Different agglomerated contaminant materials (inhomogeneous geometry characterized by different volumes).
Misalignment between tire and wheel.
Others.
All these features are classified by particular image features extracted by image processing algorithms. Concerning the welding control of a large metallic surface as for tank production, IRT provides more information about weld homogeneity. Other algorithms, such as the K‐means clustering algorithm, are applied in order to enhance for example leakages [13], or in general for radiometric defect detection in photovoltaic systems [48].
1.2.4 Production Process Mapping
Process mapping represents a very important procedure to follow for quality control. All production data provided by sensors, and also results of the image processing tools, are in general stored in DB systems and mapped by diagrams such as fishbone diagrams, p‐charts, the Plan‐Do‐Check‐Act (PDCA) cycle, and by XmR charts, thus improving the quality check during the time. The PDCA cycle, also known as the Deming cycle diagram, is a four‐step iterative management method used for the control and continuous improvement of processes and products. Steps in PDCA include:
The Plan concerning the occurring variations.
The actions to Do and to be controlled.
The Check of the executions by validating expected results.
The Act involving all the activities about the process, the standardization and stability, and improving the existing procedure.
The XmR chart, through the upper and lower control limits, provides the limits of an industrial production process mapping at the same time production results. As a further diagram to adopt for process mapping, a fishbone diagram (Ishikawa diagram) is a cause‐and‐effect approach able to track imperfections, variations, defects, or failures. This diagram is structured as a fish's skeleton with the problem at its head and the causes for the problem following the fish bones. In order to also predict defects or production failures, big data systems facilitate the analysis of the historical data. Big data systems implement NoSQL technology based on the following main categories:
key‐value distributed
document oriented
column oriented
graph oriented
However, some NoSQL DBMSs may have features that belong to more than one of these categories. Some common big data systems are Cassandra, MongoDB, and Hbase.
1.2.5 Technologies of Industry 4.0 and Industry 5.0: Interconnection and Main Limits
The interoperability of technologies is a fundamental element for the upgrade of the information system and in general for the production. Figure 1.8 shows the interoperability of different technologies involved in Industry 4.0 and Industry 5.0 environments. The diagram includes all the main technologies for control and actuation supporting production auto‐adaption managed by the AI engine.
Figure 1.8 Interoperability of different technologies involved in Industry 4.0 and Industry 5.0 environments.
The technologies of Figure 1.8 are interconnected in information by exchanging data with different software tools. The components of an innovative information system are divided into two main categories:
Transactional applications: systems and procedures supporting IT and enterprise resource planning (ERP) systems.
Decision support system (DSS): systems and procedures supporting strategic decisions and production processes management by data mining.
The ERP systems are developed for finance, logistics and in general for supply chain management. Subsequently ERP systems evolved into material requirements planning (MRP) tools implementing Business Process Re‐engineering (BPR) logics. The peculiarities of ERP systems are:
integrated solutions
functional modules
information centralization
real‐time availability
fixed functional schemes (standard models)
customer customizations
financial compliance
The implementation phases of ERP systems are summarized by the following steps:
Software selection.
formulation of the business process modeling (BPM).
mapping of existing processes (“AS IS” mapping).
definition of a new implementation scenario concerning future production processes (“TO BE” mapping).
BPR process redesign.
Production gap analysis (detailed list of missing production features).
Formalization of formal documents for the validation of new processes.
System parameterization.
Prototype creation.
Testing of the whole information systems.
The main ERP functions are:
management control
management of purchases and receptions
sales and distribution management
warehouse and logistics management
project management
production, orders, and account processing
human resources management
The upgrade of the ERP systems involves AI implementations, thus improving DSS dashboards and decision‐making processes of the whole supply chain. ERP and other software tools must be interconnected in information infrastructure enabling real‐time data processing, sensing, and actuation. According to the architecture of Figure 1.9, information systems interconnect different enabling technologies by means the use of the bus or of the ESB. DBs are synchronized and managed by an advanced ERP having different functions including data migration into the big data system. All the other facilities such as IoT, PLC sensing and actuation, AR and process simulation, are implemented through human–machine interfaces (HMIs) interconnecting the tools with the information system. The information systems are typically managed by a bus or a ESB line.
Figure 1.9 Information system interconnecting enabling technologies.
The bus system is a set of internal connections for the transmission and exchange of signals, supply voltage, and ground potentials. The exchange of signals between the microprocessor and the input and output interfaces therefore takes place via the bus