All tables
Comparisons
The 14 side-by-side tables the decks actually set up — nothing invented. Where a deck defines two things separately rather than tabulating them, the note says so and cites the slide each cell comes from.
Industry 1.0 → 2.0 → 3.0 → 4.0
class 1 · slides 4-7The four industrial revolutions on the axes Class 1 slides 4–7 actually use: dates, name, energy, and what changed.
| Industry 1.0 | Industry 2.0 | Industry 3.0 | Industry 4.0 | |
|---|---|---|---|---|
| Dates | 1760–1840 | 1830–1947 | 1947–2015 | 2015 – |
| Name | Mechanization | Mass production / technology revolution | Digitization and automation (digital revolution) | Convergence of digital, biological and physical innovations |
| Energy | Steam and internal combustion engine; coal-powered factory system | Electricity and petroleum | Nuclear and renewable explored | (not stated) |
| Materials | Iron and steel for machinery | Stainless steel, rare earth metals, plastics | Semiconductors | (not stated) |
| What replaced what | Machines and tools replaced animals and human labour | Assembly line and mass production in factories | Mechanical and analogue technology replaced by digital | Interconnectivity and advanced automation |
| Transport & comms | Canals and roads | Automobile, telegraph, telephone, radio; railways | Telecommunication, computers, Internet and WWW | M2M (machine to machine) communication |
| Signature tools | Spinning wheel mechanisation (8× output) | Spinning jenny, power loom | SCADA, PLC, CAD; robots and PLCs | IoT, AI, big data analytics, advanced robotics, cloud |
| Signature examples | The steam engine | Textile industry take-off | Factory automation | CRISPR, self-driving vehicles, 3D printing, digital twins |
| Where it started | Great Britain | (not stated) | (not stated) | (not stated) |
Conventional automation vs Industry 4.0 automation
class 3 · slides 2-5The deck's central contrast: slide 3's four characteristics against slides 4–5's six, matched up.
| Conventional automation | Industry 4.0 automation | |
|---|---|---|
| Definition | Machines, robotics and control systems performing tasks with minimal human intervention | Digital technologies, connectivity and data analytics creating smart, flexible, interconnected systems |
| Scope of a task | Task-specific: programmed to perform specific tasks repeatedly (assembly line, material handling) | Customization and flexibility (modularity): highly customized products through flexible processes that adapt to demand |
| Connectivity | Limited: systems work in isolation, limited communication with other machines | Cyber-physical systems: machines, devices and sensors connected to digital systems for real-time data exchange |
| Decision-making | Centralized: by human operators or a central control system, on pre-programmed instructions or simple rule-based logic | Decentralized: intelligent systems and algorithms let machines decide autonomously from real-time data |
| Use of data | Limited: data may be collected but is not extensively used for real-time monitoring, analysis or optimization | Data-driven insights (real-time capability): advanced analytics and ML generate actionable insights, optimize, predict maintenance |
| The virtual copy | — | Internet of Things (virtualization): a virtual copy of the smart factory from sensor data plus virtual plant and simulation models |
| Role of people | Replaces manual labour | Enhanced human-machine interaction (service orientation): humans do complex decision-making, creativity, problem-solving; machines do repetitive, data-intensive work |
| Technologies | Machines, robotics, control systems | CPS, IoT, cloud computing, AI, big data analytics, machine learning |
AI vs machine learning vs deep learning
class 4 · slides 13, 17-20The four-slide table from Class 4, condensed to its axes.
| Artificial Intelligence | Machine Learning | Deep Learning | |
|---|---|---|---|
| What it is | The study/process which enables machines to mimic human behaviour through a particular algorithm | The study that uses statistical methods enabling machines to improve with experience | The study that uses neural networks, similar to neurons in the human brain, to imitate functionality like a human brain |
| One-line definition (slide 13) | A program that can sense, reason, act, and adapt | Algorithms whose performance improves as they are exposed to more data over time | Multilayered neural networks learning from vast amounts of data |
| Relationship | The broader family, consisting of ML and DL as its components | A subset of AI | A subset of AI (and of ML) |
| As an algorithm | A computer algorithm which exhibits intelligence through decision making | An AI algorithm which allows a system to learn from data | An ML algorithm using deep (more than one layer) neural networks to analyse data and provide output |
| The maths | Search trees and much complex math | Clear idea of the logic; visualize complex functionalities like K-Means, support vector machines | Know the math but not the features; break complex functionalities into linear/lower-dimension features by adding more layers |
| Aim | Increase chances of success, not accuracy | Increase accuracy, not caring much about the success ratio | Highest rank in accuracy when trained with large amounts of data |
| Categories | ANI, AGI, ASI | Supervised, unsupervised, reinforcement learning | Unsupervised pre-trained networks, convolutional, recurrent, recursive neural networks |
| Efficiency | The efficiency provided by ML and DL respectively | Less efficient than DL; cannot work for longer dimensions or higher amounts of data | More powerful than ML; works easily for larger sets of data |
| Examples | Google's AI-powered predictions; ridesharing apps like Uber and Lyft; commercial flights using AI autopilot | Virtual personal assistants — Siri, Alexa, Google; email spam and malware filtering | Sentiment-based news aggregation; image analysis and caption generation |
Weak vs strong vs super AI
class 4 · slides 9-10, 18Three types of AI on a capability ladder, with the formal names slide 18 gives them.
| Weak AI | Strong AI | Super AI | |
|---|---|---|---|
| Capability | Focuses on one task and cannot perform beyond its limitations | Can understand and learn any intellectual task that a human being can | Surpasses human intelligence and can perform any task better than a human |
| Status | Common in our daily lives | Researchers are striving to reach it | Still a concept |
| Formal name (slide 18) | Artificial Narrow Intelligence (ANI) | Artificial General Intelligence (AGI) | Artificial Super Intelligence (ASI) |
| Examples | Virtual assistants such as Siri and Alexa; recommendation engines used by Netflix and Amazon; fraud detection software used by financial institutions | Self-driving cars; pattern and image recognition | — (none given, since it does not exist) |
Data vs big data
class 5 · slides 3-5The slide-3 definitions side by side, with the size test from slides 4 and 5.
| Data | Big data | |
|---|---|---|
| Definition | The quantities, characters or symbols on which operations are performed by a computer, which may be stored and transmitted and recorded in the form of electrical signals | Also data, but with a huge size: a collection huge in volume and growing exponentially with time |
| Size | Whatever the system handles | So large and complex that traditional data management tools cannot store or process it efficiently |
| Time | — | Cannot be stored and processed using the traditional computing approach within a given time frame |
| What it reveals | — | Patterns, trends and association, especially relating to human behavior and interaction |
| Tools | Conventional data processing techniques | Beyond conventional techniques; includes data mining, storage, analysis, sharing and visualization |
| Test | — | Relative to the system: a 100 MB attachment the email system will not carry is big data with respect to email |
Traditional data vs big data
class 5 · slides 20The slide-20 table, which is where the concrete volume ranges live.
| Traditional data | Big data | |
|---|---|---|
| Where generated | At enterprise level | Outside the enterprise level |
| Volume | Gigabytes to terabytes | Petabytes to zettabytes or exabytes |
| Source and management | Source is centralized; managed in centralized form | Source is distributed; managed in distributed form |
| Tools required | Traditional database tools for any database operation | Special kinds of database tools for any database schema-based operation |
| Data sources | ERP, SQL, transaction data, CRM transaction data, financial data, organizational data, web transaction data | Social media, device data, sensor data, video, images, audio |
Embedded system vs cyber-physical system
class 6 · slides 15, 8-10The slide-15 table, plus the distinctions slides 8–10 draw in prose.
| Embedded Systems | CPS | |
|---|---|---|
| What it is | Devices having information processing systems embedded into them | A complete system having physical components and software |
| Extent | Typically confined to a single device | A networked set of embedded systems |
| Resources | Limited resources for performing a limited number of tasks | Not resource constrained |
| Main issues | Real-time response and reliability | Timing and concurrency |
| Emphasis (slide 8) | The computational component | Communications and physical domains as well as computational |
| Design (slide 9) | Designed as stand-alone devices | Focus is on networking several devices |
| Data exchange (slide 10) | — | The most important feature: a CPS is an embedded system able to send and receive data over a network |
| Containment (slide 8) | Contained within every CPS | All CPS contain embedded systems |
5C vs RAMI 4.0 vs IIRA
class 6 · slides 21, 29-30, 33-34, 37-38The three CPS reference architectures on the axes slides 37–38 use: what each is based on, what it targets, and where it is used.
| 5C Architecture | RAMI 4.0 | IIRA | |
|---|---|---|---|
| Full name | The five-level CPS architecture | Reference Architecture Model for Industry 4.0 | Industrial Internet Reference Architecture |
| Created by | Based on automation processes models | Platform Industrie 4.0 | Industrial Internet Consortium (IIC) |
| Based on | Automation process models | SGAM | ISO/IEC/IEEE 42010 |
| Structure | Five levels: smart connection, data-to-information conversion, cybernetic, cognition, configuration | A three-dimensional map: hierarchy levels, product life-cycle, architecture layers | Four viewpoints: business, usage, implementation, functional |
| Focus | Assets' data acquisition and processing | Manufacturing plant operation, integrating the company's value chain | IIoT systems concerns in all sectors |
| Emphasis | Data acquisition for industrial devices | Horizontal and vertical integration within a factory; a Service Oriented Architecture | Interoperability among industries |
| Commonly used in | Embedded systems and small industrial environments | German manufacturing / the industrial community | All sectors / the industrial community |
| Signature concept | The five "C" levels; machines become self-aware, self-comparing, self-adaptive | Industry 4.0 Components (I4.0C) with an Administration Shell — digital twins — under a Superior System Administration Shell | Five functional domains plus crosscutting functions and system characteristics under trustworthiness |
Public vs private vs community vs hybrid cloud
class 7 · slides 29-30, 28The slide-29 table across seven factors, with the slide-30 definitions on top.
| Public Cloud | Private Cloud | Community Cloud | Hybrid Cloud | |
|---|---|---|---|---|
| Who can access it | The general public | Within an organization | A group of organizations | A mixture of public and private |
| Setup and Use | Handle In-house | Requires IT Professionals | Requires IT Professionals | Requires IT Professionals |
| Privacy and Security | Low | High | Medium | Varies from low to high |
| Control of Data | Low | High | Medium | Medium |
| Overall Reliability | Medium | High | Medium | Medium to high |
| Flexibility and Scalability | High | Medium (stable capacity) | Medium (stable capacity) | Very high |
| Cost | Lowest | Relatively high | Variable | Medium / Variable |
| Hardware | Third-party | Variable (can be on-site or third-party) | Variable | Medium / Variable |
| How work is split | — | — | — | Critical activities on the private cloud, non-critical on the public cloud |
Cloud computing vs distributed computing
class 7 · slides 20The slide-20 table, unchanged.
| Cloud computing | Distributed computing | |
|---|---|---|
| What it provides | On-demand IT services | Solving a problem over distributed autonomous computers |
| Number of types | Four: public, private, community, hybrid | Three: distributed computing systems, distributed information systems, distributed pervasive systems |
| Delivery | Delivers hosted services over the internet to its users/customers | Allows many computers to communicate and work to solve a single problem |
| What it gives you | Services such as hardware, software, networking resources | Computational tasks achieved faster than using a single computer, which takes a lot of time |
Cloud computing vs grid computing
class 7 · slides 21The slide-21 table: four rows, and the last one is the summary.
| Cloud computing | Grid computing | |
|---|---|---|
| Flexibility | More flexible compared to grid computing | Less flexible compared to cloud computing |
| Payment | The users pay for what they use (pay-as-you-go model) | The user does not have to pay for any usage |
| Scalability | Highly scalable | Less scalable |
| Orientation | Service-oriented | Application-oriented |
Cluster vs grid vs distributed computing
class 7 · slides 4-9, 20The deck defines these three across slides 4–9 and 20 rather than tabulating them together; this table assembles the axes each definition actually gives. Every cell is from a slide — where the deck gives nothing on an axis, the cell says so.
| Cluster computing | Grid computing | Distributed computing | |
|---|---|---|---|
| Definition | A group of linked computers working together closely so that in many respects they form a single computer | Middleware coordinating distinct IT resources over the network so they function and work as a virtual whole | Solving a problem over distributed autonomous computers |
| Network | Commonly, but not always, fast local area networks | LAN or WAN | Autonomous computers communicating |
| Analogy / appearance | Forms a single computer | Like an electrical grid; appears as a super virtual computer | Many computers communicating to solve a single problem |
| Why deploy it | Improve performance and/or availability, more cost-effectively than a single computer of comparable speed | Give users access to the resources they need when they need them; remote access to IT assets; aggregate processing power | Achieve computational tasks faster than a single computer, which takes a lot of time |
| Types | (not enumerated) | (not enumerated) | Three: distributed computing systems, distributed information systems, distributed pervasive systems |
| Advantages | Increasing speed/better performance; optimized resources utilization; can execute large applications | Isolated access to IT resources; building up processing control | Faster computational tasks |
| Disadvantages | Complex programming models; difficult for debug and development | (not enumerated) | (not enumerated) |
| Controls | (not stated) | Two factors: allocation and trust | (not stated) |
Edge vs fog vs cloud computing
class 7 · slides 48, 41-42The slide-48 table — the single most examinable table in the course. It separates the three by where processing happens, how much power there is, and what it is for.
| Edge Computing | Fog Computing | Cloud Computing | |
|---|---|---|---|
| Location of data processing | Data processing takes place at the edge of the network | Edge computing tasks related to LAN hardware are moved further away from the sensors using fog computing | Information is processed on the cloud server |
| Processing power & storage capabilities | Storage capacity and processing power is limited for IoT devices and sensors | Limited storage capacity and processing power | High-level and extremely powerful processing technology, and can store more data |
| Purpose | Quick analysis and real-time response | Quick analysis and real-time response | Well-suited to long-term, in-depth data analysis and storage |
IaaS vs PaaS vs SaaS
class 7 · slides 31-33The three service models, drawing on Class 7 slides 31–33 and the Class 2 slide-13 table.
| IaaS | PaaS | SaaS | |
|---|---|---|---|
| Full name | Infrastructure as a Service | Platform as a Service | Software as a Service |
| What it provides | Access to fundamental resources such as physical machines, virtual machines, virtual storage | The runtime environment for applications, development and deployment tools | Software applications as a service to end users |
| Verb (slide 33) | Host | Build | Consume |
| You manage (slide 32) | Applications, data, runtime, middleware, O/S | Applications and data | Nothing — the provider manages all nine layers |
| Provider manages (slide 32) | Virtualization, servers, storage, networking | Runtime, middleware, O/S, virtualization, servers, storage, networking | Everything |
| User (Class 2 slide 13) | System admins | Developers | End customers |
| Audience (slide 33) | System administrators and network architects | Software developers | Users |
| Examples | Amazon Web Service; AT&T, Rackspace cloud | App Engine, Azure; Google App Engine, force.com, Windows Azure | Google Docs, SalesForce; Gmail, Microsoft Online Services, Facebook |