Uzman sistemler ve deterministik yöneylem araştırması tekniklerinde bir uygulama

Expert systems and an application to tederministic operational research techniques

Tez No: 19349
Yazar: MEHMET BARUT
Danışmanlar: Y.DOÇ.DR. ORHAN KURUÜZÜM
Tez Türü: Yüksek Lisans
Konular: Bilgisayar Mühendisliği Bilimleri-Bilgisayar ve Kontrol, Computer Engineering and Computer Science and Control
Anahtar Kelimeler: Uzman sistemler, Yönetim bilgi sistemleri, Yöneylem araştırması, Expert systems, Management information systems, Operations research
Yıl: 1991
Dil: Türkçe
Üniversite: İstanbul Teknik Üniversitesi
Enstitü: Fen Bilimleri Enstitüsü
Ana Bilim Dalı: Belirtilmemiş.
Bilim Dalı: Belirtilmemiş.
Sayfa Sayısı: Belirtilmemiş.

Özet

Bilgisayar bazlı bilgi sistemleri hızlı bir şekilde gelişmektedir. Bu değişimler, bilgisayar donanım ve yazılımlarındaki gelişmelere bağlı olduğu gibi, organizasyonun bilgisayara dönük gereksinimlerinden de etkilenmektedir. Bilgisayar yazılım alanındaki son çalışmalar, Yapay Zeka <YZ> araştırmasına dayanan yeni ticari amaçlı ürünlerin geliştirilmesi üzerinde yoğunlaşmak tadır. Yapay Zeka, insanlar gibi düşünebilen zeki bilgisayar programlarının geliştirilmesiyle uğraşan bir bilim dalıdır. Bir çok altbileşeni vardır. Bunlar arasında en önemlileri; robotlar, görme sistemleri, doğal dil işleme, model algılama ve uzman sistemler sayılabilir. Yapay zekanın tarihçesi i. Bölümde kısaca verilmiştir. Bu bölümde, ayrıca, YZ altbileşenleri konusunda kısa bir bilgi verilmiştir. 2.Bölümde, Uzman sistem teknolojisi geniş bir şekilde yer almaktadır. Uzman sistemler konusunda sorutabilecek hemen hemen tüm sorulara ait bilgiler açıklanmıştır. Uzman sistemin kurulmasında etkin bir role sahip olan bilgi mühendisi ve bilgi mühendisliği konularına da değinilmişt i r. 3. ve 4. Bölümlerde, Uzman sistemlerin temel yapısını oluşturan bilgi-tabanıve çıkarım mekanizması açıklanmıştır. Uzmandan alınan bilginin bilgi tabanında uygun bir şekilde gösterimi gerekir. Bunun için birçok gösterim şekilleri vardır. Sistem, ayrıca, uygun çıkarımlarda bulunabilecek bir mekanizmaya sahiptir. 5.Bölümde, Uzman sistemlerin diğer yazılım sistemleri arasındaki ilişkisi incelenmeye çalışılmıştır. 6. Bölümde ise, uzman sistemin kurulması için gerekli araçlar ve programlama dilleri üzerinde durulmuştur. 7. Bölüm, bir uzman sistemin geliştirilme aşamalarını, her aşamada ortaya çıkabilecek problemlerle beraber vermektedir. Son bölümde ise, uygulama alanı olarak, Yöneylem Araştırması <YA> seçilmiştir. Elde edilecek uzman sistem, verilen herhangi bir problemin hangi YA teknikleriyle çözülebileceği konusunda tavsiyelerde bulunacaktır. Problemlerde doğrusallık kısıtı getirilmiştir. Uzman sistem kurma aracı olarak VP-Expert Shell'i seçilmiştir. x

Özet (Çeviri)

The CBIS field changes rapidly. Many of the changes are related to hardware and software advances, beside this, there are other forces at work, such as organizations* information requirement, a more computer literate work force, and enhanced understandings of how to use computer related technologies to improve organizational efficiency and effectiveness. One recent trend is the apperance of commercial products that are based on artificial intelligence research. Artificial intelligence <AI > is a field of study that strives to develop computer programs that think like humans. Artificial intelligence includes a number of subsets, of which robotics, vision systems, natural -language processing, and expert systems hold the greatest potential for business application. Robots are machines that are controlled by computer programs. Their use is growing rapidly, especially for Jobs that are unsafe or unpleasant for humans and involve higly repetitive physical tasks. Robots also are becoming increasingly cost-effective relative to human labor. Vision systems use a camera and a computer to provide machines with a sight capability. Progress in this area has been slow relative to other artificial intelligence areas, but it hold great potential, such as for the creation of more intelligent robots. Natural language processing refers to communicating with computers much like humans do to one another, this is not a simple task, because human communications are fraught with ambiguities, context dependencies, unspoken“commen sense”information, and the beliefs and goals of the speaker, progress is being made, however, as seen in fourth-generation software, query languages, and speech-recognition systems. XIComputer technology has introduced many new and innovative products. Perhaps the most exciting is expert system technology, from the artificial intelligence domain. Expert system technology will most likely affect everyone in the computer and software engineering fields, and most software engineers should learn about it. Expert Systems solve problems that normally require human expertise. Some of the better known applications are found in geology and medicine. Expert systems are created by knowledge engineers who acquire knowledge <facts and rules) from a human expert and embed it in a computer program. So, Chapter 1, the introduction chapter, begins with a brief history of Artificial Intelligence and discusses exactly what artificial intelligence means and implies. In addition, the subsets of artificial intelligence examined briefly. Chapter 2, introduce the expert system technology, a subset of AT, widely. No background in expert systems or AI is required for comprehentions, such as what are expert systems?, what are the basic expert systems components?, what opportunities become available through the use of expert systems? In addition, this chapter introduce the role of an expert, a knowledge engineer and the nature of expertise. The designer of an expert system is the knowledge engineer. The role has been described. Expert systems have been designed in many ways, but they share certain basic components. The components common to most expert systems are a user interface module, a knowledge acquisition facility, an inference mechanism, and a database. Often the database is managed by a data manager and is segmented in some logical scheme to support data management, system efficiency, or both. Separate explanation facilities are found in some expert system architectures, while others rely on explanation capabilities of the system's inference mechanism. In chapter 3 and 4, the purpose is to explain the nature of expert systems and knowledge bases together with features of these systems which distinguish them from traditional information processing using file based and database structures. It is stressed that the heart of the expert system is its corpus of knowledge, structured to support decision making. Various forms of knowledge representation are described and their particular applications and limitations highlighted. Reasoning and xnmethods of inference- The representation of uncertainity and its use in reasoning is also covered. This chapter as a whole acts as an introduction to the ideas behind expert systems. Chapter 5 explains the linking between expert systems and other knowledge software. The early expert systems were standalone systems. They did not link to other computer systems. The early toolkits were designed for building standalone systems, often on standalone mac hi nes. Today it is clear that many of the most valuable applications of expert system technology are those in which expert system access databases used for other purposes or link to other types of software or systems. Many important applications in the future will combine conventional data processing with expert systems. Knowledge engineering will be combined with information engineering. Much software will have an artificial intelligence compenent as a small part of its overall code. Historically, application development has been in the hands of computer based information professionals. This is now changing as more and more users, and organizational arrangements to support end-user computing. In this chapter, fistly, the software systems such as MIS, DSS, OAS, examined breifly. Then the study moves on to investigation of the linkage between expert systems and these software. Chapter 6 provide an overview of the general characteristics of AI languages and the different knowledge engineering tools that are used to develop knowledge systems. If we decide to build a knowledge system to assist or replace our expert in a particular domain, we must create a software package. It is instructive to consider exactly what levels of software might underlie the knowledge system. Most programming is done in one of a number of high level languages. Well known high level languages include BASIC, COBOL, FORTRAN, PASCAL, ans C. AI programmers commonly use high level languages such as LISP and PROLOG. PROLOG contains constructs that make it easy to write programs that manipulate logical expressions, whereas LISP consists of operators that facilitate the creation of programs that manipulate lists. These constructs are useful for developing symbolic computing programs, just as iterative constructs like“DO WHILE”loops of PASCAL are useful for numerical programming. xmKnowledge engineering tools are designed to facilitate the rapid development of knowledge systems. Tools also incorporate another aspect of knowledge engineering spesific strategies for representation, inference, and control. They contain elementary constructs for modeling the world that determine the sorts of problems the tool can easly handle. The language tool continuum is a simple way of classiying the various AI languages and tools. In general, languages are more flexible and more difficult to use to prototype a new system rapidly. Only well-trained programmers build systems starting with languages such as PROLOG. Tools are much less flexible. Major knowledge engineering decisions have already been incorporated into the tools. Consequently, if a problem matches the tool, such as VP-EXPERT to develop small but useful knowledge systems, a solution can be developed quickly. Environments stand midway between languages and tools in terms of both flexibility and ease of use. So, The Chapter 6 presents brief discussions on these programming languages and knowledge engineering tools needed to build expert systems. In addition, PROLOG as a logic programming and VP-Expert as a shell emphasised and explained briefly. In chapter 7, we turn our attention to the process involved in developing a large expert system. We shall not focus on the knowledge that goes into the system or on how that knowledge is encoded. Instead, we shall discuss the major steps necessary to plan and develop a large expert system. Unlike the developing small knowledge systems, a large expert system can be developed only by a team of people trained in knowledge engineering. In this chapter we assumed that the group undertaking the development effort has already chosen an expert system building tool. Mor over, we assume that the tool that has been selected is a narrowly focus on how frames, rule interpreters, or graphical displays are implemented. By choosing to describe the development of an expert system by means of a narrow tool, we are eliminating many complexities. This is practical perspective, since most commercial systems have been developed using just such an approach. The main focus of this chapter is on the process that occurs when a team develops a large expert system. It discusses the questions that must be considered at each step along the way, the decisions to be made and the general activities that must occur. xivSo, In this section, Prototyping is adopted and explained as a useful approach to expert systems development. The development stages are covered in detail. Properties of a problem or area that make it suitable for the development of an expert system are first discussed. The likely costs and benefits resulting from the expert system are then assessed. An important choice has to be made between using artificial intelligence languages, expert system tools and shells in the building a system, and so the characteristics of these are explained. Having constructed and assessed the prototype the final system is designed and then evaluated theoretically. The last chapter presents an application with an expert system to the field of Operational Research. The advance of computers and informatics turns organizations ever more into 'information organizations'. Operational research <OR> is a branch of science that could have been very helpful in supporting the decision makers in such organizations, both directly and indirectly by playing a leading role in the development of decision support systems and expert systems. The main topic of this application is models. But there will be some resrictions. The fist restriction was made on models that deal with the operational research. Second, we assume that there is no any risky elements. This refers to deter mi niştik models, where there is no variability in the parameters of the model, and where each takes a single volue. Third restriction is about algorithms. Several of the algorithms are very closely related to each other, and these will be dealt with first, under the common heading of mathematical programming. Within the deterministic category, this includes linear programming, - the assignment and transportation algorithms, goal programming and integer pr ogr ammi ng. In this chapter 8, we, firstly, explained the connection between operational research and expert systems, and then expressed how expert system can be entegrated into the operational research or industrial engineering environment. At the end, we turn our attention to the application. Here, VP-Expert shell has been selected as an expert system building tool. The lojik-based rules has been organized and represented in an appropriate syntax. xv

Benzer Tezler

Tez No
34086
An Expert system design for a transportation problem
Başlık çevirisi yok
BURAK ACAR
Yüksek Lisans
İngilizce
1994
Elektrik ve Elektronik Mühendisliği Marmara Üniversitesi
Endüstri Mühendisliği Ana Bilim Dalı
DOÇ. DR. LALE DURUİZ
Tez No
104335
Atölye tipi çizelgeleme için uzman-tabu arama modeli
Expert-tabu search for job-shop scheduling
FARUK GEYİK
Doktora
Türkçe
2000
Endüstri ve Endüstri Mühendisliği Sakarya Üniversitesi
Endüstri Mühendisliği Ana Bilim Dalı
YRD. DOÇ. DR. İSMAİL HAKKI CEDİMOĞLU
Tez No
14242
U-Mut: autolisple bilgisayar destekli uzman-mutfak tasarımı
U-Mut: a computer aided expert-kitchen design written in autolisp
MANOLYA KAVAKLI
Yüksek Lisans
Türkçe
1990
Mimarlık İstanbul Teknik Üniversitesi
PROF.DR. NİGAN BAYAZIT
Tez No
56024
İleri imalat teknolojilerinin ekonomik analizi ve esneklik faktörünün sayısallaştırılmasına bulanık kümeler yaklaşımı
Başlık çevirisi yok
CENGİZ KAHRAMAN
Doktora
Türkçe
1996
Endüstri ve Endüstri Mühendisliği İstanbul Teknik Üniversitesi
PROF.DR. ETHEM TOLGA
Tez No
97511
Modele de determination de prix de vente effectif flou
Bulanık etkin satış fiyatı belirleme modeli
FARUK ERTUĞ
Yüksek Lisans
Fransızca
2000
Endüstri ve Endüstri Mühendisliği Galatasaray Üniversitesi
Endüstri Mühendisliği Ana Bilim Dalı
PROF. DR. ETHEM TOLGA

Geri Dön