Wyniki wyszukiwania - BazTech

1

Machine learning in SMED

Kutschenreiter-Praszkiewicz I.

Journal of Machine Engineering

|

2018

|

Vol. 18, No. 2

31--40

EN

The paper discusses Single Minute Exchange of Die (SMED) and machine learning methods, such as neural networks and a decision tree. SMED is one of lean production methods for reducing waste in the manufacturing process, which helps to reorganize a conversion of the manufacturing process from current to the next product. SMED needs set-up activity analyses, which include activity classification, working time measurement and work improvement. The analyses presented in the article are focused on selecting the time measurement method useful from the SMED perspective. Time measurement methods and their comparison are presented in the paper. Machine learning methods are used to suggest the method of time measurement which should be applied in a particular case of workstation reorganization. A training set is developed and an example of classification is presented. Time and motion study is one of important methods of estimating machine changeover time. In the field of time study, researchers present the obtained results by using (linear) multi-linear regression models (MLR), and (non-linear) multi-layer perceptrons (MLP). The presented approach is particularly important for the enterprises which offer make-to-order products. Development of the SMED method can influence manufacturing cost reduction of customized products. In variety oriented manufacturing, SMED supports flexibility and adaptability of the manufacturing system.

2

Customer Oriented Product Planning Procedure

Kutschenreiter-Praszkiewicz I.

Applied Computer Science

|

2016

|

Vol. 12, no 2

22--32

EN

One of the most important issues in product planning is to identify customer needs and combine them with product technical and trade characteristics. Identification of customer needs was discussed, and product decomposition method was presented in the paper. The Quality Function Deployment method was suggested to be applied as a product and production process data integration tool, where engineering characteristics of a product are combined with its trade characteristics.

3

Wybrane zagadnienia technicznego przygotowania produkcji wyrobu innowacyjnego

Kutschenreiter-Praszkiewicz I.

Zarządzanie Przedsiębiorstwem

|

2015

|

Vol. 18, nr 2

23--32

EN

The paper presents innovative product planning issues with the use of QFD method and artificial neural network. Among many methods of data analysis focused on innovative product planning especially important are those one which take into consideration product engineering characteristic and time consumption and cost related to production process of product characterised by given attributes. The paper presents algorithm of innovative product data analysis, which helps identified product attributes notice by customer and producer, setting values of innovative product target attributes, alternatives identification, evaluation of standard product, determination product changes and time consumption of innovative product production process and appointed time of production process establishing.

4

Planowanie procesu produkcyjnego wyrobu innowacyjnego

Kutschenreiter-Praszkiewicz I.

Pomiary Automatyka Robotyka

|

2015

|

R. 19, nr 1

57--64

PL

W artykule przedstawiono problematykę planowania procesu produkcyjnego wyrobu innowacyjnego a w szczególności została przeprowadzona analiza zadań technicznego przygotowania produkcji TPP ukierunkowanych na automatyzację wybranych zadań planowania wyrobu innowacyjnego. Zwrócono uwagę na modułowość wyrobu, jako sposób na spełnienie indywidualnych wymagań klienta przy ponoszeniu niewielkich kosztów. Istnieje wiele sposobów udoskonalania istniejących rozwiązań prowadzących do satysfakcji klienta – literatura podaje między innymi metodę TRIZ, QFD. W artykule poddano analizie problematykę identyfikacji potrzeb klienta. Przedstawiono zagadnienia dekompozycji wyrobu. Zastosowanie koncepcji wyrobów modułowych wymaga opracowania modeli, które w uniwersalny sposób przedstawią strukturę wyrobu wraz z możliwymi alternatywami. W artykule przedstawiono model atrybutowy wyrobu. Zaproponowano zastosowanie metody QFD do identyfikacji wymagań klienta oraz określenia korelacji między wymaganiami klienta a cechami technicznymi wyrobu oraz korelacji między kolejnymi atrybutami generowanymi w ramach etapów TPP. Przedstawiono możliwość zastosowania sztucznych sieci neuronowych, jako metody generowania danych niezbędnych dla potrzeb planowania TPP wyrobu innowacyjnego. Zaproponowano regułową metodę oceny podobieństwa wyrobów.

EN

Paper presents innovative product production planning issues, especially technical product preparation tasks focused on chosen data assessment ware taken into consideration. Product modularity was taking into consideration as a cost reduced method for customer particular needs fulfilment. There are a lot of methods applicable for product development – according to literature review TRIZ and QFD are especially promising. In the paper the customer needs analysis was presented. The product modularity issue was presented. The idea of product modularity needs product models, which taking into consideration product structure and modules alternatives. The product attribute model was presented in the paper. The QFD method was apply for customer needs identification and correlation analysis for product structure and another data created during production process preparation. Artificial neural network was used for innovative product planning data prediction. Rule based method product similarity assessment was applied.

5

Product configuration using knowledge based system

Kutschenreiter-Praszkiewicz I.

Journal of Machine Engineering

|

2013

|

Vol. 13, No. 2

26--35

EN

Product configuration is focused on achieving customer satisfaction. Configuration methods should use previous experiences related to product changes One of the methods of storing experiences is elaboration of knowledge bases with the use of proper knowledge representation method. In the paper identification of customer needs was discussed, as well as product decomposition methods were presented. The Quality Function Deployment (QFD) method was suggested to be applied as a product and process data integration tool, where engineering characteristic of a product was combined with its trade characteristic. The paper shows that thanks to the application of knowledge based system in production preparation it is possible to support decision making connected with product selection and evaluation for a particular customer.

6

Application of knowledge based systems in technical preparation of machine parts production

Kutschenreiter-Praszkiewicz I.

Advances in Manufacturing Science and Technology

|

2013

|

Vol. 37, nr 1

19—29

EN

The paper presents problems of technical product preparation tasks aimed at product selection and redesign for particular application supported by Al methods. The identification of customer needs was discussed, as well as the issues of product decomposition were presented. The QFD method was suggested to be applied as the product and process data integration tool, where the engineering characteristic of a product was combined with its trade characteristic. In the industrial product characteristic, the delivery time, which can be fixed with the use of graph based scheduling method, is particularly important. The paper shows that thanks to the application of KBS methods in production preparation it is possible to support decision making connected with product selection for particular use.

PL

W pracy przedstawiono analizę zadań technicznego przygotowania produkcji (TPP) ukierunkowanych na dobór i adaptację wyrobu do danego zastosowania wspomaganą metodami Al. Analizie poddano zagadnienia identyfikacji potrzeb klienta. Omówiono zagadnienia dekompozycji wyrobu. Zastosowano metodę QFD jako narzędzia integracji danych z zakresu TPP. Charakterystykę techniczną wyrobu powiązano z charakterystyką handlową obejmującą m.in. termin realizacji. Wykazano, że zastosowanie metod KBS w przygotowaniu produkcji umożliwia wspomaganie decyzji w zakresie doboru wyrobu do danego zastosowania.

7

Integration of product design and manufacturing with the use of artificial intelligent methods

Kutschenreiter-Praszkiewicz I.

Journal of Machine Engineering

|

2011

|

Vol. 11, No. 1-2

46--53

EN

Product and process design is time- and cost consuming [15]. To reduce the cost of product development, it is necessary to integrate product and process design. The proposed product planning approach integrates activities involved in product design and manufacturing process. The aim of this paper is to develop a method of knowledge integration about customer needs, product and process characteristics. The range of analyses is limited to mechanical product type manufacturing for institutional customers. Customer needs are focused on functional characteristics of the product and the trade characteristics include product price, timing and warranty. Integration of functional requirements, product and process characteristics is needed to select the best product from a catalogue and adapt it to particular customer needs. From the given set of products, where a product is described by a set of attributes, the subset is chosen which roughly satisfies customer needs. Basing on artificial intelligent (AI) methods, data related to redesign and production processes is estimated.

8

Reliability estimation in product configuration issue

Kutschenreiter-Praszkiewicz I.

Diagnostyka

|

2010

|

nr 2(54)

37-40

EN

In the product configuration problem of mechanical product type reliability is one of the most important requirements. Product adjustment to customer expectations, which mean product customization, decides about its commercial success. Development of methods helpful in the early phase of product design is needed to product customization. The early phase of design problems is focused on deriving the optimal solution that satisfies some objective functions like e.g. reliability. The proposed product decomposition method and system reliability estimation were presented.

PL

W problematyce zarządzania konfiguracją produktu przemysłowego niezawodność jest jednym z podstawowych wymagań. Dostosowanie produktu do wymagań klienta, czyli kustomizacja decyduje o sukcesie rynkowym. Rozwój metod wspomagających wczesne etapy projektowania produktu jest niezbędny do jego kustomizacji. Wczesne fazy projektowania produktu wymagają prowadzenia optymalizacji z uwzględnieniem niezawodności jako funkcji celu. W artykule przedstawiono proponowaną metodę dekompozycji produktu oraz szacowania jego niezawodności.

9

Product configuration use chosen artificial intelligent method

Kutschenreiter-Praszkiewicz I.

Applied Computer Science

|

2010

|

Vol. 6, no 2

71--79

EN

Product adjustment to customer expectation, which means product customization, decides about its commercial success. Developing methods helpful in the early phase of product design is needed to product customization. The aim of this paper is to present application of knowledge-based systems (KBS) in product customization.

10

Application of artificial intelligence in project management under risk condition

Kutschenreiter-Praszkiewicz I.

Applied Computer Science

|

2009

|

Vol. 5, no 1

69--80

EN

Risk management problem was shown in the paper. Relationship between risk management and production process scheduling was analyzed. Different types of data analysis were presented. Toothed gear production process was taken as an example of task timing estimation.

11

Zastosowanie sieci neuronowych do wyznaczania czasu przygotowawczo-zakończeniowego operacji technologicznych.

Kutschenreiter-Praszkiewicz I.

Mechanik

|

2007

|

R. 80, nr 1

48-48

PL

Opisano metodykę i wyniki badań prowadzących do zaprojektowania architektury sieci neuronowej oraz zbudowania zbioru uczącego, a w szczególności sprecyzowania cech, które pozwolą określić czasy przygotowawczo-zakończeniowe operacji technologicznych.

EN

A description with the research results back up is provided to explain development of the neuron system architecture and creation of a learning set and in particular, to specify the features, which would provide the grounds for estimation of the process setup times.

12

Wykorzystanie sztucznych sieci neuronowych do prognozowania czasu projektowania przekładni zębatych w warunkach niepewności i ryzyka

Kutschenreiter-Praszkiewicz I.

Archiwum Technologii Maszyn i Automatyzacji

|

2007

|

Vol. 27, nr 2

113-120

PL

W artykule przedstawiono problematykę określania pracochłonności projektowania przekładni zębatych. Wydzielono etapy projektowania oraz określono wektor cech charakteryzujących prace projektowe z punktu widzenia ich pracochłonności. Z wykorzystaniem sieci neuronowych opracowano model przebiegu prac projektowych. Na podstawie analizy wrażliwości dobrano odpowiednią strukturę sieci neuronowej.

EN

The paper presents the methodology of labour consumption prediction, on the tooth gear design example. Phases of design process were established; input vector connected with labour consumption was given for each phase. Design process was modelled by neural network. Sensitivity analysis was made and a structural arrangement was given.

13

Zastosowanie algorytmów ewolucyjnych w zarządzaniu przedsięwzięciami technicznego przygotowania produkcji w warunkach ryzyka

Kutschenreiter-Praszkiewicz I.

Zarządzanie Przedsiębiorstwem

|

2006

|

Vol. 9, nr 2

39-44

EN

In this paper problem of optimisation design process is developed. There was shown how the process production preparation is changing. The latter part of the article is devoted to discuss the possible applications of Genetic Algorithm in optimisation of engineering design process. Product development and design are directly connected with the enterprise strategy and the marketing research results. The designer is to create new products or to modify the old ones according to the marketing department's demand. The product must fulfil the requested technical parameters, but also the demands concerning the economical production, quality and reliability. A company, to be competitive must continuously improve its products and processes. The development of company can be realized using its own human potential, or using transfer of technique. The transfer of technique can be realized using formal or informal methods. The formal methods are for example: purchase of licence, patents, use of consulting firm. The informal methods are: exchange of technical staff, conferences, trade, exhibition, professional training. Design process is a halfway between research process and routine organization procedure of production preparation. Development and research work is connected with uncertainty and risk. Risk is connected with internal and external factors. Internal factors include: human mistakes, information computing mistakes, machine failure. The external factors include change of law. atmospheric conditions. To prepare the plan of research work activity we can use one of the following methods: CPM, CPM-COST, PERT, PERT-COST, GERT, but only GERT respects risk and uncertainty of activity. In this paper was shown haw to join GERT and Genetic Algorithm.

14

Chosen problems of modelling engineering design process

Kutschenreiter-Praszkiewicz I.

Applied Computer Science

|

2006

|

Vol. 2, no 2

141--148

EN

An overview of the various models for engineering design process was presented. General Design Theory, Basic Model, Real Design Model were described. Type-2 fuzzy sets implementation on engineering design process was described.

15

Zastosowanie sztucznych sieci neuronowych w organizacji procesu produkcyjnego

Kutschenreiter-Praszkiewicz I.

Zarządzanie Przedsiębiorstwem

|

2003

|

Vol. 6, nr 1

24-29

EN

The paper presents methodology of time standard determination concerned with design and production part of machine. Presented methodology is used to determination of cost and term of process production. This methodology is especially important in medium and small batch production methods in machine industry. Market economy conditions force actions upon enterprises. aiming at quality improvement, production cycle shortening and cost reduction. To reach this goal it's necessary to determine precise operational times, on the base of which we can establish the costs of operations to be carried out. Good organization is determined by extend of using of enterprise resources such as material, human, machine, finance, information. The uses of schedule based on time standard well prepared help to manage the production process. This is especially important in small and medium batch production methods. In these conditions it is necessary to design a lot of different construction. In this situation methods of work measurement based on analytical estimating have greater meaning opposed to conventional ones, which use more time and has a larger cost. Methods of work measurement, known from literature, aren't sufficient. Enterprise aided computer systems need a new fast and easy to use methods of work measurement. Traditional work measurement methods are too difficult and expensive in case of small and medium batch production. Majority of methods of work measurement are useful in case of production task. But a lot of task realized particularly in technical production preparation can't measure easy. This kind of work is characterized by: variable of tasks, no repeatable conditions, long term realization, determined by a lot of factors, a lot of conceptual work. For this kind of job can't use the conventional method of work measurement based on time study. It's necessary to create a new method based on artificial intelligence methods. There are lot of methods considering an artificial intelligence methods. Author used the pattern recognition based on neural network. Those methods characterized by "teach yourself" procedures, considered continuous changes in enterprise. To determine work time standards for composed tasks determined by many conditions we can use the pattern recognition. Pattern recognition consists on allocation of a target to appropriate class. This classification realized use lo neural network. Neural network is based on model of human nerve cell. Neural network opposed to sequence data processing enables parallel data processing. Based on presented model a new method of work measurement has been elaborated. A new method of work measurement may be based on following steps. 1. Definition of purpose of time standards, for example for planning, wages... It does determine the accuracy (confidence interval) of created time standards. In this step important time kinds must be chosen. 2. Definition coefficients determining the labor consumption. Coefficients for design task are for example: use of computer aided calculation in engineering design, use of computer aided drawing in engineering design: novelty level of design problem, complexity level of solved design problem stage of primary documentation, human factor, organizational conditions connected with documentation management, kind of design work. 3. Creation of time grades. All task of design department are classified to particular time grades. One time grade includes tasks of similar labour consumption. 4. Building of database. Create teaching sets containing tasks which were done in the past and which are characterized by coefficient described in step 2. 5. The classifications of new task to grade of set use to for example neural network and determine labour consumption of new one. 6. Use of planned labour consumption to purpose. Example of using neural network to determine time standard. Example of using neural network for lime standard determination based on integration artificial intelligence packet SPHiNX2.3. Neural network enable to determine planned labour-consumption for work, which has not been carried out so far. With the aid of systematic completing of database, definition of labour-consumption will be more precise. There will be taken into consideration changes occurring m enterprise as for example- greater experience of employees, better software and hardware conversion. The first stage connected with elaboration of teaching set is more difficult and more responsible. Correct system's operation depends on quality of data, that's way participation of experts in creating the primary set is necessary. The results can be the base for development and implementation of neural network supporting management of project departments in machine industry.

16

Metodologia planowania kosztów prac przygotowania produkcji

Kutschenreiter-Praszkiewicz I.

Inżynieria Maszyn

|

2001

|

R. 6, z. 3

65-76

PL

Informacja o rzeczywistym koszcie wyrobu lub usługi ma istotne znaczenie dla skutecznego funkcjonowania przedsiębiorstwa, w warunkach stale rosnącej konkurencji. Stąd konieczność stosowania doskonalszych narzędzi - metod analizy danych, umożliwiających coraz dokładniejsze monitorowanie procesów zachodzących w przedsiębiorstwie.

EN

Market economy conditions force actions upon enterprises, aiming at quality improvement, production cycle shortening and cost reduction. To reach this goal it's necessary to determine precise operational times, on that base of which we can establish the costs of operations to be carried out. This is especially important in small and medium batch production methods. In this conditions is necessary design a lot of different construction. In this situation methods of work measurement based on analytical estimating have greater meaning opposed to conventional ones, which use more time and has a larger cost.