Smart Manufacturing: a Milestone for Industry
Rajesh AngadiProgram Manager at Cisco Systems related
to Data Science, IoT, Big Data Analytics and
Agile Product Development
Cisco Systems

How can advanced technology make manufacturing companies more competitive ? As pressure builds to provide increasingly customized products, companies are investing in simulation, additive manufacturing, autonomous robots, and the data analytics that are needed to capture and make sense of the information these systems generate. The article by Rajesh Angadi, Program Manager at Cisco Systems related to Data Science, IoT, Big Data Analytics and Agile Product Development explains to us how tomorrow's production centers will produce - and benefit from - inexhaustible streams of information. Essentially artificial intelligence-driven self-organizing Internets-of-Things, they will operate holistically and flexibly, allowing their human workers, robot assistants and additive and subtractive manufacturing systems to optimize flows of materials and energy.

Industry 4.0 is for the current trend of automation and data exchange in manufacturing technologies. It includes cyber-physical systems, the Internet of things, cloud computing and cognitive computing. Industry 4.0 is commonly referred to as the fourth industrial revolution. Industry 4.0 creates what has been called a "smart factory". Within the modular structured smart factories, cyber-physical systems monitor physical processes, create a virtual copy of the physical world and make decentralized decisions. Over the Internet of Things, cyber-physical systems communicate and cooperate with each other and with humans in real-time both internally and across organizational services offered and used by participants of the value chain .

First came steam and the first machines that mechanized some of the work our ancestors did. Next was electricity, the assembly line and the birth of mass production. The third era of industry came about with the advent of computers and the beginnings of automation, when robots and machines began to replace human workers on those assembly lines. And now we enter Industry 4.0, in which computers and automation will come together in an entirely new way, with robotics connected remotely to computer systems equipped with machine learning algorithms that can learn and control the robotics with very little input from human operators.

For a factory or system to be considered Industry 4.0, it must include:

Interoperability-machines, devices, sensors and people that connect and communicate with one another. Information transparency-the systems create a virtual copy of the physical world through sensor data in order to contextualize information.

Technical assistance-both the ability of the systems to support humans in making decisions and solving problems and the ability to assist humans with tasks that are too difficult or unsafe for humans.

Decentralized decision-making ability of cyber-physical systems to make simple decisions on their own and become as autonomous as possible.

Industry 4.0 roadmap visualizes every further step on the route towards an entirely digital enterprise. In order to achieve success in the digital transformation process, it is necessary to prepare the technology roadmap in the most accurate way. In today's business, Industry 4.0 is driven by digital transformation in vertical/horizontal value chains and product /service offerings of the companies. The required key technologies for Industry 4.0 transformation such as artificial intelligence, internet of things, machine learning, cloud systems, cyber security, adaptive robotics cause radical changes in the business processes of organizations.

Manufacturing organizations are able to accumulate large amounts of plant floor production and environmental data due to advances in data collection, communications technology, and use of standards. The challenge has shifted from collecting a sufficient amount of data to analyzing and making decisions based on the huge amount of data available. Data analytics(DA) can help understand and gain insights from the big data and in turn help advance towards the vision of smart manufacturing. Modeling and simulation have been used by manufacturers to analyze their operations and support decision making. Manufacturing has always had Big Data. We have been collecting data with historians, MES (Manufacturing Execution systems) for decades. It is just a new buzz word for the marketers. Manufacturing is an untapped market for Big Data. There is lots of data, lots of different types of data, and hardly any of it is being used for analysis today. It is not uncommon in manufacturing to hear of Smart Connected Assets like jet engines producing petabytes of data each flight or Manufacturing Execution Systems(MES) collecting millions of process variable measurements from the plant each shift; however, running reports on large data sets does not qualify as Big Data analytics in manufacturing.

Smart manufacturing system requires capabilities and technologies for designing and improving the overall system performance through diagnostic and prognostic assessment based on (big) data analytics. The right insight derived from big data could lead to right actions for enhancing sustainability, productivity, flexibility, and competitive advantages and enabling sustainable and agile manufacturing. The contributions from different engineering disciplines to bring out common issues and specific challenges that address predictive modeling for Smart Manufacturing Systems. An initial set of topics includes (but is not limited to) manufacturing Key Performance Indicators (KPIs), Methods, Technologies, and IT Infrastructure for Big Data and Data Analytics; Standards and Protocols; and Business Best Practices for Smart Manufacturing Systems. Manufacturers have been collecting and storing data for years, but now big data technologies enable more constructive use of this information, including how to increase throughput, boost yields, improve efficiency, and reduce downtime. Big data is characterized by huge data sets and varied data types (e.g., images, text , and machine log files), which the production line is producing at a much faster rate than ever before. When this data is analyzed using new tools available in the market, manufacturers can gain valuable insights derived from finding patterns, extracting meaning, and ultimately making decisions that lead to greater efficiency. Big data analytics, Cloud and Internet of Things (IoT) technologies are the substantive foundation that enables advanced levels of smart manufacturing performance. That's why companies are working hard to drive a data revolution in smart manufacturing that will yield new productivity and efficiency gains. IoT Solution Blueprint is easier than ever to learn how big data analytics applied to factory equipment and sensors can bring operational efficiencies and cost savings to manufacturing automation processes. However, many machine tools operate in relative silos, so it is a major challenge to collect, analyze, and act on data generated across the factory floor. This is why companies have assembled various Cloud, Internet of Things and big data technologies that provide the connectivity, security, interoperability, and analytics capabilities that enable higher-performance manufacturing automation.

Big Data Analytics in manufacturing is about using a common data model to combine structured business system data like inventory transactions and financial transactions with structured operational system data like alarms, process parameters, and quality events, with unstructured internal and external data like customer, supplier, Web, and machine data to uncover new insights through advanced analytical tools. This definition of Big Data Analytics differs from the traditional approach most manufacturers and vendors have taken to dealing with manufacturing data. In most cases, manufacturers have invested heavily in data collection and visibility, mainly through legacy MES, EMI (Enterprise Manufacturing Intelligence), and Data Historians.

Big Data analytics promise is all about plucking insights, knowledge and trends from deep reservoirs of raw data. But the reality is that massive troves of data are difficult for commercial and institutional computer systems to collect, sort, manage and analyze in a cost effective and speedy way. Curating and "cleaning" data to improve quality requires access to powerful computing power. Data analytics infrastructures entail complex techniques such as machine learning, visualization and cloud computing - to name a few. Businesses, government and other organizations looking to big data analytics to gain insights and solve challenges typically do not have the core competencies required to process and analyze giant databases.PARC, a Xerox company, (R & D services), Cisco (Networking and IT leader), Hitachi Data Systems (big data storage) and Quantiply (enterprise architecture) have created the Big Data Foundry (BDF) that pools together experts in software and hardware who can help enterprise harness data in a more powerful and efficient manner. The foundry offers a Big Data Rapid Experimentation Platform made up of readily available technology and an "algorithmic toolbox" designed to accelerate experimentation and reduce risk. Our mission is to help organizations develop innovative data driven methods to find answers to business problems and develop models that focus on solutions that can make a difference.

Rapid Experimentation Cloud
Big Data Foundry brings collective expertise and technologies to help you experiment and scale new ideas with reduced risk and time-to-value. We provide all the tools and technologies you need to connect, explore, experiment, and validate your hypotheses and problems for smart manufacturing.
  • Reference architectures, data scientists and tools that reduce technological complexity and deployment costs.
  • Pre-assembled reference architectures to validate your business solutions on target reference architectures before investing in these complex tools and technologies.
  • Support development of data driven infrastructure to empower your employees to sense, analyze, interpret and act fast on the data insights .
Data Sciences Innovation
Big Data Foundry member companies collaborate to help you engage worldclass data scientists, data engineers, cloud experts, innovators, and ethnographers all in one place to help you find new routes to revenue and profitability in your organization. With Big Data Foundry as a strategic partner, you no longer need to delay exploring and implementing new Big Data technologies.

Internet of Things (IoT) which help for realtime visibility and business intelligence is well within the reach. Companies around the world are adopting new plant architectures which takes advantage of the pervasive connectivity and actionable data provided by the IoT. By converging factory -based operational technologies with global IT networks, companies are increasing uptime and improving operational equipment effectiveness intelligence - from the plant floor through the supply chain which means less downtime, higher productivity, greater resiliency, and the agility to respond to rapidly changing customer and market needs.

Smart Manufacturing Operations offering, we can:
Optimize your equipment for greatest profit- Smart Manufacturing Operations offering enables the operation of plant equipment to its maximum capacity, increasing revenues and profits.

Shift from reactive to predictive maintenance-Using machine-to-machine data, Smart Manufacturing Operations maintains a real-time picture of operating assets versus capacity and service needs. You can operate equipment at full capacity to maximize revenues, while scheduling service for the least production impact. You can shift production work between plants in your portfolio by viewing plant capacity across your company.

Accommodate critical orders- Our solution offers plant managers a window into production-line capacities to schedule new orders at a plant or plant portfolio without interrupting the flow of work in process. This means you will be better prepared to meet the expectations of key customers for elevated levels of service and responsiveness. Operate at peak energy efficiency- Smart Manufacturing Operations enables plant managers to directly track energy efficiency, sustainability and environmental compliance. Upper-limit alerts give managers time to minimize energy costs and avoid environmental, health and safety fines.

Business success is widely attributed to the use of advanced analytics. Smart Manufacturing Operations is a powerful business information and analytics solution that uses data from plant and corporate systems and external sources to create operational insights presented through specialized, preconfigured and userconfigurable dashboards and reports, giving plant managers the information they need to optimize many parameters essential to meeting demand and reducing cost.

SMART MANUFACTURING OPERATIONS STRATEGY
Operations Strategy will work with you to develop a clear strategy, architecture and roadmap to guide your transformation.

It will specifically address critical manufacturing functions, including:
  • Operating equipment effectiveness
  • Supply chain and inventory
  • Energy utilization and efficiency
  • Health and safety
  • Environmental compliance
  • Financial performance
SMART MANUFACTURING ANALYTICS IMPLEMENTATION
To transform your operation, it will:
  • Tailor and extend our predesigned analytics models for your operation
  • Source and prepare your data for the analytics models
  • Upgrade or install and integrate big data and business intelligence tools with plant and corporate data sources such as ERP, plant historians and laboratory systems
  • Tailor and extend preconfigured reports and dashboards, and put them into production
Emerging Technology as Artificial Intelligence
Artificial Intelligence (AI) is the science concerned with the creation of machine intelligence which is able to perform tasks, only performed by people. Much of this machine intelligence is symbolic and heuristic. Artificial intelligence (AI) is intelligence exhibited by machines. In computer science, an ideal "intelligent" machine is a flexible rational agent that perceives its environment and takes actions that maximize its chance of success at some goal. Colloquially, the term "artificial intelligence" is applied when a machine mimics "cognitive" functions that humans associate with other human minds, such as "learning" and "problem solving. AI research is divided into subfields that focus on specific problems or on specific approaches or on the use of a particular tool or towards satisfying particular applications. The central problems (or goals) of AI research include reasoning, knowledge, planning, learning, natural language processing(communication), perception and the ability to move and manipulate objects. Approaches include statistical methods, computational intelligence, soft computing (e.g. machine learning), and traditional symbolic AI. Many tools are used in AI, including versions of search and mathematical optimization, logic, methods based on probability and economics . The AI field draws upon computer science, mathematics, psychology, linguistics, philosophy, neuroscience and artificial psychology.

Demands towards Smart Manufacturing Systems in an enterprise are as follows:
  • A possibility of collecting and processing different types of information from all sources, both internal and external, in order to acquire and model knowledge necessary to make decisions at all levels of decision process in an enterprise. At the same time a possibility of modeling knowledge and processes, based on human thinking, is required .
  • In a decision process, at decision selecting, the decision maker's subjective evaluation based on his experience and intuition should be taken into account in IMS.
  • There should be a possibility of preliminary information handling and analysis with analytical methods as well as modern artificial intelligence technologies.
  • A possibility of detecting emergency and critical situations and of prompt reaction to them. There must be a possibility of situational data analysis in real time, necessary in an emergency inside the production system or in its surroundings.
  • A possibility to allow for complexity and comprehensiveness of decision-making issues in strategic management support.
  • Taking into account the lack of stability and change dynamics, both in the surroundings and inside the enterprise, the IMSs under design should have the capability for learning from experience and adapting the experience to intensive alteration of working conditions.
  • In intelligent manufacturing systems, the following selected contemporary methods and techniques of knowledge and decision process modeling should be integrated:
  • Artificial neural networks - the most fascinating tool of artificial intelligence, capable of modeling extremely complex functions and, to some extent, copying the learning activity in the human brain.
  • Fuzzy logic - technologies and methods of natural language formalization, linguistic and quality knowledge processing and fuzzification.
  • Genetic algorithms and methods of evolutionary modeling - learning algorithms based on theoretical achievements of the theory of evolution, enriching the artificial intelligence techniques.
The combination of these tools, in which knowledge is represented symbolically, with the traditional expert system will make it possible to create complex programmatic tools for solving difficult decision-making problems at each stage of enterprise functioning.

Artificial Intelligence and Control Engineering
Artificial intelligence (AI) relates to control engineering is when embedded software helps with situational awareness. The software algorithm looks at feedback from a situation, then actuates the decision (closedloop control) without human consultation, or the software recommends a course of action with human consultation (open-loop control). Control engineering or control systems engineering is the engineering discipline that applies control theory to design systems with desired behaviors.

The practice uses sensors to measure the output performance of the device being controlled and those measurements can be used to give feedback to the input actuators that can make corrections toward desired performance. Control engineering is the engineering discipline that focuses on the modeling of a diverse range of dynamic systems (e.g. mechanical systems) and the design of controllers that will cause these systems to behave in the desired manner. Although such controllers need not be electrical like many are and hence control engineering is often viewed as a subfield of electrical engineering.

There are two major divisions in control theory, namely, classical and modern, which have direct implications over the control engineering applications. The scope of classical control theory is limited to single -input and single-output (SISO) system design, except when analyzing for disturbance rejection using a second input. The system analysis is carried out in the time domain using differential equations, in the complex-s domain with the Laplace transform while modern control theory is carried out in the state space, and can deal with multipleinput and multiple-output (MIMO) systems. This overcomes the limitations of classical control theory in more sophisticated design problems. Control engineering was all about continuous systems. Development of computer control tools posed a requirement of discrete control system engineering because the communications between the computer-based digital controller and the physical system are governed by a computer clock. The equivalent to Laplace transform in the discrete domain is the Z-transform. Today, many of the control systems are computer controlled and they consist of both digital and analog components.

Therefore, at the design stage either digital components are mapped into the continuous domain and the design is carried out in the continuous domain, or analog components are mapped into discrete domain and design is carried out there.

In manufacturing, a machine running a webbased process may have similar situational awareness. There may be a perfectly good reason to leave the machine running when the last material runs through the rollers and an operator is standing in a certain location. If the machine is unattended at that particular moment, embedded code may begin an orderly shutdown as the best response. Control Engineering relates to the next big thing (TNBT) which is the second generation of smartphones, which have the software capacity to provide situational awareness. TNBT devices will be able to recognize what is going on inside your area or site and determine when something is out of normal but not yet in alarm. Information for this awareness may come from traditional fixed sensors or even by listening for sound patterns such as hisses, whistles, clangs, and bangs. TNBT devices will become true operator assistants; always watching and always listening for out-of-normal conditions or for events that require manual intervention .

Seven Artificial Intelligence (AI) tools have proved to be useful with sensor systems (part of IoT's): Knowledge-based systems, fuzzy logic, automatic knowledge acquisition, neural networks, genetic algorithms, casebased reasoning, and ambient-intelligence. Applications of these tools within sensor systems have become more widespread due to the power and affordability of present-day computers. The appropriate deployment of the new AI tools will contribute to the creation of more competitive sensor systems and applications.

Artificial Intelligence (AI) helps computing in four ways:
  • 1. Natural language understanding to improve communication.
  • 2. Machine reasoning to provide inference, theorem-proving, cooperation, and relevant solutions.
  • 3. Knowledge representation for perception, path planning, modeling, and problem solving.
  • 4. Knowledge acquisition using sensors to learn automatically for navigation and problem solving
  • 5. Artificial intelligence's ability to function as a safety measure and provide another set of eyes, so to speak, can be extremely beneficial to worker safety in manufacturing. It can also enhance our ability to understand what's happening around us and offer solutions that might not be readily available
Will artificial intelligence (AI) take control of human race during Smart Manufacturing?
The answer to this question seems to be positive. Several experts of AI have similar comment as "everything that humans can do machines can do". Stephen Hawking also warned us during an interview with BBC that "The development of full artificial intelligence could spell the end of the human race." Ex Machina, a recent enthralling science-fiction film presents the possibility of a robot that has cognitive capability to think, feel and even manipulate human beings. Self-driving cars, Siri on your iPhone, weather forecasts, face recognition on your Facebook photos, etc are all examples. A Japanese company with Deep Knowledge found out an Artificial Intelligence(AI) as one of the directors due to its ability to predict market trend that is "not immediately obvious to humans". Replacing human with robot in manufacturing is a trend that we can't stop or avoid. As technology advances, the low cost , high-accuracy and efficiency of robot is going to benefit the human society as a whole on a broader level.