|
|
|
|
|
  
|
MACHINE INTELLIGENCE TECHNOLOGIES FOR HEAVY INDUSTRY
|
||
Machine Intelligence
|
Background Automate or emigrate have become the practices of North American heavy industry during the 1990's in order to compete with the inexpensive labor, rich resource base and low regulatory environments in Asia and South America. To automate means to develop human/machine systems with specialized problem-solving capabilities. It also means to capture human expertise and knowledge in a computer. Automation technologies have always played an important role in increasing the competitiveness of heavy industry. In the 1960's, on-line computer systems were introduced to centralize plant process control and information flow. In the 1970's and 1980's microprocessor technology brought distributed computing and cost effective redundancy to computerized control systems. Microprocessor-based programmable logic controllers replaced most of the analog logic systems and allowed introduction of advanced, modern control theory in industrial practices. As a result, in the 1990's, computer-based control systems are common in the industry. Today's computer-based process control systems acquire data, and perform complex analysis and control actions based on process models or statistical process control. The systems are networked to provide specific information at various technical and managerial levels. During the 1990's, in the search for further improvements in production and scheduling methods, the industry is increasingly seeking computer-assisted management of human assets including management of production expertise, staff training, etc. Many companies have recognized that their competitive advantages are no longer based on production technology, capital or labor costs. Instead, their competitive advantages are based on the knowledge of their employees, i.e. on employees' ability to process information and data in all aspects of production and business including marketing, scheduling, and control of production processes. A key computer technology applied to management of knowledge and human resources has been Artificial Intelligence (AI), specifically its application fields of expert systems, machine learning, neural networks, and fuzzy logic.
Rationale for Intelligent Software Systems in Heavy Industry Conventional control and automation helps prevent costly shutdowns, increase yield and quality, and result in operating the process and equipment more effectively and efficiently. However, automation of some production aspects is often not easy because plant operations exhibit complex interactions under rapidly changing circumstances. Many of the plant operations can not be mimicked by simple models or are not well understood. The actions of operators are ambiguous, imprecise, and based on experience and intuitive knowledge that can not be programmed or modeled using mathematical methods or simple algorithms. All of these factors make the use of conventional computer technologies difficult and often ineffective for automation of production processes. To derive better performance from computers and to supplement the conventional control technologies with more advanced features, the industry has turned, therefore, to artificial intelligence methods which can facilitate the management of information, knowledge and production decisions. Artificial Intelligence (AI) is a branch of computer science that studies human-like intelligence and capabilities of reasoning. They offer a number of benefits for automating production processes and procedures. They can:
They can be viewed as:
All three views of AI technologies are correct, yet they also limit our understanding of intelligent software software' capabilities. For example, one of the key motivations behind the use of intelligent technologies is the fact that they can deal with human cognitive limitations, i.e., human failure to monitor all information, to resolve complex and conflicting situations, to identify high-revenue opportunities or to prevent high-cost mistakes. Another motivation behind the use of intelligent software software is the increasing need for higher quality and greater availability of production expertise to deal with the increased complexity of production problems. intelligent software software assist in training novice operators, provide decision support, or mimic human reasoning in narrowly defined domains. They have the ability to keep track of thousands of pieces of information while commanding the knowledge of several experts.
Application Types and Situations Intelligent software systems play a number of roles in heavy industry. Selected examples are discussed below. Process Control: These tasks usually involve automation of low-level control (loop control) in a real-time system. The implemented systems are concerned with fault detection, diagnosis and alarming, and with operating the control devices in the control loops. Integral functions of intelligent software software are sensor diagnostics, handling of erroneous or missing data, and performing temporal reasoning. Real-time AI control systems acquire, analyze and display data intelligently; make inferences and reason about the process; and implement complex control decisions in real time. Process Monitoring: Application of advanced process control in heavy industry has brought better optimization, stability and control quality of industrial processes. Yet some processes are too complex or are not sufficiently understood to implement conventional, advanced control methods. intelligent software software result in better formalization of domain knowledge and better understanding of tasks and processes. They monitor, compare and analyze the process behavior of events that are crucial to successful operation and suggest any corrective action that should be implemented by the operators. Scheduling and Planning: More varied orders, just-in-time manufacturing, and the desire to ensure delivery times and good end-product quality are some factors that have increased the need for operational flexibility and underscore the importance of scheduling and planning in the industry. intelligent software software offer several advantages in developing computerized scheduling systems. Instead of presenting one optimization schedule, AI-based scheduling systems present several schedules with their evaluation indexes. The operator can then select the "best" optimum schedule. Fault Diagnosis and Maintenance: Fault diagnosis and maintenance constitute a significant fraction of operating cost. The problem of fault diagnostics is compounded by the fact that good diagnostic expertise is rare and not available 24 hours a day. Artificial Intelligence systems offer a number of advantages for working with diagnostic problems. First, they can monitor and analyze hundreds of sensors, determine any anomalies in their functions and identify probable causes of the discrepancies between expected and actual operating conditions. They can incorporate the expertise of the best maintenance personnel, information from equipment manuals and previous records of equipment/process failures. The systems can be available to the operators and maintenance personnel at all times and at the right locations.
Performance and Benefits Productivity (Control): Improved stability of the process, quick adaptation to process changes and feed variation, and improved control of plant disturbances are the most frequent benefits reported from the implementation of intelligent software software. Other reported benefits are elimination of the effects of shift changes, reduced duration of work disruptions, and increased efficiency of materials flow and utilization. The value of these benefits depends on the case or situation, but good indications of typical gains in productivity are reported paybacks of several months to a year. Productivity (Monitoring & Diagnosis): The most frequently reported benefits from the implementation of intelligent software software for process diagnosis are improved, more consistent decisions by non-experts, decreased quality variation in the product, expertise being kept in the company, and reduced downtime and repair time, the net result being higher productivity. Productivity (Scheduling): Increased productivity due to improved staffing scheduling; the ability to reschedule production for new, urgent orders; and improved schedule quality, such as reduction of machine waiting time are the most frequently reported benefits of intelligent scheduling. Quality: Better training of new staff and service personnel, development of better understanding of process issues and equipment behavior, and improved diagnosis of quality problems are examples of how implementation of intelligent software software can result in quality improvement. Energy Efficiency: Energy use in energy intensive industries can be affected by a number of production faults. When production stops, large amounts of energy can be wasted in terminating operations and then resuming them. intelligent software software can offer significant direct and indirect energy savings by improving process diagnosis, stability and consistency in operations and by improving the control response to operating changes.
Summary Heavy industry is taking a new approach to optimize work and production. No longer does heavy industry focus only on optimizing human and hardware resources, but it optimizes knowledge resources as well. In the past, work was organized according to a control paradigm, i.e. a single controlled way of performing a job was always defined. Now the objectives, goals and desired outcomes are defined and it is recognized that due to the complexity of the business environment, several approaches or processes may be equally or similarly effective at achieving the desired goals - work is now organized according to a goal paradigm. Intelligent software systems play an important role in the execution of a goal-oriented paradigm in heavy industry. intelligent software software offer flexibility, adaptability and versatility, so that a variety of approaches may be used to meet a specific goal, depending upon the circumstances and the organization’s requirements.
For more information, please contact us.
|
||
| REDUCT & Lobbe
Technologies Inc. P.O. Box 800, 186 - 8120 No.2 Road., Richmond, BC, Canada V7C 5J8 ph: (604) 275-3711 fax: (604) 275-3715 email: dispatch@reduct.com |
