Monitoring and Control of Unstructured Manufacturing Big Data

Abstract

Unstructured manufacturing big data silos are challenging for enabling various data-driven applications such as digital threads and digital twins in manufacturing. The management of big data silos requires to address the issues of large volume, data inconsistency, data redundancy, information silos and data security. This research developed a systematic approach to managing data silos using the state of art big data software. Applying this approach in the product life cycle can control data silos, data consistency, redundancy, timely update and enable the automatic workflow of each system.

A Systematic Review of Data Quality in CPS and IoT for Industry 4.0

“…The reviewed papers address wide-ranging data quality issues. They include, amongst others, outliers (isolated, erroneous values) [2, 8, 9, 12, 13, 15, 19, 28, 29, 34ś36, 40, 42, 43, 45, 47, 49, 50], missing values [1ś3, 6, 9, 14, 18, 19, 21, 25, 28, 32, 33, 35ś39, 43ś46, 50], duplicated records [9,14,19], noise in data [5,19,30,33,37,42,45,48], data drift [14], data discontinuity [17], data imprecision [25], data timeliness (freshness) [1,3,10,16,21,22,26,38,39], high dimensionality [9,19,42,43], data inconsistency [1,3,4,6,10,25], and data veracity [6,7,11,20,23,27,31,33,38,39]. Data quality issues are mainly addressed using data quality dimensions, i.e., attributes representing a single aspect of the data quality [147].…”

“…Only ten papers mentioning programming languages and solutions (i.e., [4,5,9,10,21,23,26,27,35,39]) address data quality techniques for CPS and IoT applications as their primary goal. For instance, Lin et al [23] propose a new approach for repairing corrupted data in IoT applications.…”

“…The data collected, processed, and exchanged at each stage of CPS and IoT applications might have a diferent structure, format, and velocity and be stored in data silos not available to all the system users but only to some users. Cui et al [10] propose a systematic approach, i.e., Data Control Module (DCM), that uses state-ofart big data software to manage data silos in manufacturing. A data silo concerns timely monitoring of data changes, redundancy, inconsistency, and insecurity.…”

“…Nine primary studies [4,8,10,16,23,24,31,40,49] cover edge-cloud orchestration for data processing (not necessarily data quality management). Four of these studies [23,31,40,49] provide a data quality management technique (data cleaning and repair) in the architecture layers from L1 to L7.…”

Determining the Level of Use of the Industry 4.0 Solutions in the COVID-19 Pandemic Era: Results of Empirical Research

“…The literature shows the diversity of Industry 4.0 technologies. This collection includes, among others, (1) big data and analytics [20][21][22][23], (2) autonomous robots [24], (3) simulations [25][26][27][28][29][30], (4) horizontal and vertical system integration [31,32], (5) Internet of Things-IoT [31], (6) cyber-security [33,34], (7) the cloud, (8) additive manufacturing, (9) augmented reality, (10) artificial intelligence [35], (11) mobile technologies, and (12) RFID and RTLS technologies [12,30,[36][37][38][39][40][41][42]. Each solution could be independently implemented as a separate project for the organisation.…”