Hazırlık sürelerinin analizi ve azaltılması

Abstract

Kesikli parti tipi üretim yapan sistemlerde "tam zamanında" idealine ulaşmak için aşılması gereken engellerden biri ve en önemlisi hazırlık sürelerinin azaltılarak küçük parti üretiminin yapılabilmesidir. Hazırlık sürelerinin azaltılması birinci derecede, hazırlık işlerinin doğru ve tam olarak tanımlanmasına bağlıdır. Bundan sonra yapılacak sistematik bir çalışma ile hazırlık sürelerinin azaltılması gerçekleştirilebilir. Hazırlık süreleri genellikle direk maliyete etki ederler, böylece hazırlık sürelerinde sağlanacak azalma doğrudan hazırlık maliyetlerini ve toplam üretim maliyetlerini iyi yönde etkiler. Hazırlık süreleri firmalar tarafından istenmeyen israf sürelerdir. Bunun nedeni ürüne hiçbir ek değer katmayan ama yapılmak zorunda olunan işlerdir. Bu yüzden bazı firmalar hazırlık işlerini işçilik kayıpları olarak görmektedirler. Bunun yanında hazırlık süreleri azlatma çalışmaları hiçbir risk içermemektedirler. Yani üst yönetime ilgiden başka hiçbir yük yüklemeyen iyileştirme çalışmalarıdır. Bu çalışmada hazırlık sürelerinin azaltılması için her koşulda izlenebilecek bir yol önerilmiştir. Ayrıca, hazırlık sürelerinin azaltılmasında karşılaşılan en büyük sorun, izlenecek sistemli bir yolun olmamasıdır. Dolayısıyla, bu konuda odaklanılmıştır. Bunun yanısıra, üretim ortamlarımızda geleneksel tezgahlar yerlerini Bilgisayar Nümerik Kontrollü (CNC) tezgahlara bırakmaktadır. CNC tezgahların çoğunlukta olduğu üretim ortamlarında, hazırlık sürelerinin azaltılması, tezgahlarda kullanılan takımların, efektif olarak planlanmasına kuvvetlice bağlı kalmaktadır. Bundan dolayı yapılan uygulamada kullanılan takım planlaması yaklaşımları da ayrıca her koşulda hazırlık sürelerini azaltma çalışmaları için kullanılabilmektedir. Hazırlık sürelerini azaltma için geliştirilen sitematik yaklaşımın bir cam kalıp üretim sisteminde uygulaması yapılmıştır.

The issue of setup time reduction is important for firms seeking to incorporate procedures and concepts such as flexible manufacturing systems (FMS) and just- in-time (JIT) manufacturing to improve manufacturing productivity. Setup time reduction is a fundamental tenet of just-in-time (JIT) manufacturing and flexible manufacturing systems (FMS). With the great success of Japan 's manufacturing there has been a sudden interest in theoretical models for the reduction of setup time (cost). This has the potential for producing many benefits such as reduced inventory costs, improved quality, and increased equipment utilisation. In fact, Japan 's ability to control setup time and slash lead time were key elements in its successful implementation of JIT. There are also many benefits of setup time reduction. These benefits include small-lot production capabilities which yield savings in storage, handling, and inventory carrying costs; reduced lead times; increased quality; increased flexibility; and increased capacity. Setup time is defined as the time it takes to go from the production of the last good piece of a prior run to the first good piece of a new production run. Setup, therefore, not only includes changing fixtures, dies, and / or tooling, but also tear down, cleanup, inspection, trial runs, and any material handling, administrative work, idle time, etc., that occurs between the production of good parts. Assuming that the cost per setup is equal to the setup time multiplied by a standard labour hourly cost. Thus, there is a positive correlation between setup time and cost per setup. Using setup time reduction interchangeably with setup cost reduction, or simply using blanket term, setup reduction, instead of distinguishing between setup cost reduction and setup time reduction. Recently, many authors indicated that significant reductions in setup time can be realised by applying common sense and experience to setup processes, with little or no capital investment. Often this setup depends on which jobs follow and this is referred to as sequence dependent setup time. However, one of the most popular approaches to job shop problems is to apply a simple priority rule to the queue at each machine. These simple priority rules however are usually fixed and, as such, do not respond to changes in the manufacturing system. Intuitively it would seem that the amount of setup time should be a factor to be taken into account when sequencing jobs in a practical manufacturing environment. The sequence dependent setup time had a significant impact on the shop performance. Ordinary sequencing rules could not provide as good shop performance as setup-oriented rules that accounted for sequence dependency. xiii The concept of eliminating waste by reducing setup time is rooted in the SMED (Single Minute Exchange of Die) methodology developed by Shingo (1987). Figure 1 depicts the SMED methodology and typical responses to SMED in terms of programs and efforts in the machining environment. Step 1 Separate internal setup actions from external setup actions Internal Setup -> Setup While Machine Is Stopped External Setup -» Setup While Machine Is Operational Step 2 Convert internal setup actions to external setup actions Internal Setup Actions Must Not Be Interrupted To Perform External Setup Step 3 Eliminate adjustment Adjustment Consumes As Much As 50% of The Total Setup Time (Setup time is the time from the completion of last good part A to the acceptance of the first good part B) Step 4 Single minute setup Responses to SMED methodology (i) Identify and Separate Internal and External Setup Actions A. Videotaping of setups B. Classify / analyse data (ii) Convert Internal Setup To External Setup A. Fixture setup, tool setup, gauge acquisition external B. Implementation of machine resident common tool packets (Hi) Eliminate Adjustment - Keyed location of fixtures, angle plates and other work-holding devices to establish pre-defined location on worktable(e.g. quick change tooling) Note : The two problems associated with adjustment - Part / Fixture location not equal to programmed location - Machine dimensions not equal to programmed dimensions (iv) Make Setup One-Touch Figure 1. Single Minute Exchange of Die (SMED) methodology. Steps 1 and 2 of SMED can be accomplished in machining as shown, and by other methods, while Step 3 promotes the concept of eliminating internal setup time by eliminating adjustment Adjustment is eliminated in machining by a combination of pre-defining the fixture location relative to machine tool work table using sine keys (or other quick setup fastening and locating devices), and quick setup fixturing to pre-define part positioning in a fixture. In this way Step 4 of SMED may be accomplished. Tool setup actions generally take the most time in total setup time. Because, manufacturing with NC machine, tool length adjustment process takes more time xiv than other adjustments. In table 1, tool length adjustment process with NC machine is shown. 1. Make a trial cut. 2. Check (inspect) the resulting machined dimension. 3. Mentally compute an offset given by equation (1). equation 1: Offset - programmed dimension - actual dimension. 4. Enter a tool length adjustment offset into the controller. 5. Repeat steps' 1-4 until machined and programmed dimensions are equivalent (using as many parts as necessary). Table 1. Tool length adjustment process with NC machine. To adapt the consequence of setup time for the manufacturing process and the enterprise success, it is necessary to consider systematically the setup process in the production. In this thesis shown a method, which enables to analyse the setup process and derive suggestions for setup time reduction upon the basis of ascertained weaknesses. This method must consider all influences on the setup process. They can be divided into setup components (production facilities) and production organisation, as well as personnel. Organisation -> a. production control b tool organisation c. planning of setup d. technical document e. maintenance Personnel -> a. qualification b. motivation Production -> a. machine Facilities b. machine control c. tools d. fixture e. clamping devices f. measuring tools g. auxiliary devices h. periphery i. material These are areas of influences on the machine near setup process. To analyse a machine near setup process it is advisable to structure the procedure step by step. The first step is to determine the investigation field. This means, the work centres should be selected, on which setup process recording should be occurred. The data recording on the selected machines forms the second step. The evaluation and valuation of the recorded data are the third step. Output of this step is the listing of the main weaknesses and disturbances in setup process on the machine. xv The fourth step is to derive the suggestions to eliminate weaknesses and reduce machine down time caused by these weaknesses of setup process. The realising of the selected suggestions forms the final of analysis. The setup process was in the setup analysis divided into so-called setup elements and setup steps independently. The setup elements are formed out of the combination of process neutral setup objects and setup activities. One setup step association with one setup element, is dependent on external or internal feasibility of separate setup elements. The activities of preparation and ending work describe the setup operations on machine, which could be done parallel to machine running time (external). The setup steps of clamping down and clamping up contain the necessary activities of disassembling, rebuilding, installing or adjusting for tools, fixture or workpiece. After finishing these activities it is theoretical possible to begin the production of next lot Putting into practice this turns out to be impossible, because normally it needs one or more test runs, in order to obtain the required quality of workpieces. For the performing of the data recording on the selected work centres, the aid and proceeding have to be determined, so as to register the setup process data with required accuracy and to obtain the data in clear valuable form. The observer records what activity has been carried out by the personnel on the machine. This activity has to be converted into suitable code with the predefined list of setup object and setup activity. E.g. the observed activity "installing tool" results in the setup element code C-33 according to predefined list. The results of evaluation of setup elements form the fundament to derive the suggestions, which lead to reduce machine down time caused by weaknesses associated with setup objects. So, it is necessary to find out the cause of weaknesses located with the help of data evaluation. In suggestions deriving, it is important to distinguish between the effect and cause. The results of setup data evaluation describe the effect. However, the cause, which has to be removed. Finally, derived suggestions realising, the weaknesses may be removed to reduce setup time. XVI In the case study of thesis, the weaknesses of setup process associated with "tool" setup object. To remove these weaknesses, derived suggestions are "1-make classification and standardisation of tools, 2-job scheduling to minimise the number of tooling changeovers must be made, 3-tool clustering must be made and 4- tool planning must be made." To make job scheduling, TSP (travelling salesman problem) approach was used. To make tool clustering, two clustering approaches were used. One is ROC algorithm and other one is a neural network approach. In tool planning, KTNS (keep tool needed soonest) rule was used. Travelling Salesman Problem : In the TSP, a salesman must visit each city in his territory and then return home. In our case, the worker must perform each job and then return to the starting condition. Several mathematical formulations exist for the TSP. One approach is to let xij be 1 if city j is visited immediately after i and 0 otherwise. A formal statement is then JV N for all i JX = 1 for all j i=l no subtours Xij = 0 or 1 Use of Eastman 's algorithm with Branch and Bound Technique, the scheduling problem was solved. Tool Planning : In the case study, for tool planning we have two assumptions. First, assuming that tools are not needed elsewhere, there is no advantage in ever having less than M tools on the machine. Second, assume an ordering for jobs on machine is given. In KTNS we only remove as many tools as necessary to make way for the next job. The tools removed are those that will not be needed again until the longest time (most jobs) in the future.