (Product Data Quality): Representation of Data Quality for Product Data and Specifically for Shape Data

Abstract

In the communication and sharing of product data, if the difference of the required data quality and the data quality actually incorporated into data is significant, it causes various problems. It is often the case that a creator of low quality data does not realize it unless it is harmful for his job. In most cases, low quality data passed to subsequent processes, such as manufacturing process, cause problems since these are not appropriate from the machining precision point of view or the detailed shape modeling point of view. In these cases, rework or repair of data is necessitated before commencing the target process, which results in significant economy loss and delay of product development. Today’s product model data are dumb data because design intents and data quality incorporated are not explicitly represented. Receiving systems cannot know whether the data passed possess sufficient quality for the target job or not. Another problem is that engineers in later processes, such as the manufacturing process, cannot issue data quality related request beforehand in a commonly agreed manner. The problems mentioned above are caused by the lack of a commonly agreed representation of product data quality (PDQ) information. Our proposed solution is designed to enable the communication and sharing of data quality information. This paper reports the development of a PDQ standard (ISO 10303-59), which is a resource part of ISO 10303 Standard for the Exchange of Product Model Data (STEP) (2008, “ISO 10303-59, Industrial Automation Systems and Integration. Product Data Representation and Exchange. Part 59 Integrated Generic Resource: Quality of Product Shape Data,” International Standard Organization, Geneva). The objective of ISO 10303-59 is to establish a PDQ model and to enable the use of PDQ data independently or in combination with product data. The developed PDQ information model represents concepts such as data quality criteria, measurement requirements, and measured results. Based on the PDQ model, the PDQ for shape data model, which is a specialization of the PDQ model to 3D shape data quality, is also developed.