PAIS-DQ: Extending process-aware information systems to support data quality in PAIS life-cycle

Abstract

The successful execution of a Business Process implies to use data with an adequate level of quality, thereby enabling the output of processes to be obtained in accordance with users requirements. The necessity to be aware of the data quality in the business processes is known, but the problem is how the incorporation of data quality management can affect and increase the complexity of the software development that supports the business process life-cycle. In order to gain advantages that data quality management can provide, organizations need to introduce mechanisms aimed at checking whether data satisfies the established data-quality requirements. Desirably, the implementation, deployment and use of these mechanisms should not interfere into the regular working of the business processes. In order to enable this independence, we propose the PAIS-DQ framework as an extension of the classical Process-Aware Information System (PAIS) proposal. The PAIS-DQ addresses the concerns related to data quality management activities by minimizing the required time for the software developers. In addition, with the aim of guiding developers in the use of PAIS-DQ, a methodology has been also provided to facilitate organizations to deal with complex concerns. The methodology renders our proposal applicable in practice, and has been applied to a case study where a service architecture implementing the standard ISO/IEC 8000-100:2009 parts 100 to 140 is included.

DMN for Data Quality Measurement and Assessment

“…Moreover, some DMN extensions let the integration of the decision making process non-only incorporating dataflow [17]. The necessity to incorporate data quality measurement in business processes was early identified in [16]. However, to the best of our knowledge, there is no solution that use DMN as a mechanism to model and evaluate data quality assessment and measurement requirements and it is still a challenge how bring the gap between the human description of data quality requirement description, with and automatic data quality assessment and measurement.…”