• Non ci sono risultati.

Functional view of stack components

Nel documento UNIVERSITA’ DEGLI STUDI DI PARMA (pagine 80-83)

3.3 A new RM of I4.0 for SMEs: the RMI4.0

3.3.2 Functional view of stack components

The view of the stack IoT-BD-CPS that this thesis wants to represent is that related to the smartification of the systems towards the realization of the SF. This view leverages some instrument:

• The ‘IoT architecture’ (L. Da Xu et al., 2014). This architecture is composed of four layers:

o the ‘Sensing Layer’ to percept the status of objects and systems and uniquely integrate them, via actuators, sensors, RFID tags and other devices capable of acquire data (e.g. PLC)

o the ‘Network Layer’ that transfers data captured via ‘Sensor Layer’ through wired or wireless network to the next ‘Service Layer’, mapping and connecting objects and enabling their capability of sharing data

o the ‘Service Layer’ that makes use of technologies (i) supporting services and applications (e.g.

data storage, exchanging and management of data) required by the users or applications (e.g.

middleware, platforms), and (ii) routing the interoperability among heterogenous devices o the ‘Interface Layer’ that allows to interconnect and manage objects easily, and to display

information in a clear and comprehensible way for interaction of the user (both machines and humans) with the system.

68

• The ‘Big Data framework classification’. This classification is provided by Al-Gumaei et al. (2019) who have analyzed four frameworks for BD, i.e. the ‘Big Data Taxonomy’63, the approach of Ellingwood (2016), and the ‘Big Data Landscape’64:

1. ‘Data ingestion frameworks’, which deal with transferring raw data from data sources to the big data system and handle format and integration issues

2. ‘Data storage frameworks’, which include distributed file systems and databases that persistently store varieties of big data formats

3. ‘Computation frameworks’, which are capable of (i) processing large datasets and (ii) concurrently routing their elaborations to machines. Elaborations relate to both batch processing of blocks of data and stream processing of continuously processed data

4. ‘Analytics frameworks’, which consists of algorithms and computations used (i) to unlock value from big data and (ii) to make predictions about future trends based on past events

• The ‘5C architecture’ of Lee et al. (2015). It is an extension of the ‘3C architecture’, namely

‘Computation’ ‘Communication’-and ‘Control’ (Ahmadi et al., 2018; Hu et al., 2012). The architecture is composed by five hierarchical functions each one characterized by some attributes of CPSs corresponding to specific technologies to adopt for realizing them:

1. ‘Smart Connection’: is the bottom hierarchical level characterized by the data acquisition through sensor network, controllers, as well as enterprise manufacturing systems. It requires standards and protocols since the variety of data. It relates to the system condition monitoring 2. ‘Conversion’: is the hierarchical level dealing with transforming data into information. It

consists of suitable algorithms, and relates to system self-awareness

3. ‘Cyber’: is the middle layer acting as a central hub, which routes information to every connected system, forming the system network. Digital twining and Analytics (e.g. data mining) are needed for elaboration and synthesis of information gathered. This layer enables the CPSs and allows them to self-comparisons

4. ‘Cognition’: this layer deals with providing users with the proper knowledge about the system acquired, for prioritizing and optimizing decisions

5. ‘Configuration’: this layer realizes the feedback from the cyber space to physical space, and make machines self-configure and self-adaptive

• The ‘Wisdom hierarchy’ (Rowley, 2007). Rowley (2007) ‘Data-Information-Knowledge-Wisdom (DIKW) hierarchy’ (Ackoff, 1989) considering the source specificity of each hierarchy item, and then has made it into the ‘Wisdom hierarchy’ by mapping the ‘DIKW hierarchy’ onto several notable information system hierarchies. Author (Rowley, 2007) refers the D-I-K-W hierarchy to the

‘Transaction Processing ‘Management Information Decision Support

System’-‘Expert System’ hierarchy of derived information system. The information system hierarchy built by Rowley (2007) is described by ISA 95 automation pyramid65 and its characterization through communication networks (Ikram & Thornhill, 2010; Tountopoulos et al., 2018).

o The ‘Transaction Processing System’ refer to the ‘Production Processes’ ISA 95 level. It operates within the ‘Field Network’ in which data are collected from processes running

63 "Big data taxonomy", October 2014. Retrieved in Al-Gumaei et al. (2019). Not founded on the web.

64 Source: Big data landscape 2018. Available from: http://www.qaware.de/fileadmin/user_upload/QAware-Big-Data-Landscape-2018.pdf. Last access: 2020.09.18

65 Retrieved in Åkerman (2018)

69

through sensors, devices producing signals (e.g. RFID), and other field devices, within both wired and wireless networks, and also using ‘Collection’ and ‘Management’ functionality of the Cloud which refers to capturing and aggregation of data, and storage-preservation-access functionalities

o The ‘Management Information System’ refers to the ‘Sensing & Manipulating’ and ‘Monitoring

& Supervising’ ISA 95 level. It realizes control and processing of operations exploiting PLC, SCADA, HMI, and ‘Preparation’ functionalities of the Cloud, namely pre-processing of raw data collected at the previous hierarchy layer

o The ‘Decision Support System’ refers to the ‘Manufacturing Operations Management’ ISA 95 level. It exploits MES for operations management and relates to ‘Processing’ functionality of the Cloud, which uses data analytics, simulation, modelling and related technologies for providing management and operations decision maker with suitable instruments

o The ‘Expert System System’ refers to the ‘Business Planning and Logistics’ level of the ISA 95 pyramid, which leverages ERP systems and Cloud ‘Distribution’ functionality for visualization and representation of the system state addressing business decision-making processes

Then, it is used for linking the ‘Wisdom hierarchy’ and other architectures and frameworks of I4.0. The full combination of structures is provided in the bullet list below. DIKW meanings are provided by Ackoff' definitions (1989) (in italic font), and then are mapped to types of information systems as made by Rowley (2007). The meaning of each element towards I4.0 technology stack is provided from the original works considered:

• “Data are defined as symbols that represent properties of objects, events and their environment. They are the products of observation. But are of no use until they are in a useable (i.e. relevant) form. The difference between data and information is functional, not structural”. Data in the information system hierarchy of Rowley (2007) are contained in the ‘Transaction Processing System’. They are acquired at the ‘Sensing layer’ of IoT by means of sensor belonging to the bottom function ‘Smart Connection’ of CPS, and then are transferred to the ‘Network layer’ of IoT for ‘Data ingestion’ and ‘Data storage’ within BD frameworks.

• “Information is contained in descriptions, answers to questions that begin with such words as who, what, when and how many. Information systems generate, store, retrieve and process data. Information is inferred from data” within the middle architecture functions ‘Conversion’ and ‘Cyber’, since the relationship with the

‘Management Information Systems’ level of Rowley (2007). ‘Conversion and Cyber functionalities’

are realized by means of ‘Service layer’ and ‘Computation frameworks’ of IoT and BD respectively.

• “Knowledge is know-how and is what makes possible the transformation of information into instructions. Knowledge can be obtained either by transmission from another who has it, by instruction, or by extracting it from experience”.

In the derived information system hierarchy (Rowley, 2007), it matches to the ‘Decision Support System’, which in the 5C architecture of CPS is related to the high function ‘Cognition’ realized through ‘Analytics framework’ of BD still within the ‘Service layer’ of IoT architecture.

• Finally, Intelligence and Wisdom belonging to the ‘Expert System’ of information system (Rowley, 2007) are reached, and they refer to the ability of increasing efficiency and effectiveness. “Wisdom adds value, which requires the mental function that we call judgement. The ethical and aesthetic values that this implies are inherent to the actor and are unique and personal”. It matches the higher CPS function ‘Configuration’

realized through ‘Interface layer’ of IoT by means of ‘Analytics frameworks’.

70

Table 3.3 recaps how DIKW hierarchy and architecture of I4.0 are interconnected for achieving wisdom within SFs of I4.0.

Table 3.3 - Mapping I4.0 architecture onto DIKW hierarchy, towards system smartification

DIKW hierarchy

Information system hierarchy

level (Rowley, 2007)

ISA 95 pyramid;

network and technologies; Cloud

functionalities

IoT layers (S. Li, Da Xu, &

Zhao, 2015)

BD frameworks (Al-Gumaei et

al., 2019)

CPS functionality (J. Lee, Bagheri, et

al., 2015)

Data Transaction Processing System

Production Processes; Field network and field devices (generally sensors); Collection

and Management

Sensing Network

Data ingestion

Data storage Smart Connection

Information Management Information System

Sensing & Manipulating and Monitoring &

Supervising; Control and Operations network via PLC, SCADA, HMI;

Preparation Service

Computation Conversion Cyber

Knowledge Decision Support System

Manufacturing Operation Management; Management

network and MES;

Processing Analytics

Cognition

Wisdom Expert System

Business; Business Planning and Logistics via ERP;

Distribution

Interface Configuration

Nel documento UNIVERSITA’ DEGLI STUDI DI PARMA (pagine 80-83)