Artificial Intelligence and expert systems

Artificial Intelligence and Expert Systems  | Categories of expert systems ....

Meet Goldfire-Optimal Decision Engine from Invention Machine

 

Artificial Intelligence and Expert Systems -  Artificial Intelligence (AI) is a branch of computer science,  which  focuses on the development of computer systems to simulate the processes of problem solving and duplicate human brain functions. Expert systems (ES) are a category of programs based on the  theory and methods of  artificial intelligence.

The first expert systems appeared in the late sixties. Today, they exist in many forms, from medical diagnosis to investment  analysis and from counseling to production control. Due to the advances of the last  decade, today's expert systems users can choose from dozens of commercial software packages. At present, we accept as routine such expert systems as weather forecasting, online mapping and driving directions, diagnostic systems for automotive repair shops, and so on.

Many large corporations use expert systems in their business.  The list of the companies  using  expert systems technology is long and varied: NASA, HP,  Lockheed,  Boing, DaimlerChrysler AG, various power, gas and oil stations, etc.

Expert systems tend to be more effective than other computer based applications, because they:

• may combine  the knowledge of many experts in a specific field,

• can store  an unlimited amount of  information, and works  much faster, than a human,

• are  available 24 hours a day, and can be used at a distance over a network,

• are able to explain their information requests and suggestions,

• can process client's uncertain responses and, by combining several pieces of uncertain information, may still be able to make strong recommendations,

• can accumulate the knowledge of high level employees for any company, which is  especially useful  when the company needs to fire  them due to worsened market conditions.

Categories of Expert Systems

Many applications are not covered here, and some manufacturers and resellers belong in multiple categories. Due to the space limitations,  manufacturers are included in only one category.

 Interpretative ES

Expert systems in this category interpret information from sensors to describe a situation. For example, this might be the interpretation of process measuring information in a chemical plant. Interpretative ES deals with real data. Therefore, these ES must not only retrieve, but also analyze, filter and restore missed data, as well as mine data, transform data to another format, and so on.
Interpretative expert systems can handle various types of data, such as data for the recognition of voice, visual, smell, touch or gestures.
Examples within this category of ES include:

• Netrics Data Quality Software, from Netrics, Inc., which enables cleansing of data for mission-critical and business-critical operations. The Netrics line of products may deliver a near-perfect accuracy of data, for “dirty” or error-prone data. Netrics develops tools which allow matching imperfect data in real-time, despite inconsistencies, incompleteness, variations and errors.
• Nuggets and eNuggets toolkits, from Data Mining Technologies, Inc., are desktop applications, which build models to uncover hidden facts and relationships for industries such as healthcare, biotechnology, and telecommunications. These tools can predict new data and trends, and reveal which variables make decisions with the most impact.

ES for Abnormal Condition Recognition

Expert systems perform abnormal condition recognition (ACR) or, in other words, diagnosis of the condition of an object or process using a description of the situation, knowledge of the architecture (structure and functions) of the subject, and other methods, to establish the probable causes of incorrect operations.

These expert systems oftentimes serve as consultants, who not only make the diagnosis, but also help with debugging. ACR can assess the technical condition of a complex industrial system, or the health condition of a human . The technical diagnostic type of expert system is combining artificial intelligence techniques with equipment fault diagnosis techniques. It uses special knowledge and a method provided by experts which simulates the inference process of the experts, to solve various complicated and important problems.

MatheMedics Inc. has developed medical diagnostic programs for the back, chest, abdomen, and other sign/symptom domains, and has considerable experience with the creation of efficient medical expert and decision support systems.

Gensym Corporation has developed the G2 platform and new products, NeurOn-Line, Optegrity, and e-SCOR, which enable real-time expert system applications.

Transformer Oil Analyst (TOA), developed by Delta-X Research, warns the staff of any suspicious symptoms or trends. Whenever an equipment or insulating fluid problem is suspected, TOA enables the user to quickly review both the historic (previous) and current status.

Shop Automated System of Technical Diagnostics (SASTD) was created by Intellectual Systems, Inc. in Russia, while the author was founder and President of this company. SASTD is real-time hybrid ES for a pipe line gas compressor station.

   ES for Prediction and Forecast

The prediction and forecast category of ES enables the forecasting of possible outcomes of observable situations. Examples may include the prediction of crop damage from harmful insects, estimation of the demand for oil in the global market place, or prediction of the next armed conflict. Also included in this category are weather forecasting, scoring and predictive modeling, risk management and other financial forecasting, to name a few. Examples of products in this category are:

• NeuralWorks Predict, developed by Neuralware Inc., is an integrated tool for rapidly creating and deploying prediction and classification applications. This system is based on neural network technology, combined with genetic algorithms, statistics, and fuzzy logic, to automatically find optimal or near-optimal solutions for a wide range of problems.

• NeuroXL Predictor, developed by NeuroXL Inc., is a neural network forecasting tool that quickly and accurately solves forecasting and estimation problems for a wide variety of tasks, including stock price prediction, sales forecasting, and sports score prediction.
• Forecasting Software developed by the Vanguard Corporation, combines different decision support methods to find the best forecast for the needs of an individual or entire organization.

ES for Design and Management Solutions

 

After considering a great number of variable parameters for solving innovation tasks, ES can assist in the optimization of design, and find the best management decisions.

Usually such expert systems are important parts of CAD/CAE/SIM systems. In this category are such products as the Goldfire complex of tools, developed by Invention Machine. These Goldfire tools provide:

• Innovation Trend Analysis

• Research

• Process Analysis

• Root Cause Analysis

ES for Control and Monitoring

These categories exist as the portion of control systems that help to monitor operations, and to control certain functions of the machinery in various types of enterprises. Many corporations that have developed Supervisory Control and Data Acquisition Systems (SCADA) have used expert systems for different control levels of industrial plants. For example, Emerson Process Management, Inc. is developing the following products:

• The DeltaV Advanced Control Suite, which includes simulation and optimization subsystems

• The Ovation expert control system, which gives users higher levels of plant availability, reliability and environmental compliance

• WDPF control systems, which integrate plant process control, local and wide area SCADA systems, PLC networks, maintenance management, and so on.

The British company, Knowledge Process Solutions Inc., has developed an intelligent plant technology (IntelliPlant) that includes model-based multivariate control and optimization subsystems for minimizing costs. The control and optimization technology includes the ability to develop statistical models of equipment behavior, based on data mining technology. These technologies have a low cost with fast implementation speed.

ES for Analysis and Reporting

 

There are a several popular applications on the American market:

• The Analysis, DSS and Reporting Suites, developed by MicroStrategy Inc., are Business Intelligence applications for integrating, reporting, analysis, and information delivery abilities in one platform

• PowerCenter and Informatica Platform, from Informatica Corporation, is designed for mass deployment and adoption of business analytics

The Israeli company, Expert Solutions International, produces the expert system, Logist, which enables organizations to rapidly create robust applications, well-integrated with their business activities, while achieving the following goals:

• Preserve the organization’s hidden knowledge assets, in the form of business rules

• Create dynamic applications that can adapt to market speed, thereby sharpening the organization’s competitive advantage

• Simplify knowledge sharing within the organization

• Empower the business with knowledge

• Lower implementation and assimilation costs

ES for Manufacturing

 

Manufacturing is one of most robust areas of expert systems and AI-related activities. It is very difficult to find statistical data about the implementation of these systems in the various industries of developed countries of the world. However, it is possible to make some extrapolation. According to a 1994 study involving ninety-eight Fortune 500 industrial companies, expert systems were utilized in manufacturing by 52% of the companies, where ES was perceived to be very useful in process planning, product design, layout, facility design and maintenance. Undoubtedly, there are now more than 52% of solid companies using expert systems.

Manufacturing ES is a part of Computer-Integrated Manufacturing (CIM) or Intelligent Manufacturing Systems (IMS). Manufacturing ES may be applied to almost any manufacturing area, including planning, design and system configuration, process control, quality control, and so on.

One popular product on the market is G2 e-SCOR, developed by Gensym Corporation. The G2 e-SCOR product enables the simulation of various configurations, tests the robustness of the supply chain, and identifies the required service levels. This product can help identify the weak links and areas for improvement in the supply chain, thereby saving time and money.

ES for Marketing and Trade Management

 

A key use of artificial intelligence in marketing and trade management ES is to determine marketing potential and business opportunities.

The Ward Systems Group, Inc. puts out AI Trilogy, a new package of business and scientific software. AI Trilogy contains the NeuroShell Predictor, NeuroShell Classifier, and GeneHunter software together in one package.

Data Mining Technologies Inc.  provides several types of modeling services for developing marketing and trade management ES, including:

• Prospect segmentation modeling

• Customer segmentation modeling

• Customer cloning

• Prospecting for new customers

• Finding high-value new customers

• Maximizing return from loyalty programs

MarketResearch.com is the world’s largest and continuously-updated collection of market research, with more than 200,000 reports from over 650 leading global publishers. Much of this was accomplished with the help of ES.

ES for E-Learning

 

E-Learning Systems (ELS) is based on special tools, and supplies individual or group education and training for different disciplines. These learning tools and services make the learning process faster and more adaptive to the challenges of the business world. ELS can include Learning Management Systems and Course Management Systems. Following are some developers and providers of e-Learning systems:

• GeoLearning, Inc.  is a leading developer of e-learning delivery platforms and web-based training solutions for organizations around the world, including Fortune 1000 companies, government agencies, and so on.

• CompanyCollege.com  is the e-learning solution from Business Training Library Inc., the provider of affordable training solutions for small companies; this company offers easy to use, completely customizable online learning to the Business Services, Financial Services, and Healthcare fields, among others.

 

For more information  see  the book "Artificial Intelligence Around Us".