This deliverable contains a guideline to manage data quality in manufacturing. It establishes a conceptual basis by introducing several concepts, such as data and information, data life cycle, information needs, data and information quality, and production system levels. The guideline uses the Plan-Do-Study-Act (PDSA) cycle and focuses on the Plan and Do steps. Section 3.1 outlines an information flow analysis for producers to understand which data quality factors the organization must manage. Section 3.2 suggests three types of measures to manage data quality factors. Awareness measures aim to raise awareness of data quality issues and factors among employees. They require the least effort but are also not very reliable unless strictly controlled. Programmatic measures are functions in software that force users into behavior that ensures high data quality. Examples are input validations and auto-complete. These measures are much more reliable but may be costly to implement. Organizational measures cover complex cases where other measures are not feasible. They focus on larger-scale organizational activities (e.g., work instructions, training, and new roles) to promote behavior that minimizes data quality issues.
