Bulanık kümeler ve endüstri mühendisliği uygulamaları
Başlık çevirisi mevcut değil.
- Tez No: 55757
- Danışmanlar: PROF.DR. AHMET FAHRİ ÖZOK
- Tez Türü: Yüksek Lisans
- Konular: Endüstri ve Endüstri Mühendisliği, Industrial and Industrial Engineering
- Anahtar Kelimeler: Belirtilmemiş.
- Yıl: 1996
- Dil: Türkçe
- Üniversite: İstanbul Teknik Üniversitesi
- Enstitü: Fen Bilimleri Enstitüsü
- Ana Bilim Dalı: Belirtilmemiş.
- Bilim Dalı: Belirtilmemiş.
- Sayfa Sayısı: 37
Özet
Birinci bölümde L.ZADEH'in ortaya koyduğu teori genel şekilde açıklanmış, işlemsel operatörlerden, mantık operatörlerinden bahsedilmiş ve bunların kullanımı açıklanmıştır. Bu bölümde ayrıca endüstriyel uygulamalardan bahsedilmiştir. Bulanık Mantık temelde çokdeğerli mantık, olasılık kuramı, yapay zeka ve yapay sinir ağlan üzerine oturtulmuş olayların oluşum olasılığından çok oluşum derecesi ile ilgilenilen bir kavrandır. Olasılık ve bulanıklık kavranılan arasındaki en önemli farklılık bulanıklığın bir deterministik belirsizlik olmasıdır. Bulanık mantık yaklaşımı günümüzde bir çok uygulama alam bulmaktadır.Bunların başlıcaları; kontrol algoritmaları, tıbbi tahliller, karar verme, ekonomi, mühendislik, çevre, literatür, yöneylem araştırması, örnek tanımlama, pisikoloji, güvenilirlik, güvenlikdir. Bulanık Kümelerin Avantajları: İnsan düşünüş tarzına yakım olması, matematiksel modele ihtiyaç duymaması, uygulamaların hızlı ve ucuz olması, bulanık mantık çok karmaşık, lineer olmayan, belirsizlikler içeren, geleneksel yöntemlerle gerçekleşemeyen sistemlerin oluşturulmasına olanak tanır. İkinci bölümde ağırlıklı olarak Endüstri Mühendisliği kapsamında Bulanık Kümeler yaklaşımının kullanılabilirliği incelenmiş ve yapılan uygulamalara değinilmiştir. Son olarak bu bölüm içerisinde uygulama örneklerinden 'Personel Seçimi' konusu detaylı olarak incelenmiştir. Personel seçiminde dilsel sübjektif değerlendirmelerin sağlıklı yapılabilmesi için Bulanık Kümelerin kullanımı örneklendirilmiş ve süreç adım adım açıklanmıştır. Endüstriyel Proseslerin Kontrolü ve Kalite Kontrol, Üretim Çizelgeleri ve Stok Kontrolü,İnsan Faktörü ve Yöneylem Araştırması genel uygulamalardır. vu
Özet (Çeviri)
Abstract: The propose of this thesis is to show that the Fuzzy Logic applications in industrial environment. This thesis consist of basic approaches, general logic rules and industrial applicability of fuzzy logic. In first part of the work, Fuzzy Logic theory, operation and logic operators and their using are explained in general. The field of industrial engineering professes to be the discipline which is most concerned with humans, human systems and human decision making. However, to paraphrase Dr. Lotfi Zadeh, many of the techniques used in solving industrial engineering problems, i.e., human-centered problems, are adaptations of tools and techniques for solving mechanistic systems. Since Dr. Zadeh first published his formulation of the theory of fuzzy logic sets in 1965, there has been a frenzy of activity in applying his theories to problems arising in a variety of fields. But, while there has been some activity in the industrial engineering community to develop application of fuzzy methodologies. The characteristic function of a crisp set assigns a value of either 1 or 0 to each individual in the universal set, thereby discriminating between members and nonmembers of the crisp set under consideration. This function can be generalized such that the values assigned to the elements of the universal set fall within a specified range and indicate the membership grade of these elements in the set. Larger values denote higher degrees ofset membership. Such a function is called a membership function and the set defined by it a fuzzy set. Let X denote a universal set. Then, the membership function fÂ. by which a fuzzy set A is usually defined has the form fA:X->[0,l], where [0,1] denotes the interval of real numbers from 0 to 1, inclusive. Fuzzy sets are often incorrectly assumed to indicate some form of probability. Despite the fact that they can take on similar values, it is important to realize that membership grades are not probabilities. One immediately apparent difference is that the summation of probabilities on a finite universal set must equal 1, while there is no such requirement for membership grades. vuiThe application areas covered are quality control and control of industrial processes, production scheduling and inventory control, human factors, organizational design and financial management, and operations research. The theory of fuzzy sets allows the existence of a type of uncertainty due to vagueness or fuzziness rather than due to randomness alone. Another word for this type of uncertainty would be“imprecision”. In this most basic sense, a fuzzy set is a set where objects have gradual rather than abrupt transition from membership to nonmembership. Industrial engineering basically concern itself with decision making- for example, in the planing, design, or operation of man-machine systems. These systems can be classified as manufacturing systems, health care systems, transportation systems, energy systems, information systems etc., but in each case, planning, design, and operation implies making and implementing decisions. Typically, these decision are made trough the use of models or representation of the system under consideration. The model can be classified as“mental models”or mathematical models, but in either case the model users have implicit or explicit criteria under consideration. The model basically permits the ranking of various alternative decision (or design) with respect to various implicit or explicit criteria. Models are formulated through two types of interdependent inputs.: 1) Data gathered about the problem situation, and 2) The values, perceptions, biases, etc. of model builders. For example, in layout design problem, data concerning materiel flows must be obtained and integrated with other information in the building of model. As another example, the value system of the designer may be used in either explicit or implicit tradeoff among multiple conflicting objectives. Both of the above types of inputes are characterized by imprecision or vagueness, and their vagueness is not always due to randomness (and therefore cannot be related to probability theory). For example, in the design of a quality control system, the designer may be vague about the tradeoff he/she is willing to make between quality and cost. Fuzzy set methodologies allow a model builder to capture and consider this inherent vagueness within the modeling process. In effect, the modular no longer must ignore this inherent vagueness. In second part, Industrial Engineering applications are investigated. Mainly, Personnel Selection Process is explained. An algorithm for personal selection was developed by Liang and Wang. The method first aggregates decision makers linguistic. Assessments about subjective criteria weightings and rating to obtain the fuzzy suitability index and its ranking value. Further, coming the subjective and objective ranking values, the final ranking values for personnel suitability evaluation are obtained. Than the most suitable personnel can be selected. Since either test-oriented objective approach or interview-oriented subjective approach has its advantages and disadvantages, the method combining both subjective and objective assessment proposed in this study is desirable and convincing for personnel selection. IXIn thesis it's mentioned some selected application are below: Quality Control and Control of Industrial Processes: The primary usefulness of fuzzy sets, subset and linguistic variables lie in their ability to treat, in a quantitative manner, the natural vagueness which exist in these types of problems. In one type of quality control problem, acceptance sampling, there is an effort to determine the acceptability of an entire lot of parts, comments, or products based upon a representative sample taken from the lot. Classical statistical methods have been employed to design acceptance sampling plans. If a lot of items is sampled and a specified number of“defective”items are found in the sample, the entire lot is either rejected as defective or the entire lot is 100 % inspected to detect all defective items. The problem is that there is often no clear-cut definition of a defective item. Rather, there exists a“degree of defectiveness”which is routinely disregarded in traditional acceptance sapling plans. The concept of degree of defectiveness can be handle quite easily, in theory, by the use of fuzzy sets. However, in practice, the definition of membership function for these fuzzy sets may be very difficult. Similarly, control charts are used in quality control applications to graph the variability of a process variable over time. The typical control chart consists of a center line, which represents the average value of the process variable, and upper and lower control limits usually defined as a function of the standard deviation of the process. When the process variable value falls outside either control limit the process is considered to be out of control and the search for an assignable cause for this condition is undertaken. Fuzzy control limits may be constructed as fuzzy bands equal to the difference between the value of quality characteristic where the nonconformance cost is maximum and the value where it is zero. Production Scheduling and Inventory Control: Management science have been interested in industrial scheduling problems for at least three decades; however, as McKay, Safayeni and Buzacott point out, their efforts have had little impact on the industrial world of job- shop scheduling. They list a number of reason for this lack of impact but among these is the vagueness of decision routinely made on an ad hoc basis by human schedulers. These individuals who must perform detailed scheduling and dispatching tasks use their intuition along with a great deal of sensory data as input to a mental model of shop floor at any point in time These inputs are played against well-known constraints and algorithms to arrive at a scheduling decision in a dynamic environment. One promising approach which may enable the development of more effective scheduling models is use of the decision inputs and variable relationships, i.e., they are helpful in constructing fuzzy algorithms which produce production schedules. Along the lines of the preceding discussion, many inventory control models have been developed by operation researchers over the last three decades. However, the parameters required by these models are extremely difficult to estimate, especially carrying costs, shortage costs, demand patterns, setup costs, variable production costs and lead time. Each of these parameters could be modeled as fuzzy numbers to arrive at membership function associated with desirable lot sizes and order release times. Human Factors: It is defined as the interdisciplinary study of optimization of work systems with respect to human characteristics, expectation and behavior, investigatescomplex relationship between people, machines and physical environment. The propose of human factors investigation is to remove the incompatibilities between human capacities and requirements of the tasks they perform in order to provide for a safe, healthy, productive and convertible working condition. One of main problems that makes the above goal difficult to achieve is the natural fuzziness inherent to complex man- machine relationships. According to Zadeh“great complexity of such systems calls for approaches that are significantly different in spirit as well as in substance from the traditional methods which are highly effective when applied to mechanistic systems, but are far too precise in relation to systems in which human behavior plays an important role.”In general, fuzziness in man-machine system is due to: 1) Inability to acquire and process an adequate amount of information about systems. 2) Vagueness of the relationships between people and working environments, and 3) Vagueness of human thought processes. From the human factors point of view, fuzziness, or vagueness, arises at all levels of cognitive processes and therefore, is implicit in human behavior. Although most human characteristics have very complex contextual dependencies which are readily expressible in tabulations of numbers, people comprehend vague concepts as if those concepts were represented by fuzzy sets, and manipulate them according to the rules of fuzzy logic. While human factors specialists understand vague concepts such as“high workload”,“low illumination”, or“heave load”they cannot use these terms in their work if only traditional methods of data analysis are applied. Fuzzy set theory, which allows for mathematical modeling and manipulation of vague information, and evaluation of uncertainty due to fuzziness, rather than randomness alone, can be a powerful tool to deal with human-based uncertainty. Personnel selection problem very good example for fuzzy logic applications. To assure that the right people are placed in the right jobs, personnel selection has always been an important issue for all organizations. Many individual attributes, e.g. organizing ability, creativity, personality, emotional steadiness, comprehension, leadership, general aptitude, etc., are considered for personnel selection. These attributes can be broadly classified into two categories: 1) subjective attributes these qualitative definitions, e.g., personality, leadership, past experience, and 2) objective attributes- these can be assessed quantitatively, e.g. general aptitude, job related knowledge, analytical ability, etc. Many precision-based approaches for personnel selection have been conducted and many of them were focused on the design and evaluation of selection pretests. The test-oriented research can make tremendous contributions in dealing with job situation which the job requirements can be quantified. Another approach for personnel selection is though interview. In the real world, to evaluate personnel suitability through interview or test, measures of the subjective criteria, e.g.,personality, leadership, emotional steadiness, past experience, self confidence and intelligence may not be precisely defined for the decision makers. In addition, personnel suitability rating under each of the subjective criteria as well as the XIimportance weights of the subjective criteria are very often assessed by linguistic D terms, such as, 'good', 'very low', 'poor', etc. Due to that the subjective evaluation data can be more adequately expressed in fuzzy linguistic variables. A fuzzy multiple-criteria decision making method is used to integrate various linguistic assessments and weights to obtain the fuzzy suitability index. The index can take into account the ambiguities involved in the evaluation process for various subjective criteria. Than, by combining the subjective ranking value with the objective ranking value by a subjective criteria weight, the final rating for personnel suitability assessment can be conducted. A stepwise description of the multiple-criteria personnel selection procedure is given in the following. Step 1. Form a committee of decision makers, and identify the required selection criteria. Step 2. Devide personnel selection criteria into subjective and objective categories. Step 3. Assign importance weights to the subjective criteria. Step 4. Assign suitability rating to the candidates versus varies criteria. Step 5. Tabulate the weightings of the subjective criteria, then obtain the aggregated weighting. Step 6. Tabulate the preference ratings by the decision-makers, then pool them together to get the aggregated fuzzy ratings of candidate under subjective criterion. Step 7. Tabulate the test scores associate with different candidates versus various objective criteria. Step 8. Aggregate fuzzy ratings and aggregated weighting with respect to all the subjective criteria, then obtain the fuzzy suitability index for all the subjective criteria. Step 9. Calculate the subjective ranking values of all the candidates. Step 10. Calculate the ranking value of each candidate's test scores of all the objective criteria. Step 1 1. Calculate the final ranking values. Step 12. Select the candidate with the maximum final ranking value. xii
Benzer Tezler
- Modele de determination de prix de vente effectif flou
Bulanık etkin satış fiyatı belirleme modeli
FARUK ERTUĞ
Yüksek Lisans
Fransızca
2000
Endüstri ve Endüstri MühendisliğiGalatasaray ÜniversitesiEndüstri Mühendisliği Ana Bilim Dalı
PROF. DR. ETHEM TOLGA
- Birikimli kanı dereceleri yaklaşımına dayalı yeni bir grup karar verme yöntemi
A novel group decision making approach based on the cumulative belief degrees
BİLAL ERVURAL
Doktora
Türkçe
2018
Endüstri ve Endüstri Mühendisliğiİstanbul Teknik ÜniversitesiEndüstri Mühendisliği Ana Bilim Dalı
DOÇ. DR. ÖZGÜR KABAK
- Assessment of challenges of the internet of things in supply chain during pandemic using interval-valued neutrosophic analytical hierarchy process
Aralıklı nötrosofik analitik hiyerarşi sürecini kullanarak pandemik sırasında tedarik zincirindeki nesnelerin interneti zorluklarının değerlendirilmesi
MARZHAN DOSZHANOVA
Yüksek Lisans
İngilizce
2020
Endüstri ve Endüstri MühendisliğiMarmara ÜniversitesiEndüstri Mühendisliği Ana Bilim Dalı
PROF. DR. SELİM ZAİM
- Ayrıştırılmış bulanık küme tabanlı çok kriterli karar verme yöntemi geliştirme ve uygulamaları
Developing decomposed fuzzy set based multi-criteria decision making methods and applications
ÖZLEM ARSLAN
Doktora
Türkçe
2024
Endüstri ve Endüstri MühendisliğiYıldız Teknik ÜniversitesiEndüstri Mühendisliği Ana Bilim Dalı
PROF. DR. SELÇUK ÇEBİ
- Hata türü ve etkilerinin bulanık kümeler yaklaşımı ile analizi
Analysis of failure modes and effects with fuzzy sets approach
ÖNDER ÖZDEMİR
Yüksek Lisans
Türkçe
2004
Endüstri ve Endüstri MühendisliğiYıldız Teknik ÜniversitesiEndüstri Mühendisliği Ana Bilim Dalı
PROF. DR. HÜSEYİN BAŞLIGİL