Tekstil endüstrisinde makina girişimi
Machine interference
- Tez No: 14276
- Danışmanlar: PROF.DR. ATAÇ SOYSAL
- Tez Türü: Yüksek Lisans
- Konular: Endüstri ve Endüstri Mühendisliği, Industrial and Industrial Engineering
- Anahtar Kelimeler: Belirtilmemiş.
- Yıl: 1990
- Dil: Türkçe
- Üniversite: İstanbul Teknik Üniversitesi
- Enstitü: Fen Bilimleri Enstitüsü
- Ana Bilim Dalı: Belirtilmemiş.
- Bilim Dalı: Belirtilmemiş.
- Sayfa Sayısı: 65
Özet
ÖZET MAKİNA GİRİŞİMİ Tekstil sanayiinde bir işçiye birden fazla makina verilmektedir. Bu da standart randımanın hesaplanmasında ortaya Makina Girişimi adı altında bir problem çı karmaktadır. Standart randıman değeri Teşvikli Ücret Sistemlerinde, Üretim Planlama ve Kontrolunda, iş Gücü Planlamasında ve Maliyet hesaplamalarında kullanılmaktadır. Bu nedenle standartlar hesaplanırken titiz bir çalışma yapıl malıdır. Makina girişimi kaybı, makinaların servis görmeleri için işçi beklemeleri sıra sında verilen kayıptır. Makinaların sistematik servisli ve tesadüfi servisli olma larına göre çözüm şekilleri değişmektedir. Bu tez kapsamında makinalarda düzenli duruşlu periyodik servis, tesadüfi du- ruşlu makinalarda ise periyodik ve tesadüfi servisler için çözüm yollan veril miştir. Tesadüfi duruşlu makinalarda makina girişimi hesaplarının çok daha karmaşık olduğu ve çözüme ulaşmak için olasılık yöntemlerinin kullanıldığı görülmüştür. Tesadüfi makina girişimi probleminin yaşandığı bir işletme örnek olarak ince lenmiş ve makina devrinin basitleştirilmesi çalışmaları yapılmıştır.
Özet (Çeviri)
SUMMARY MACHINE INTERFERENCE It is unlikely that any industrial engineer working in the textile industry is not familiar with the phenomenon of machine interference and the problems asso ciated with it. Taking a simple case and ignoring complications for the moment, we may pic ture a weaver in charge of a group of looms and imagine that at the instant he is observed all his looms are running. His aim is to keep them running and pro ducing as much of the time as possible, but any loom is liable to stop at random. When that happens, he knows what to do. If he is free, he goes at once to the stopped loom and fixes or corrects the cause of the stoppage, and he restarts the loom. Again all looms are running. The time loss is a characteristic of the material being manufactured and is independent of the number of looms in group. Imagine that the next time the stop occurs, before the weaver can get the stopped loom back into production, another loom stops. The second loom has to wait untill the weaver is free to attend to it. How long that will be, will depend on the stage reached by the weaver in his work on the first loom. The time loss representing waiting time imposed by the overlap of the ma chines stoppages is called“ MACHINE INTERFERENCE ”. Production is lost not only while the machines are under attention but also while the machines are stopped awaiting attention. The rates of production per machines, and per operator, thus depend not only on the rate of occurrence of stops and on the“Attention Time”per stop, but also on the number, n, of machines per operator. A multimachine assignment is one where more than one machine is operated by a single operator. The primary purpose of this thesis is to give methods that will provide help in determining how many machines to assign to an operator and how to establish equitable production standarts on these multimachine as- singments. VITherefore, the mathematical prediction of the rate of production from informa tion about the stoppage-rates, the attention times and the number of machines per operative is shown. There are several ways in which such a prediction can be useful in practice; 1. We may wish to determine the optimum number of machines per operator on economic considerations, and to do this we must be able to work out the rate of production corresponding to any given n. 2. We may contemplate a modification to the machines to eliminate certain stoppages. What should the new value of n be, and will the gain in production be sufficient to justify the modification? 3. We may have measured rates of production from which to make a comparison to two alternative processes, but for one process there were say 10 machines per operator and for the other 15 machines per operator. Is an apparent diffe rence between the processes due solely to the differing values of n ? The solution to many multimachine assingments involves the calculation of ma chine interference idleness which are caused by assigning more than one ma chine to a single operator. Machine interference idleness is the time that a machine is idle because the operator is servicing another machine in the group. Operator idleness is the time that the operator is idle because all the machiness in the group are run ning automatically. Two general types of solutions to multimachine assignment problems are pre sented in the thesis, depending on whether machines are randomly serviced or systematically serviced by the operator. Looms have been mentioned as a type of machinery that provides the best known example of interference, however, there are many other machines in the textile industry and in other industries, which have similar characteristcs and are subject to loss from similar causes. These machines all have a running time which is long in relation to the operator attention time, so that they can be assi gned in numbers to one operator. Once this is done, overlapping stops can occur in most cases, but not necessarly in all. If the stops are regular, they can be scheduled and interference may be avoided. vuMost machines have some regular stops and some random stops, and the effect is practically the same as if all were random. Spinning machines and cartoning machines are the examples of machines that have random servicing demands. In the case of weaving with looms, the occur rence of yarn breakage and therefore loom stoppage is entirely random. Solu tions based on the laws of probability are used for these assignments. In Chapter 4, probabilistic analysis of the problem is discussed. Other machines, such as plastic-mold presses and gear-hobbing machines have regular ( systematic ) servicing demands. The loading and automatic processing times of these machines are predictable as to order of occurance and elapsed time required for both operating and servicing. Solutions to problems involving these machines are determined in Chapter 2. When stops occur in regular sequence, it is obvious that servicing should be synchronised to conform to this sequence. A definite patrol cycle results, and is designated as cyclic servicing. When stops are not synchronised or where they occur at random, the job method may call for operator to always tend the nea rest stopped machine, thus representing random servicing. However, even where stops opccur at random, cyclic servicing is often preferred and this is dis cussed in Chapter 3. Considerable effort has been spent by mathematicians, in measurement of ma chine interference. This problem, the correct assignment of man to machine, has excited the imaginations of many competent mathematicians who have re markably approached reality through their exploration of these probabilistic model. Their successes are numerous and research is continuing, but often these formulas only approximate the real situation. The prevalent mathemati cal techniques used in solving machine interference problems are binomial, normal, and exponential. In the case of the ideal automatic machines which have only scheduled stops, is relatively easy to calculate interference as shown in Chapter 2. But the direct timing efforts in random machine interference have generally proved fruitless, because of the difficulties encountered in timing simultaneous events, rating the operator, and the like, and because of the necessity of develo ping interference curves or tables covering the whole range of possible assi gnments entailing different numbers of machines and work loads. vinThe solutions discussed in the thesis, are based on quite restrictive assump tions and there are numerous complications arising in practical applications which can make these assumptions seriously untrue. For example solutions are given for a number of idealized versions of the problem on the assumption that the operative is always available to attend to stopped machines. In most textile processes, however, the operative will have other duties to per form and will be permitted to take some time for relaxation. The amount of the interference as well as the output from a system depend on: 1. The distribution of loading-unloading times and processing times, 2. Availability of the operator, 3. The priority rule used to select the next job to be processed, 4. The number of machines assigned to the operator. Useful approximations to the practical case, like in the example, are given in the Paper too. In many practical applications the most unrealistic assumption in the analysis is the one that the operative is always available to restart stopped machines, that is that the restriction on service is purely one of capacity and not one availabili ty, m fact, in machine interference problems some allowance must always be made for relaxation and personal needs, and quite often also the operative will have other duties. We call all demands on the operative's time, other than the restarting of stop ped machines, ancillary work. It has also been assumed that the selection of a particular machine for service is un correlated with the service time that will be necessary to restart that ma chine. There are, however, various ways in which these assumptions may fail It may happen that the operative tends to work quickly when the number of machines awaiting attention is large. In general, a tendency for the service time to be shorter when the number of machines awaiting attention is large will increase the rate of production. IXIn some applications namely where the machines are spread over a considerable area and where it is important to examine each machine frequently whether or not it has stopped (spread ancillary work), the operative may adopt a scheme of regular patrolling. There may be absence of statistical uniformity. For example, while the arrivals may be completely random over quite long time periods, the mean rate of arri val may vary in time. In some applications this may be systematic variation through the day, with machine stops it is more likely to be superimposed ran dom variations between different batches of raw materials, etc... An example, a problem of the machine interference from a real world situation is given in Chapter 5. The example is from textile industry. In the company, tire cord fabric is produ ced. The cord fabric, as a semi-finished product, is a critical input of the tire manufacturing industry. It is manufactured in three stages: The first step is the Ply and Cable twisting operation to optimize strength and fatigue resistance properties. The second stage is weaving. Greige fâbric rolls are produced in accordance with customer specifications Chemical treating and hot stretching is the final process. Machine Interference Problem occurs in the first and second stages in the com pany. The solutions of Dale Jones are used to tackle with the problem. There are two stoppages in the process. One of them occurs in regular sequence,and the other one, the breaks of the yarn, have random servicing demands. It is as sumed that all fit to random servicing requirements. In this study, the machine cycle was also simplified. The question may be asked: Why used mathematics? Why not measure interfe rence as it actually occurs on the job? A boy came to the bank of aswift stream an saw a trout. He threw in a cork to mark the spot and went to get his rod. When he got back, the trout had gone, and so had the cork. The idea of measuring interference looks about as useful as the idea of marking the spot in a strem. It would be wearisome to list the change causes that effectinterference. Even with the most ingenious system of checks and corrections, there would be little hope of obtaining meaningful results. Calculated machine interference thus provides a more valid and logical basis than mere observation in providing proper time study allowance and determi ning expected efficiencies in multimachine assingments. XI
Benzer Tezler
- Mekanik sistemlerde temas problemlerinin incelenmesi
The investigation of contact problems in mechanical systems
İSMAİL KAYA
Doktora
Türkçe
2024
Makine Mühendisliğiİstanbul Teknik ÜniversitesiMakine Mühendisliği Ana Bilim Dalı
PROF. DR. CEMAL BAYKARA
- Makina halılarının kalite özellikleri üzerine bir araştırma
An Investigation of the quality properties of machine manufactured carpeds
MEHMET ALİ ERKESİM
Yüksek Lisans
Türkçe
1995
Tekstil ve Tekstil Mühendisliğiİstanbul Teknik ÜniversitesiPROF.DR. MUSTAFA KÖSEOĞLU
- Descriptive and predictive statistical modeling of a ring spinning process
Ring iplik eğirme prosesinin betimsel ve kestirimsel olarak istatistiksel modellenmesi
REYHAN İŞSEVER
Yüksek Lisans
İngilizce
2016
Kimya MühendisliğiBoğaziçi ÜniversitesiKimya Mühendisliği Ana Bilim Dalı
DOÇ. DR. BURAK ALAKENT
- Tekstil endüstrisinde doğrusal programlama ile üretilen planlama ve bilgisayar uygulaması
With lineer programming production planning in textil industrie and computer application
SEVİL ARICANLI
Yüksek Lisans
Türkçe
1996
Bilgisayar Mühendisliği Bilimleri-Bilgisayar ve KontrolYüzüncü Yıl ÜniversitesiBilgisayar Mühendisliği Ana Bilim Dalı
YRD. DOÇ. DR. YÜCEL YILMAZ
- Endüstriyel borlama ve tekstil endüstrisinde bir uygulama
Industrial boriding and an application on textile industry
YUSUF KARAMAN
Yüksek Lisans
Türkçe
2003
Makine MühendisliğiSüleyman Demirel ÜniversitesiMakine Eğitimi Ana Bilim Dalı
DOÇ. DR. ABDULLAH ÖZSOY