Artificial intelligence is a science provides great concern for creating machines that can imitate human behavior and taking his intelligence into account. Creating these machines increase with the development of artificial programming technique. In the past time people thought creating these smart machines is impossible and that is a kind science fiction but today the dream becomes realistic when the scientists success in creating those smart machines which they have the ability to imitate entire human characteristics such as: understanding the speech, contact with humans, and responding to the human commands.
Most of the textbook define AI as "the study and design of intelligent agents", and also John McCarthy defines AI as "combination of science and engineering to create intelligent machine". Recent days AI becomes main essential aspect of creating the technological industry, and the main supplier of alternative solutions for most issues in computer science. There are currently no consensuses on how closely the brain should be simulated. Early of twentieth century many researches explored the link between neurology and information theory and cybernetics, W.Grey Walter's turtles and Johns Hopkins beats create machines that used electronic network to exhibit rudimentary intelligence, other researchers are gathered for meetings of technology in Princeton university, they discovered another approach but this approach was largely abandoned.
This picture illustrate AI brain's simulation approach
Although there are some clearly pure applications of AI such as industrial machines, or the IntellipathTM pathology diagnosis system currently approved by the American Medical Association and distributed in hundreds of hospitals worldwide for the most part, AI does not produce stand-alone systems, but it provide additional information and technology to existing applications, databases, and suitable environments factors to make them friendlier, intelligent, and more sensitive to user behavior and changes in their environments (agility manufacturing). The AI amount of an application for example, a logical inference or learning module is generally a huge system, dependent on essential infrastructure. Industrial Research &Development with relatively short time-horizons could not have formal justification work of the kind and volume that has been required to initiate the infrastructure for the civilian and military achievement that AI interested on today. And beyond the thousand of currently distributed applications, ongoing endeavors that draw upon these decades of federally-sponsored fundamental research point towards even more impressive future capabilities:
1. Autonomous vehicles: A DARPA-funded onboard computer system from Carnegie Mellon University drove a van all but 52 of the 2849 miles from Washington DC to San Diego, averaging 63 miles per hour day and night, rain or shine.
2. Computer chess: Deep blue, a chess computer initiated by IBM researchers, defeated world champion Gary Kasparov in a landmark performance.
3. Mathematical theorem proving: A computer system at Argonne National Laboratories proved a long-standing mathematical conjecture about algebra using an approach that would be considered imaginative if done by humans.
4. Scientific classification: A NASA system learned to capture very weak signals as either planets or celestial bodies with high degree accuracy, by studying examples classified by experts.
5. Advanced user interfaces: PEGASUS is a spoken language interface connected to the American Airlines EAASY SABRE reservation system, which allows passengers to acquire flight information and make flight reservations via a large on-line dynamic database that can be treated through their PC over the internet.
3. BACKGROUND
With the tremendous increasing demands for products and services implementation of modern technology become very crucial in order to accommodate markets and customers needs, so entire manufacturing corporate move toward utilizing AI devices to respond to the customer needs and globalize its business. In addition to that customers convert their views from buying products and using services with lower price and bad quality to higher price and good quality products or services, so that means utilizing accurate equipments to ensure quality has been built in our products. On the other hand now a day's most of governmental regulations state huge endeavor must be devoted to the environment this creates major constraints in front of traditional approaches. All of these conditions encourage and illustrate the major role of artificial intelligence application in manufacturing.
4. AI IN MANUFACTURING SYSTEM
4.1 Future of AI in Manufacturing Systems
Artificial Intelligence was seen by many scholars as the solution to all problems pertaining to manufacturing technologies. The mere notion that the human factor which everyone's fingers points at following any problems can be replaced by robust and accurate thinking machines is undoubtedly mind boggling to say the least. The literature has talked this issue intensively in the past and continues to look at improving the application of AI in manufacturing settings. One of the early endeavors into finding applications for AI in manufacturing took place even before personal computers were marketed as discussed by (Biegel, 1986) who discussed the future potential for AI applications in manufacturing and discussed the possibility of applications like: Knowledge Representation, Voice Recognition, Voice Synthesis, Natural Language Understanding, Machine Vision Systems, Pattern Recognition, Robotics, Computer-Aided Instruction, Automatic Programming and Expert Systems. In this section we will discuss some of these applications by comparing the author's initial outlook to the actual implementations in today's manufacturing systems.
4.2 Knowledge Representation
Biegel, in his article (Biegel, 1986), elaborated on this issue of knowledge representation which according to him was a major concern due to number of manufacturers who have many interrelated main-frame databases and fail to connect it to work together. He suggested that AI might be used in the future to not only standardize the format of how data is stored but also to determine what needs to be stored in the first place since this will help in saving the scarce space for storage at that time. We can see today that this approach is widely used in the concept of Relational Databases which effectively chooses what needs to be stored and what doesn't. Although it's not thought of as an application of Artificial Intelligence it indeed is one. This can be improved upon in the future by giving the computers the liberty of choosing where and who should have the permission to view the data which can contribute to the field of rights-management.
4.3 Voice Recognition
Perhaps one of the earliest fields to be looked into in the area of AI applications, yet, although the author claimed that it has a lot of potential due to the fact that at that time it was in the infancy stage it is still to this date not robust enough to be considered seriously in a manufacturing setting. The fact that this is voice recognition as oppose to voice comprehension means that it relies fully on the quality of the input and thus similar sounding words will result in confusion. In our opinion, the correct direction forward is to start in analyzing and understanding the context of the word in the speech so that more accurate results can be obtained. Only then will we have a future where operators talk robots into what they want to get done. Voice synthesis is the process of converting written text into computer voice that can be understood by the human operators. These systems are widely available and rely on programming and software implementations of how word's phonetics should sound.
4.4 Natural Language Understanding
One of the areas that continue to be thought of as pure science fiction is the ability for computers to not only recognize but also understand natural human language. The biggest obstacle to this as discussed by Biegel is that computers rely on syntax to help them understand what they are required to do. This over-reliance on syntax makes it nearly impossible for computers to fully understand what humans want to say. They can however be programmed to recognize certain phrases in instructions but this is the closest mean of approaching the problem at the moment. Nevertheless, through the advances of Neural-Computing and the continues effort to understand how our brains work, there might come a day where we tell our gadgets what to do.
4.5 Machine Vision Systems
Machine Vision is an area that received more actual implementation and gave better results recently thanks to the advances in image processing algorithms and high-resolution cameras. Many manufacturing settings already employ machine vision in quality control, precise cutting equipment and stocking systems. One of the famous examples used every day not only in manufacturing setting but also in our everyday life is Bar-Code readers which facilitated the transfer of data to amazing extent.
5. AI IN MANUFACTURING DESIGN
5.1 AI application in designing mould
The application of artificial intelligence has been applied in manufacturing for these recent years. In the industry, one of the areas that have applied artificial intelligence is design for manufacturing. The intent of this application of AI techniques is to solve some product development problems. The objective is to provide initiatives that promise high levels of success in manufacturing industry. journal of intelligent manufacturing throughout the world.
5.2 AI application in designing mould
There's a lot of design involve in manufacturing for different product, and one of them is mould design. One of the most important phases of mould design is conceptual mould design. By using the application of decision making diagram, it can be the solution of mould design faulty but the application is not user friendly and the selection of single design solution is separated from selection of other without taking in consideration the influence among them. Recently, the application of artificial intelligence has proved that it is a very efficient tool for making high level complexity decisions, such as those in conceptual mould design phase.
The analysis of the decision-making diagrams shows that the designer is expected to answer with YES or NO repeatedly which can be very tiring. Most of the questions can be joined into one general form which the input parameters are entered such as moulding dimensions, thermoplastic and material density in order to reduce the question to a minimum amount. But still, there is the disadvantage of using computer program which is the static knowledge base represented in fact by the decision-making diagrams.
5.3 Expert systems
Artificial intelligence is a very powerful tool and it is suitable to solve this kind of problem by using the VisiRule software based on the programming language Prolog for the development of intelligent applications such as expert systems (ES). Expert systems represent one of the artificial intelligence methods, by creating a computer programs that use the stored human knowledge for problem solving regarding a certain domain in the same way they would be solved by humans.
The ES success criterion is successful solving of everyday, complex problems that require human expertise. Human problem-solving method still represents a big enigma in the field of artificial intelligence. Therefore, numerous methods have been developed which try to simulate it. Although human mental processes are too complex to be described by an algorithm, there is still the majority of experts are capable of expressing their knowledge for problem solving in the form of the so-called "rules". This term in AI is the most often applied type of presenting knowledge to the computer and is defined as "if-then" structure which brings certain information on the IF side of the rule into matching process which cause a rule firing. The rules are relatively simple to be created and understood. The "if" part is called the antecedent, premise and condition, meanwhile the "then" part is called consequence, conclusion and action. The application of one or another mechanism depends on the type of the problem solved by ES. In VisiRule tools, both inference mechanisms are available and for problem solving in this paper the forward chaining mechanism has been applied. The diagram layout is presented based on which the Prolog program code is generated in the background. It also presents a section from the conversation with the user which is the basis for determining the fan gate.
The example of conceptual mould design presents the VisiRule tools which in a simple and graphical way that allow the design and maintenance of the knowledge base. Instead of manually entering, changing, deleting or controlling the relations in the program code of hundreds or thousands of knowledgebase rules, the work of the intelligent system designer is in this case made easier due to the fact that the entire knowledgebase and relations among the facts are presented by graphic layout and the inference mechanism programming code is automatically generated in the background. The described expert system for conceptual mould design contains the knowledgebase which solves the problems of types and positions of gates on the mould for over 20 different ones.
The emergence of artificial intelligence (AI) and expert systems (ES) technology from the laboratory environment into general practice offers a promising approach for the development of intelligent application in DFM and CE. These technical developments would be particularly valuable for design and manufacturing engineers. With them, a product design can be analyzed and evaluated with recommendations based on available knowledge prior to and during product development. In addition, an intelligent system can help to capture and transmit critical manufacturing knowledge from experts, helping to neutralize the effects of losing experienced design and manufacturing personnel.
6. AI IN SIX SIGMA
6.1 Introduction to Six Sigma
Six Sigma at many organizations simply means a measure of quality that strives for near perfection. Six Sigma is a disciplined, data-driven approach and methodology for eliminating defects (driving toward six standard deviations between the mean and the nearest specification limit) in any process from manufacturing to transactional and from product to service.
The statistical representation of Six Sigma describes quantitatively how a process is performing. To achieve Six Sigma, a process must not produce more than 3.4 defects per million opportunities. A Six Sigma defect is defined as anything outside of customer specifications. A Six Sigma opportunity is then the total quantity of chances for a defect. Process sigma can easily be calculated using a Six Sigma calculator.
The fundamental objective of the Six Sigma methodology is the implementation of a measurement-based strategy that focuses on process improvement and variation reduction through the application of Six Sigma improvement projects. This is accomplished through the use of two Six Sigma sub-methodologies: DMAIC and DMADV. The Six Sigma DMAIC process (define, measure, analyze, improve, control) is an improvement system for existing processes falling below specification and looking for incremental improvement. The Six Sigma DMADV process (define, measure, analyze, design, verify) is an improvement system used to develop new processes or products at Six Sigma quality levels. It can also be employed if a current process requires more than just incremental improvement. Both Six Sigma processes are executed by Six Sigma Green Belts and Six Sigma Black Belts, and are overseen by Six Sigma Master Black Belts.
According to the Six Sigma Academy, Black Belts save companies approximately $230,000 per project and can complete four to 6 projects per year. General Electric, one of the most successful companies implementing Six Sigma, has estimated benefits on the order of $10 billion during the first five years of implementation. GE first began Six Sigma in 1995 after Motorola and Allied Signal blazed the Six Sigma trail. Since then, thousands of companies around the world have discovered the far reaching benefits of Six Sigma.
6.2 Six Sigma manufacturing technique pros
There are Six Sigma manufacturing technique pros which, the Six Sigma places a clear focus on achieving measurable and quantifiable financial returns of an organization. A six sigma project is never approved unless the bottom line has been defined. Six sigma utilizes the tools and techniques for fixing problems in business process in a very sequential order. There is a certain amount of discipline to the way that problems are handled. All the tools and techniques have a unique role in where, when, why, and how they are used and applied. This makes all the difference is the success and failure of the program. Since the main focus is systematic, statistical thinking and proven statistical tools and techniques are encouraged. The character of leaders and their passion within the six sigma techniques and tools is greatly valued and extremely important. The commitment from passionate leaders makes this program so successful. When individuals believe and are devoted to a better change, good things happen. The method in which six sigma takes in problem solving brings many different factors into the equation; cultural changes, and customer focus which are two of the main ones. Six Sigma integrates all aspects into the program and creates improvement within the company. Six Sigma creates a foundation and development for master managing. Decisions that are made for Six Sigma techniques and tools are made only with the facts and data rather than just assumptions and hunches. With six sigma techniques and tools measurements are always in place and essentially considered part of the norm. The focus is clearly on achieving measurable returns to the bottom line of an organization. Projects do not get approval unless the bottom line as been defined clearly.
6.3 Six Sigma manufacturing technique cons:
There is Six Sigma manufacturing techniques cons which, there is a challenge largely with always finding good quality data especially in some instances where there is no data available at all. This alone can take much of the time in itself. Prioritization becomes an issue as it is a critical factor in whether or not the six sigma program succeeds or fails. There are few tools available to help with this process so most of the prioritization is based on subjective judgment. There is a large discrepancy between what a statistical error is. Since six sigma is 3.4 defects per million and a defect is described as anything that does not meet a customer's needs. Non standard procedures in the certification process of "black belts", "green belts" etc are inconsistent across different companies. Many consulting firms claim to be experts in six sigma when they do not understand the tools and techniques at all. There are many pros and cons to implementing six sigma tools and techniques in your manufacturing plant, but make sure that you know all there is to know and that it is right for your company before you implement anything.
6.4 A Better Example of Six Sigma Deployment
To stay competitive, top management at another company determines costs must be reduced and customer satisfaction increased. A Six Sigma program is initiated to address these urgent issues and to prepare for future issues. At a briefing, middle managers learn that the major business targets (and big Ys) are to reduce costs and customer complaints at plant A. Local management at plant A is asked to take steps within given budgets to meet the new targets.
After receiving Six Sigma awareness training and participating in workshops to learn basic tools and better understand their jobs within the overall program, the local managers now know that after accepting their business assignments, they must move from the business level into the process level and become process owners seeking process improvement potential. Instead of launching massive improvement projects with broadly stated goals, the new process owners work together in a team to drill down from the original business issues to discover the individual process problems. To do this, they use their dashboard process performance results and other high level data as well as standard Six Sigma tools such as voice of the customer research and SIPOC (suppliers, inputs, process, output and customers) diagrams in a form of pre-analysis.
The pre-analysis identifies a series of little-y issues such as employee discipline slow cycle times, high product defect rates and machine downtimes. These, in turn, are causing plant A's high costs at the big Y level. After prioritizing these problems and identifying the corresponding processes that require immediate attention, individual project charters are drafted to address them. These are then distributed to Black Belt candidates, who are about to begin their training. They are then asked to form a team and formally kick off their assigned projects to validate the root causes for the well-defined y-variables.
Depending on the number of Black Belts available, finding all the potential for achieving the full business goal of reduced costs at plant A could take one or two years. But the project pipeline is now flowing. And a system has been put in place to implement Six Sigma further down in the organization.
7. CONCLUSION
Artificial intelligence is a sub-field of computer science concerned with understanding the nature of intelligence and constructing computer systems capable of intelligent action.
As the manufacturing industry becomes increasingly competitive, manufacturers need to implement sophisticated technology to improve productivity. Artificial intelligence, or AI, can be applied to a variety of systems in manufacturing. It can recognize patterns, plus perform time consuming and mentally challenging or humanly impossible tasks. In manufacturing, it is often applied in the area of constraint based production scheduling and closed loop processing.
AI software uses genetic algorithms to programmatically arrange production schedules for the best possible outcome based on a number of constraints, which are pre-defined by the user. These rule-based programs cycle through thousands of possibilities, until the most optimal schedule is arrived at which best meets all criteria.
