Toplam kalite anlayışı içinde istatistiksel proses kontrolün rolü ve kalite geliştirme amaçlı uygulanması
Başlık çevirisi mevcut değil.
- Tez No: 75346
- Danışmanlar: DOÇ. DR. COŞKUN ÖZKAN
- Tez Türü: Yüksek Lisans
- Konular: Endüstri ve Endüstri Mühendisliği, Industrial and Industrial Engineering
- Anahtar Kelimeler: Belirtilmemiş.
- Yıl: 1998
- Dil: Türkçe
- Üniversite: İstanbul Teknik Üniversitesi
- Enstitü: Fen Bilimleri Enstitüsü
- Ana Bilim Dalı: Endüstri Mühendisliği Ana Bilim Dalı
- Bilim Dalı: Belirtilmemiş.
- Sayfa Sayısı: 166
Özet
ÖZET Bu çalışmada, globalleşen dünya içinde çok çetin olan rekabet şartları altında varlıklarını daha düşük maliyet ve daha yüksek kalite ile üretim yaparak sürdürmek isteyen her işletmenin, vazgeçilmez uygulamalarından biri olan İstatistiksel Prosos Kontrol konusu incelenmiştir. Birinci bölümde Kalite, Kalite Kontrol, Toplam Kalite gibi temel kavramlar konusunda kısa bilgiler verilmiştir. Sonra İstatistiksel Proses Kontrolün tanımı, Toplam Kalite Yönetimi içindeki yeri, kullanım amaçları, kullanım alanları ve faydaları incelenmiştir. İstatistiksel Proses Kontrol; bir ürünün en ekonomik ve yararlı bir şekilde üretilmesini sağlamak amacıyla, istatistik prensip ve tekniklerinin üretimin tüm aşamalarında kullanılmasıdır. İPK, ürün için belirlenmiş spesifikasyonlar, tezgah ve proses yetenekleri kısıtları altında, prosesten alınan yakın geçmişe ait veriler kullanılarak mevcut üretimin firesiz ve yeniden işleme ihtiyacı doğmadan sürdürülmesini sağlar. İkinci bölümde, bir İPK uygulaması sırasında kullanılacak temel istatistiksel bilgileri ve formülleri hatırlatılarak, kalite uygulamalarında kullanılan İstatistiksel Teknikler ele alınmış ve en çok kullanılan Temel İstatistiksel Teknikler incelenmiştir. Üçüncü bölümde, İPK çalışmalarının en önemli araçlarından biri olan Kontrol Diyagramları incelenmiştir. Kontrol Diyagramları 'nın çeşitleri, yapısı, kullanımı ve yorumu ile ilgili bilgiler sunulmuş, uygulayıcılarına getireceği faydalar irdelenmiştir. Kontrol Diyagramları; ürünün gerçek kalite spesifıkasyonlarını, geçmiş deneylere (verilere) dayanarak saptanan limitlere göre kronolojik kıyaslamaya yarayan grafiklerdir. Kontrol diyagramları, arzu edilen niteliklerde ürün veya hizmet üretebilmek için prosesin istatistiksel olarak kontrol ve analiz edilmesinde kullanılmaktadır. Dördüncü bölümde, ilk üç bölümde ele alınan ve incelenen bilgilerin ışığında bir İstatistiksel Proses Kontrol uygulaması için izlenmesi gereken bir faaliyet planı geliştirilmiştir. Prosesin tanımlanması ile başlayan bu planda, kontrol altında tutulacak karakteristiklerin belirlenmesi, ölçme alet ve yöntemlerinin seçimi, tezgah ve proses yeterlilik analizleri gibi faaliyetlerle ilgili yapılması gereken işlemler sıralanmıştır. Son bölüm olan beşinci bölümde ise geliştirilen faaliyet planına uygun olarak TEE SULTANÇİFTLÎĞİ İşletmesinde yapılmış olan pilot uygulama anlatılmaktadır. Hedef, bu uygulamanın tüm işletme geneline yaygınlaştırılarak zaten yüksek olan ürün kalitesinin daha da arttırılması ve maliyetlerin düşürülmesidir. XIII
Özet (Çeviri)
SUMMARY STATISTICAL PROCESS CONTROL The concept of quality with respect to customer satisfaction has been with us since the human being produces for others. There are two important things for a product and a production process. One of them is the cost of the product, which im plies the cost of scrap and rework. The other important thing is the quality level of the product. While markets become more global today, both of them are very impor tant for a company. Companies must reduce their costs and increase the level of the quality, in order to maintain their market share. If the cost of the product is not low enough or the quality of the product is not satisfactory, customers don't buy it. In or der to increase the quality level and to reduce the costs, some quality tools are needed. One of them and may be the most effective of them is Statistical Process Control. The purpose of this study is to provide a thorough coverage of the Statistical Process Control. I tried to analyse the structure of the Statistical Process Control and to improve a plan for activities. In the first chapter, information about basic concepts, such as Quality, Quality Control, Total Quality etc. is given. Then, definition of statistical process control, its place within TQM, application purposes, application fields and advantages is analysed. It should be evident that everyone is, or at least should be, in favour of quality. But, the concept of the quality does not mean the same thing to everyone who uses the word. As a matter of fact, the word quality may have different meanings to the same person, depending upon the context in which he uses it. For such a short and simple word, it is quite difficult to find widely acceptable definition. As a short definition we describe the quality, fitness for use. Taguchi describes the quality as the loss (from function variation and harmful effects) a product causes to society after being shipped, other than any losses caused by its intrinsic functions. Quality Assurance and Quality Control address the mean and techniques of producing quality products. Quality assurance means to assure quality in product so that a customer can buy with confidence and use it for a large period of time with confidence and satisfaction (satisfy the requirements of customers). Total Quality Control (TQC) in an effective system for integrating the quality development, quality maintenance and quality improvement efforts of various groups in an organization so as to enable marketing, engineering, production and service at the most economical levels which allow for full customer satisfaction. Total Quality Control and Total Quality Management are different from each other. Total Quality Management is both a philosophy and set of guiding principles that represent the foundation of a continuously improving organization. XIVStatistical Quality Control is the application of statistical techniques in all stages of an operation in order to meet established standards of quality in the most economical manner. A process, which is operating within certain limits that can be specified numerically in terms of the quality characteristic being measured is said to be in control. When any quality characteristic is being observed or measured, the observed value will vary from observation to observation. Some variation is inevit able and is an inherent part of the process being observed. The acceptable variation that is inherent to the process is due to a wide range of random causes and is called random variation. Variation due to causes that are not part of this random system leads to excessive variation, which is not acceptable. Such non-random causes of variation are referred to as assignable causes. A process exhibiting variation that is subject to assignable causes is said to be out of control. Statistical process control, which is a part of statistical quality control, takes a process under control. SPC fol lows a process with considering the determined specifications for the product by using the recent data for the process concerned under machine and process capability constraints. In the second chapter, some basic statistical terms are given. Statistical techniques that are used in quality practices are discussed. In attempting to control any quality characteristic it is necessary to make observations or measurements that result in numerical data. Whenever numerical data are collected and analayzed for the purpose of making some decision and taking some action, statistical techniques must be used. In this chapter the basic statistical techniques are discussed. These tech niques can be classified into three groups. Generally, the most commonly used tools are the seven basic quality tools. These tools are : 1. Cause-effect diagram 2. Stratification analysis 3. Check sheet 4. Histogram 5. Scatter diagram 6. Pareto analysis 7. Control charts. Ishikawa states that as much as 95 percent of quality related problems in the factory can be solved with these seven fundamental quantitative tools. These tools are briefly described in the second chapter. In the third chapter, control charts are analysed. Types of control charts, structures of them, applications and comments are presented and the advantages for the users are discussed. Statistical process control charts help us to detect, diagnose and correct production problems in a timely fashion. The result is substantial improvement in product quality. On the other hand, control charts also tell us when to leave a process alone, thus preventing unnecessary adjustments that tend to increase the variability of the process, rather than reduce it. In addition to detecting the presence of special causes, control charts helps us estimate the natural tolerance of a production process, which in turn permits us to estimate process capability. A natural tolerance is defined as the interval covering ±3 a from the mean of a measured process quality characteristic. Knowing our process capabilities helps us to select the appropriate machines, tools and so on, to meet engineering specifications consistently in our production operations. XVSPC calls for understanding the important distinction between variables data and attributes data. When a record is made of an actual, measured quality character istic, the quality characteristic is said to be“expressed”by variables. On the other hand, ifa record shows only a summary or classification with regard to any specified set of requirements, either expressed or implied, it is said to be a record by“attrib utes.”Trouble is a common state of affairs in manufacturing. Whenever the trouble consists of difficulty in meeting quality specifications that are expressed in terms of variables, the Shewhart control charts for X and R are indispensable tools in the hands of the trouble shooter. They provide information on three matters, all of which need to be known as a basis for action. These are 1. Basic variability of the quality characteristic 2. Consistency of performance 3. Average level of the quality characteristic No production process is good enough to produce all items of product exactly alike. Some variability is unavoidable; the amount of this basic variability will depend on various characteristics of the production process, such as the machines, the materials, the operators. Where both upper and lower values are specified for a quality characteristic, as in the case of dimensional tolerances, one important question is whether the basic variability of the process is so great that it is impossible to make all the product within the specification limits. When the control charts shows that this is true and the specifications cannot be change, the alternatives are either to make fundamental change in the production process that will reduce its basic variability or to face the fact that it will always be necessary to sort out the acceptable product. Sometimes, however, when the control chart shows so much basic variability that some product is sure to be made outside the tolerances, a review of the situation will show that the tolerances are tighter than necessary for the functioning of the product. Here the appropriate action is to change the specifications to widen the tolerances. Variability of the quality characteristic may follow a change pattern, or it may behave erratically because of the occasional presence of assignable causes that can be discovered and eliminated. The control limits on chart are so placed as to disclose the presence or absence of these assignable causes. Although their actual elimination is usually an engineering job, the control chart tells when, and in some instances suggests where, to look. As previously mentioned, the action of operators in trying to correct a process may actually be an assignable cause of quality variation. A merit of the control chart is that its tell when to leave a process alone as well as when to take action to correct trouble. The elimination of the assignable causes of erratic fluctuation is described as bringing a process under control and is responsible for the many of the cost savings resulting from statistical quality control. Even though the basic variability of the process is such that the natural tolerance range is narrower then the specified tolerance range, and even though the process unsatisfactory because the average level of the quality characteristics is too low or too high. This also will be disclosed by the control chart. In same cases the correction of average level may be a simple matter, such as changing a machine setting; in other situations, such as increasing an average level of strength, it may call for a program of research and development work. Once the control chart shows that a process is brought under control at a satisfactory level and satisfactory limits of variability, one may feel confident that the product meets specifications. This suggests the possibility of basing acceptance XVIprocedures on the control chart, using it to determine whether this happy state of affairs is continuing. Under these favorable circumstances substantial savings are often possible in costs related to inspection. Where inspection consists of destructive tests, it may be possible to reduce the number of items tested, thus saving both in testing cost and in the cost of the product destroyed. Most routine inspection of manufactured products is inspection by attributes, classifying each item inspected as either Accepted or Rejected ( with possibly a further division of rejects into spoilage and rework). This statement applies both to %100 inspection and sampling inspection. In such inspection it is common practice to make a record of the number of the items rejected. The practice of recording at the same time the number of items inspected is not so universal. However, if quality performance at one time is to be compared with that at another time, the record of total number inspected is just as necessary as the record of number rejected. The ratio of the number items rejected to the number of items inspected is the fraction rejected. Thus the Shewhart control chart for fraction rejected generally makes use of data that either are already available for other purposes or that can readily be made available. Simple statistical calculations provide control limits that tell whether as signable causes of variation appear to be present or whether the variations from day to day (or lot to lot, vendor to vendor, or whatever the classification basis may be) are explainable on change grounds. Control chart- for attributes (the p chart) is somewhat less sensitive than the charts for variables (X and R charts) and does not have as great a diagnostic value. Nevertheless, it is an extremely useful aid to production supervision in giving information as to when and where to exert pressure for quality improvement. It is a common experience for the introduction of a p chart to be responsible for substantial reductions in the average fraction rejected. In some instances the p chart will dis close erratic fluctuations in the quality of inspection, and its use may result in improvement in inspection practices and inspection standards. Moreover, the p chart often serves to point out those situations needing diagnosis of trouble by the control chart for variables. In addition to its use in process control, the p chart may be of great value in dealing with outside vendors. Vendors may differ both in the quality level submitted and in the variability of that quality level. It is particularly desirable to know whether the quality of product submitted by a vendor today is a reliable indication of what may be expected to be submitted next month. The p chart gives useful guidance on this point. Shewhart control chart for non-conformities per unit applies to two specialized stiuations. One is the case where a count is made of the number of non conformities of such type as blemishes in a painted or plated surface of given area, weak spots in the insulation of rubber-covered wire of a given length, or imperfections in a bolt of cloth. The other is the case of inspection of fairly complex assembled units in which there are a great many opportunities for occurrences of non-conformities of various types, and total number of non-conformities of all types found by the inspectors is recorded for each unit. As in other types of control charts, the control limits are set in away to detect the presence or absence of assignable causes of variation, and they therefore tell when to take action on the process and when not to do so. Experience indicates that xvuerratic variation in inspection and that the control chart for non-conformities per unit generally proves helpful in standardizing inspection methods. Although this type of control chart applies only to a limited number of manufacturing situations involving quality, it has broad application to many other types of situations commonly met in everyday life. In the fourth chapter, an activity schedule for SPC tool is prepared by using the data handled in the first three chapters. This schedule which begins with the definition of the process, gives the operations to be done for determination of the characteristics to be controlled, selection of the controlling tool and methods or machine and process capability analysis in an order. The output from all repeated operations or continuous processes varies to some extent. When that variation is due only to random causes, the process said to be in control. The range over which this variability occurs is often referred to as the process capability. A measure of this variability, usually in terms of the process standard deviation, is one of the parameters of the process. Capability studies are conducted for the purpose of estimating the process parameters and using this information to determine the ability of the process to meet standards and specifications. This information can also be used to establish new standards for the process or to modify or revise old ones. These standards can then be used as a basis for constructing control charts. When data have been collected from a process for a sufficient period of time, it becomes possible to describe the process statistically. This description involves estimates of the mean and standard deviation and assumption about the form of the distribution. This in turn leads to the estimated capability limits of the process, that is, the limits within which at least 99 percent of the observed values of the process variable can be expected to fall. In order to determine process capability, the process being studied must be in control. The concept of process capability refers to a process operating under normal conditions and not subject to any but random causes of variation. Consequently, when the data obtained for the capability study contain an extreme or unusual value and investigation determines that it is the result of an assignable cause which can be corrected, that value can be eliminated from the data. This would be true of any estimation procedure. In estimating process parameters for the purpose of determining process capability, certain practical problems sometimes arise. A process may be out of statistical control and the causes of the condition cannot be economically corrected. When this situation occurs, they suggest that capability limits be determined for the process as it exists, that is, with these assignable causes present. These limits will be inflated because the process will not be operating at its best. However, its better to exercise control at limits that recognize the inherent instability of the process than to abrogate any attempt to control the process at all. One capability limits have been established for the process, management has the responsibility for deciding whether or not they are satisfactory. If they are satisfactory, then decisions can be made based upon the existing capability. These decisions could include the acceptance or rejection of contract proposals, the pricing of product, the establishment of control charts, etc. If control charts are established based on the standard values and the limits determined by the capability study, their purpose is then to maintain control of the process at the existing limits. XVMIf management decides that the existing capability limits are not satisfactory, they are then faced with the problem of deciding whether or not the process can and should be improved. Process improvement may involve changes in working conditions, procedures, equipment, personnel, or other factors, bearing upon the performance of the operation. The amount of improvement which may be possible and the economic advantage resulting should be compared with the cost of obtaining the improvement. This cost should include the cost of new and improved equipment, the cost involved in establishing location, layout, or operating procedures, the costs involved in establishing and implementing new personnel training procedures, the costs related to the replacement of inefficient or ineffective personnel, and so forth. If improvement in process capability is possible and economically feasible, then control of the process can be exercised at whatever higher level can be achieved. It would, however, be futile to attempt to establish control of a process at a level that is beyond the capability of an existing process without first improving the process capability. While conducting a capability study, the first step is the collection of data from the process. Once the data have been collected, the statistical techniques should be used to compute the mean and the standard deviation of the sample values. Most process variables have a normal or approximately normal distribution. When the process variable does not have a normal distribution, then some other appropriate distribution should be used to calculate the capability limits of the process. If a normal distribution is assumed, the capability limits of the process are estimated from the sample mean and the sample standard deviation. The fifth chapter, which is the last one of this study includes a pilot study in Turkish Electrical Industries Sultançiftliği Electrical Motors Company, according to the developed activity plan. The main goal of this project is to implement this study in the whole factory. Further gains will be a reduction in the costs and an improve ment in the quality level that is already high. XIX
Benzer Tezler
- Toplam kalite kontrolunun sigara teknolojisinde uygulama olanaklarının incelenmesi
Başlık çevirisi yok
METİN ÖNER
Yüksek Lisans
Türkçe
1988
Endüstri ve Endüstri Mühendisliğiİstanbul ÜniversitesiÜretim Yönetimi Ana Bilim Dalı
DOÇ. DR. MEHMET GÜNEŞ GENÇYILMAZ
- Havacılık sektöründe istatistiksel proses kontrol: Uçak bakım süreçlerinin iyileştirilmesine yönelik bir uygulama
Statistical process control in the aviation sector: An implementation to improve aircraft maintenance processes
HAMİT HAMİDETTİN KUMURKAN
Yüksek Lisans
Türkçe
2021
Mühendislik Bilimleriİstanbul Teknik Üniversitesiİşletme Mühendisliği Ana Bilim Dalı
DOÇ. DR. NİHAN YILDIRIM
- Otel işletmelerinde toplam kalite yönetiminde bir araç: Kalite kontrol çemberleri
A Tool in total quality management at hotel organisations: Quality control circles
HÜLYA BAĞIRAN
Yüksek Lisans
Türkçe
1997
TurizmDokuz Eylül ÜniversitesiTurizm İşletmeciliği Eğitimi Ana Bilim Dalı
PROF. DR. ALP TİMUR
- Bir toplam kalite yönetimi değerlendirme modeli: Türkiye uygulaması
A Total quality management evaluation model: Application in Turkey
DİLEK ERERTEM
- 1011 Ordudonatım Ana Tamir Fabrikaları'nda istatistiksel kalite kontrolü ve uygulaması
Statistical quality control and its application in 1011 Ordudonatım Ana Tamir Fabrikası
BORA ÖZTEKİN
Yüksek Lisans
Türkçe
2003
İşletmeKara Harp Okulu KomutanlığıSavunma Yönetimi Ana Bilim Dalı
DR. TAYFUN ÖZDEMİR