Measurement information management for industry 4.0

Abstract

The measuring procedure includes measurement uncertainty by default. When reporting measurement findings, it is vital to assess measurement uncertainty. In dimensional metrology, a coordinate measuring machine (CMM) is a common measurement tool. To assess measurement uncertainty, a measurement expert might employ a variety of techniques and software. When using a CMM to measure an artefact, there are several factors that contribute to uncertainty in dimensional measurements. These factors may include environmental conditions, the characteristics of the artefact being measured, the design of the CMM and its components, the skills and actions of the operator, the measurement methods employed, and the calibration of the reference standards. For the purposes of risk analysis and decision-making, it is important to be able to evaluate measurement uncertainty. Industries often rely choices on data from quantitative measurements. To assess the uncertainty, detailed knowledge (metadata) regarding the measurement process is needed. It is becoming more important in manufacturing and metrology for Industry 4.0 to organise pertinent metadata and associated information to present knowledge in a semantically appropriate, interoperable, reusable, and accessible way. Understanding measurement process and uncertainty is crucial for measurement, acknowledging measured data and evaluating the accuracy of the process. Data traceability is vital for the quality of the measuring process in Industry 4.0 and other organisations. The need to arrange appropriate metadata or knowledge in a way that promotes semantic presentation, reuse, accessibility, and interoperability is increasing. Ontology in knowledge engineering is a promising technology to fulfil this requirement. The results and information from the CMM measurement process’s uncertainty evaluation procedure may be viewed as knowledge in terms of an ontology so that the user may evaluate and update the sources of uncertainty values or data to achieve uncertainty improvement for the existing measurement. Although various studies from knowledge engineering, such as maintaining the knowledge of CMM inspection planning, have been suggested to satisfy this requirement. Past work still needs to address the uncertainty during the measurement process fully. Some early studies have been conducted to create web-based software that evaluates measurement uncertainty, allowing the user to select the mathematical model and supporting the decision-making. These application results are available in amorphous formats, including excel sheets. Obtaining and updating critical information from these spreadsheets for subsequent usage is challenging. Because the uncertainty sources and their values change over time, user must investigate and invest time to acquire the data to reduce the measurement uncertainty. This research has addressed the existing challenges mentioned above. An ontology of CMM measurement knowledge (OCMK) system has been developed. This system consists of the ontology development of CMM measurement and uncertainty of results and a web application to integrate this ontology. The web application is connected to Jena Fuseki. It works as a server for the developed Ontology and Python Flask API. The user interfaces for uncertainty evaluation have been developed by Python, HTML, Bootstrap, JQuery and AJAX for the application. The user can use the GUM Framework to analyse the CMM measurement uncertainty by entering the mathematical model in a web application. The original ontology’s result class was expanded and updated dynamically to include the information or outputs from the uncertainty estimation. Sources and values of the uncertainty are utilised to support further analysis and decision-making concerning the measurement and product quality verification. After evaluation user can customise the results, which are directly added to the ontology. It also provides the facility to search the knowledge semantically. The NLP BERT transformer is utilised to get the chatbot’s functionality, such as question-answering, in the web application for the OCMK system. The three case studies of experimental data have also been implemented for this system.