Organizational process maturity model for IoT data quality management

Abstract

Data quality management (DQM) is one of the most critical aspects to ensure successful applications of the Internet of Things (IoT). So far, most of the approaches for assuring data quality are typically data-centric, i.e., mainly focus on fixing data issues for specific values. However, organizations can also benefit from improving their capabilities of their DQM processes by developing organizational best DQM practices. In this regard, our investigation addresses how well organizations perform their DQM processes in the IoT domain. The main contribution of this study is to establish a framework for IoT DQM maturity. This framework is compliant with ISO 8000-61 (DQM: process reference model) and ISO 8000-62 (DQM: organizational process maturity assessment) and can be used to assess and improve the capabilities of the DQM processes for IoT data. The framework is composed of two elements. First, a process reference model (PRM) for IoT DQM is proposed by extending the PRM for DQM defined in ISO 8000-61, tailoring some existing processes and adding new ones. Second, a maturity model suitable for IoT data is proposed based on the PRM for IoT DQM. The maturity model, named IoT DQM3, is proposed by extending the maturity model defined in ISO 8000-62. However, in order to increase the usability of IoT DQM3, we consider adequate the proposition of a simplification of the IoT DQM3, by introducing a lightweight version to reduce assessment indicators and facilitate its industrial adoption. A simplified method to measure the capability of a process is also suggested considering the relationship of process attributes with the measurement stack defined in ISO 8000-63. The empirical validation of the maturity model is twofold. First, the appropriateness of the two models is surveyed with data quality experts who are currently working in various organizations around the world. Second, in order to demonstrate the feasibility of the proposal, the light-weight version is applied to a manufacturing company as a case study.